Advances in The Human Side of Service Engineering : Proceedings of the AHFE 2017 International Conference on The Human Side of Service Engineering, July 17-21, 2017, The Westin Bonaventure Hotel, Los Angeles, California, USA 978-3-319-60486-2, 3319604864, 978-3-319-60485-5

Table of contents :
Front Matter....Pages i-xiii
Front Matter....Pages 1-1
Using Gamification to Enhance User Motivation: The Influence of Gender and Age....Pages 3-10
The Interaction of Causal Attribution of Performance and Compliance with Decision Support Systems in Cyber-Physical Production Systems - An Empirical Study Using a Business Simulation Game....Pages 11-23
Gamified Self-service Checkouts: The Influence of Computer-Related Causal Attributions on User Experience and Motivation....Pages 24-36
Defining Recrutainment: A Model and a Survey on the Gamification of Recruiting and Human Resources....Pages 37-49
SciencOmat: A Gamified Research Platform for Evaluating Visual Attractiveness....Pages 50-60
Front Matter....Pages 61-61
Alarm Management: A Systems Approach to Patient Safety....Pages 63-72
How Is the Quality of Life of Patients with Cerebral Palsy Improved? Qualitative and Quantitative Evaluation of a Communication and Learning Assistance System Based on ICTs....Pages 73-81
Prevalence and Risk Factors of Respiratory Problems Among Solid Waste Collectors in the Philippines....Pages 82-89
Prevalence, Severity, and Risk Factors of Work-Related Musculoskeletal Disorders Among Stevedores in a Philippine Break-Bulk Port Terminal....Pages 90-100
Front Matter....Pages 101-101
Applying Value Proposition Design for Developing Smart Service Business Models in Manufacturing Firms....Pages 103-114
Wisdom Service Systems: Harmonious Interactions Between People and Machine....Pages 115-127
Do You Like According to Your Lifestyle? A Quantitative Analysis of the Relation Between Individual Facebook Likes and the Users’ Lifestyle....Pages 128-139
An Online Tool to Promote Citizenship Roles in Urban Community Organizations....Pages 140-150
Front Matter....Pages 151-151
Factors Influencing Academic Success for Design Students: A Study of Curricular Expectations and Ethical Issues....Pages 153-161
Humanity-Embedded Service Science Education....Pages 162-172
Developing T-Shaped Civil Engineers Through Involvement in Out-of-Class Activities....Pages 173-177
T-Shape Professionals Co-working in Smart Contexts: VEGA(ST) – Venice Gateway for Science and Technology....Pages 178-190
Front Matter....Pages 191-191
Work-Related Factors Affecting Sustained Alert State Among Bank Security Personnel in the Philippines....Pages 193-202
The Negative Effect that Benefit Delay in Service Consumption Causes to a Customer and the Countermeasure Against It....Pages 203-214
A Task Analysis of Small-Scale Jewelry Craft Workers to Investigate the Effects of Work System Elements and Sitting Mobility on Body Discomfort....Pages 215-226
Front Matter....Pages 191-191
The Consequences of Lean Six Sigma on Banking Improvement: A Study at a Front-Line Unit of a Bank Company in Indonesia....Pages 227-237
Back Matter....Pages 239-240

Citation preview

Advances in Intelligent Systems and Computing 601

Louis E. Freund Wojciech Cellary Editors

Advances in The Human Side of Service Engineering Proceedings of the AHFE 2017 International Conference on The Human Side of Service Engineering, July 17–21, 2017, The Westin Bonaventure Hotel, Los Angeles, California, USA

Advances in Intelligent Systems and Computing Volume 601

Series editor Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland e-mail: [email protected]

About this Series The series “Advances in Intelligent Systems and Computing” contains publications on theory, applications, and design methods of Intelligent Systems and Intelligent Computing. Virtually all disciplines such as engineering, natural sciences, computer and information science, ICT, economics, business, e-commerce, environment, healthcare, life science are covered. The list of topics spans all the areas of modern intelligent systems and computing. The publications within “Advances in Intelligent Systems and Computing” are primarily textbooks and proceedings of important conferences, symposia and congresses. They cover significant recent developments in the field, both of a foundational and applicable character. An important characteristic feature of the series is the short publication time and world-wide distribution. This permits a rapid and broad dissemination of research results.

Advisory Board Chairman Nikhil R. Pal, Indian Statistical Institute, Kolkata, India e-mail: [email protected] Members Rafael Bello Perez, Universidad Central “Marta Abreu” de Las Villas, Santa Clara, Cuba e-mail: [email protected] Emilio S. Corchado, University of Salamanca, Salamanca, Spain e-mail: [email protected] Hani Hagras, University of Essex, Colchester, UK e-mail: [email protected] László T. Kóczy, Széchenyi István University, Győr, Hungary e-mail: [email protected] Vladik Kreinovich, University of Texas at El Paso, El Paso, USA e-mail: [email protected] Chin-Teng Lin, National Chiao Tung University, Hsinchu, Taiwan e-mail: [email protected] Jie Lu, University of Technology, Sydney, Australia e-mail: [email protected] Patricia Melin, Tijuana Institute of Technology, Tijuana, Mexico e-mail: [email protected] Nadia Nedjah, State University of Rio de Janeiro, Rio de Janeiro, Brazil e-mail: [email protected] Ngoc Thanh Nguyen, Wroclaw University of Technology, Wroclaw, Poland e-mail: [email protected] Jun Wang, The Chinese University of Hong Kong, Shatin, Hong Kong e-mail: [email protected]

More information about this series at http://www.springer.com/series/11156

Louis E. Freund Wojciech Cellary •

Editors

Advances in The Human Side of Service Engineering Proceedings of the AHFE 2017 International Conference on The Human Side of Service Engineering, July 17–21, 2017, The Westin Bonaventure Hotel, Los Angeles, California, USA

123

Editors Louis E. Freund Department of Industrial and Systems Engineering San Jose State University San Jose, CA USA

Wojciech Cellary Department of Information Technology Poznan University of Economics and Business Poznan Poland

ISSN 2194-5357 ISSN 2194-5365 (electronic) Advances in Intelligent Systems and Computing ISBN 978-3-319-60485-5 ISBN 978-3-319-60486-2 (eBook) DOI 10.1007/978-3-319-60486-2 Library of Congress Control Number: 2017943054 © Springer International Publishing AG 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Advances in Human Factors and Ergonomics 2017

AHFE 2017 Series Editors Tareq Z. Ahram, Florida, USA Waldemar Karwowski, Florida, USA

8th International Conference on Applied Human Factors and Ergonomics and the Affiliated Conferences Proceedings of the AHFE 2017 International Conference on the Human Side of Service Engineering, July 17−21, 2017, The Westin Bonaventure Hotel, Los Angeles, California, USA Advances in Affective and Pleasurable Design Advances in Neuroergonomics and Cognitive Engineering Advances in Design for Inclusion Advances in Ergonomics in Design Advances in Human Error, Reliability, Resilience, and Performance Advances in Human Factors and Ergonomics in Healthcare and Medical Devices Advances in Human Factors in Simulation and Modeling Advances in Human Factors and System Interactions Advances in Human Factors in Cybersecurity Advances in Human Factors, Business Management and Leadership Advances in Human Factors in Robots and Unmanned Systems Advances in Human Factors in Training, Education, and Learning Sciences Advances in Human Aspects of Transportation

WonJoon Chung and Cliff (Sungsoo) Shin Carryl Baldwin Giuseppe Di Bucchianico and Pete Kercher Francisco Rebelo and Marcelo Soares Ronald L. Boring Vincent G. Duffy and Nancy Lightner Daniel N. Cassenti Isabel L. Nunes Denise Nicholson Jussi Kantola, Tibor Barath and Salman Nazir Jessie Chen Terence Andre Neville A. Stanton (continued)

v

vi

Advances in Human Factors and Ergonomics 2017

(continued) Advances in Human Factors, Software, and Systems Engineering Advances in Human Factors in Energy: Oil, Gas, Nuclear and Electric Power Industries Advances in Human Factors, Sustainable Urban Planning and Infrastructure Advances in the Human Side of Service Engineering Advances in Physical Ergonomics and Human Factors Advances in Human Factors in Sports, Injury Prevention and Outdoor Recreation Advances in Safety Management and Human Factors Advances in Social & Occupational Ergonomics Advances in Ergonomics of Manufacturing: Managing the Enterprise of the Future Advances in Usability and User Experience Advances in Human Factors in Wearable Technologies and Game Design Advances in Communication of Design Advances in Cross-Cultural Decision Making

Tareq Z. Ahram and Waldemar Karwowski Paul Fechtelkotter and Michael Legatt Jerzy Charytonowicz Louis E. Freund and Wojciech Cellary Ravindra Goonetilleke and Waldemar Karwowski Tareq Z. Ahram Pedro Arezes Richard Goossens Stefan Trzcielinski Tareq Ahram and Christianne Falcão Tareq Ahram and Christianne Falcão Amic G. Ho Mark Hoffman

Preface

The 5th International Conference on the Human Side of Service Engineering (HSSE 2017) held at The Westin Bonaventure Hotel, Los Angeles, California, USA, from July 17 to 21, 2017. The conference was organized within the framework of the International Conference on Applied Human Factors and Ergonomics (AHFE) as an Affiliated Conference. If there is any one element to the engineering of service systems that is unique, it is the extent to which the suitability of the system for human use, human service, and for providing an excellent human experience has been and must always be considered. This conference included over 60 presentations during the three-day event in sessions including Gamification in Service Engineering, Advances in Health Care Facilities Design and Quality and Safety Practices, Shaping the Human Side of T-Shaped Professionals, Industry 4.0 and the Smart Service World, and Service Science and User Experience Design. While many of the presentations were contributed as slides only, 21 of the presenters also submitted full papers that were reviewed and accepted by the co-editors for publication in this Proceedings. The book is divided into five major sections as follows: Section Section Section Section Section

1: 2: 3: 4: 5:

Gamification in Service Engineering Innovations in Quality and Safety Practices in Health Care Systems Smart Service World Service Science Education and T-Shaped Professionals Worker and Consumer Factors in Service Engineering

A major area of expansion in this conference is in Gamification in Service Engineering, led by Freimut Bodendorf and Christian Zagel, which appears as the first section of this Proceedings. As service systems look for new ways to engage participants, gamification has become a fruitful strategy for research. New applications of the concepts are reported in five full papers in this section. New to the conference this year were two sessions led by Jim Spohrer focusing on the topic of the T-Shaped Professional. This has been a subject of substantial

vii

viii

Preface

interest in the areas of personnel development and education in recent years. The concepts of “depth in a discipline or system” and breadth across fields of study, cultures, organizational responsibilities, and other experiences contributing to managerial, empathy, communications, teamwork and similar skills have been leading topics in discussions about educating service systems professionals. Of the 11 presentations in these sessions, we are pleased to have two representative papers included in this Proceedings. We intend to expand our content in this field in the future. Other major areas of discussions at this meeting were in the fields of service design and improvement in the healthcare sector, led by Lukasz Mazur, Kendra Johnson, Alison Amos, Gregg Tracton, and Carolyn Brown, and service innovations in industry (referred to as Industry 4.0) led by Walter Ganz. Other session chairs included Aura Matias, Regiane Romano, Shrikant Parikh, and Debra Satterfield. We are deeply indebted to all session chairs for their contributions to HSSE 2017. We hope that you find this volume useful and interesting, and invite all to join us for next year’s meeting. Information is available at www.ahfe-hsse.org. Our sincere thanks and appreciation go to the Board members listed below for their contribution to the high scientific standard maintained in developing this book. Alison Amos, USA Clara Bassano, Italy Caio Márcio Almeida e Silva, Brazil Freimut Bodendorf, Germany Carolyn Brown, USA Bo Edvardsson, Sweden Walter Ganz, Germany Dolly Goel, USA Kazuyoshi Hidaka, Japan Kendra Johnson, USA Eunji Lee, Norway Christine Leitner, UK Aura C. Matias, Philippines Lukasz Mazur, USA Prithima Mosaly, USA Maria Victoria Palacin-Silva, Finland Shrikant Parikh, India Birgit Penzenstadler, USA Paolo Piciocchi, Italy Jari Porras, Finland Regiane Romano, Brazil Debra Satterfield, USA Yuriko Sawatani, Japan Ahmed Seffah, Finland

Preface

ix

Jim Spohrer, USA Kinley Taylor, USA Gregg Tracton, USA Christian Zagel, Germany July 2017

Louis E. Freund Wojciech Cellary

Contents

Gamification in Service Engineering Using Gamification to Enhance User Motivation: The Influence of Gender and Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sophie Jent and Monique Janneck The Interaction of Causal Attribution of Performance and Compliance with Decision Support Systems in Cyber-Physical Production Systems - An Empirical Study Using a Business Simulation Game . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philipp Brauner, Ralf Philipsen, André Calero Valdez, and Martina Ziefle Gamified Self-service Checkouts: The Influence of Computer-Related Causal Attributions on User Experience and Motivation . . . . . . . . . . . . . Adelka Niels and Christian Zagel Defining Recrutainment: A Model and a Survey on the Gamification of Recruiting and Human Resources . . . . . . . . . . . . Oliver Korn, Florian Brenner, Julian Börsig, Fabio Lalli, Maik Mattmüller, and Andrea Müller SciencOmat: A Gamified Research Platform for Evaluating Visual Attractiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christian Zagel, Alexander Piazza, Yoan Petrov, and Freimut Bodendorf

3

11

24

37

50

Innovations in Quality and Safety Practices in Health Care Systems Alarm Management: A Systems Approach to Patient Safety . . . . . . . . . . Vinaya Sebastian and Carolyn Hamasaki Brown

63

xi

xii

Contents

How Is the Quality of Life of Patients with Cerebral Palsy Improved? Qualitative and Quantitative Evaluation of a Communication and Learning Assistance System Based on ICTs . . . . . . Jonnathan Berrezueta-Guzman, Vladimir Robles-Bykbaev, and Luis Serpa-Andrade

73

Prevalence and Risk Factors of Respiratory Problems Among Solid Waste Collectors in the Philippines . . . . . . . . . . . . . . . . . . . Lizbeth A. Mariano and Aura C. Matias

82

Prevalence, Severity, and Risk Factors of Work-Related Musculoskeletal Disorders Among Stevedores in a Philippine Break-Bulk Port Terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Keneth Sedilla and Aura Matias

90

Smart Service World Applying Value Proposition Design for Developing Smart Service Business Models in Manufacturing Firms . . . . . . . . . . . . . . . . . . . . . . . . . 103 Jens Neuhüttler, Inka C. Woyke, and Walter Ganz Wisdom Service Systems: Harmonious Interactions Between People and Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Md. Abul Kalam Siddike, Kazuo Iwano, Kazuyoshi Hidaka, Youji Kohda, and Jim Spohrer Do You Like According to Your Lifestyle? A Quantitative Analysis of the Relation Between Individual Facebook Likes and the Users’ Lifestyle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Alexander Piazza, Christian Zagel, Julia Haeske, and Freimut Bodendorf An Online Tool to Promote Citizenship Roles in Urban Community Organizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 Adrian Haurat and Li Liu Service Science Education and T-Shaped Professionals Factors Influencing Academic Success for Design Students: A Study of Curricular Expectations and Ethical Issues . . . . . . . . . . . . . . 153 Debra Satterfield Humanity-Embedded Service Science Education . . . . . . . . . . . . . . . . . . . 162 Fu-Ren Lin Developing T-Shaped Civil Engineers Through Involvement in Out-of-Class Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Denise R. Simmons

Contents

xiii

T-Shape Professionals Co-working in Smart Contexts: VEGA(ST) – Venice Gateway for Science and Technology . . . . . . . . . . . 178 Paolo Piciocchi, James C. Spohrer, Linda Martuscelli, Maria Cristina Pietronudo, Michele Scocozza, and Clara Bassano Worker and Consumer Factors in Service Engineering Work-Related Factors Affecting Sustained Alert State Among Bank Security Personnel in the Philippines . . . . . . . . . . . . . . . . . . . . . . . . 193 Yoshiki B. Kurata and Aura C. Matias The Negative Effect that Benefit Delay in Service Consumption Causes to a Customer and the Countermeasure Against It . . . . . . . . . . . 203 Kazuhiro Fujimura A Task Analysis of Small-Scale Jewelry Craft Workers to Investigate the Effects of Work System Elements and Sitting Mobility on Body Discomfort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Hazel Caparas and Aura Matias The Consequences of Lean Six Sigma on Banking Improvement: A Study at a Front-Line Unit of a Bank Company in Indonesia. . . . . . . 227 Nilda Tri Putri, Andi Gunawan, and Agus Sutanto Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239

Gamification in Service Engineering

Using Gamification to Enhance User Motivation: The Influence of Gender and Age Sophie Jent(&) and Monique Janneck Electrical Engineering and Computer Science, University of Applied Sciences, Mönkhofer Weg 239, 23562 Luebeck, Germany {sophie.jent,monique.janneck}@fh-luebeck.de Abstract. The number of solo-self-employed and persons with flexible, individualized working conditions has grown over the last years. An online coaching application is currently being developed for this target group to increase well-being, reduce stress, and convey job design skills. An important point in designing this application is to keep users’ motivation high. A possible solution is gamification. To investigate possible effects of gamification, we asked 300 participants from our target group in an online survey how they would be motivated by gamification elements in an online coaching application. In this paper we report results regarding the influence of age and gender on the effects of gamification on user motivation. The results show that the use of progress bars, collecting points, and the activation of new content is seen as the most motivating. Male users would feel more motivated by gamification than female participants. Furthermore, motivation through gamification decreases with age. Keywords: Gamification


Motivation
Gender
Age
Online survey

1 Introduction Over the last years, the number of solo-self-employed and persons with individualised, flexible working conditions, like project work, home office and work while traveling, has grown [1, 2]. The advantages of flexible working conditions are individual working hours, increased self-responsibility, and autonomy [3, 4]. However, despite the benefits of flexible work, there are also challenges, like balancing work and leisure and designing one’s own work tasks and conditions. Therefore, these persons need comprehensive skills for job crafting [5] in order to stay healthy and work efficiently [6]. As part of a larger research project an online coaching application is currently being developed as a web application for this target group to increase well-being, reduce stress, and convey job design skills. An important point in designing this application is to keep users’ motivation high, especially because this target group has long working hours and high workloads. A possible solution is gamification, i.e. using game-design elements in a non-game context [7]. Previously, a mini-study was carried out based on an earlier prototype of the online coach to examine which gamification elements would motivate users of our target group over a longer period of time [8]. The study suggested that user motivation depends on the type of the game elements used and that the effects differ between women and men. © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_1

4

S. Jent and M. Janneck

To confirm these results based on a larger sample and also to investigate the influence of age on the effects of gamification on user motivation, we conducted an online survey with 300 participants. The results are reported in this paper. The paper is structured as follows: In the next section we give an overview of related work in this field and describe commonly used game-design elements. Our research questions and methods are explained in Sect. 3. Results are presented and discussed in Sects. 4 and 5.

2 Related Work Gamification is “the use of game design elements in non-game contexts” with the goal of increasing motivation and user activity [7, p. 2]. The use of gamification in applications has increased in recent years. Especially game-design elements like points, badges, leaderboards, progress bars [9], countdowns [10] and content activation [11] are used in many applications. • Points are a basic component in many games [12]. Users can collect points for successful execution of certain activities [13]. • Badges are visual awards that users obtain for certain achievements. Users are motivated to collect further “trophies” [14]. Furthermore, badges represent possible activities [15] and may draw the user’s attention to additional tasks not yet completed. • Leaderboards are ranking systems, presenting a list of users that are ordered according to their scores [16], e.g., points collected. Leaderboards add a social component and invoke a sense of competition [17]. • Progress bars show to what extent a task has been completed and represent positive development [18]. • Countdowns limit the time available for carrying out specific tasks and thus implement time pressure as a game factor [19]. • Content activation (or unlocking) means that the player gains access to content or functionality according to certain rules, e.g., only after prior tasks were fulfilled. Hamari et al. [9] reviewed 24 studies, in which gamification was used. The authors conclude that gamification may have a positive effect on psychological outcomes (e.g., motivation, enjoyment and attitude) and behavioral outcomes (e.g., effectiveness of learning). However, use context andusers’ characteristics have a strong influence on the results. A previous mini-study with 16 participants (7 female, 9 male, age: 28–60 years) investigating motivational effects of different gamification elements showed that progress bars, activation of new contents, collecting badges and points would achieve the most positive motivational effects in the long term [8]. A leaderboard would have a less positive effect. The results also show that gamification elements have different effects on women and men: male users feel more motivated by points and badges than female users. Motivation was also affected by the type of badges (e.g., frequency of registration, performing exercises). Interestingly, the participants’ general gaming behavior had no influence on the results.

Using Gamification to Enhance User Motivation

5

So far, there are hardly any studies focusing on demographic differences regarding the effects of gamification. This lack of data was already observed by Koivisto and Hamari [20], who were among the first to investigate demographic differences regarding gamification. Participants of their study were members of a social fitness network where users can collect points and badges. The findings demonstrate that women gain greater social benefits from the use of gamification. Furthermore, ease of use of the service is negatively affected by higher age. The influence of gamification elements on the purchase decisions for a product was studied by Bittner and Schipper [21]. In general, a positive influence was observed, although purchase intentions as well as perceived usefulness and enjoyment were found to be lower for older than for younger people.

3 Methods We conducted an online survey presenting different gamification elements (badges, points, a leaderboard, progress bars, activation of new content) and asked the participants how they would be motivated by these elements in an online coaching application. They were able to choose between “very motivated”, “motivated”, “partly motivated”, “hardly motivated” and “not motivated”. Counters were not considered in the survey because they will not be used in our coaching application, as the tasks are not supposed to be performed under stress such as time pressure. Furthermore, variables such as age, gender, personality traits and also general working characteristics were included in the questionnaire. In the survey, 300 people participated (147 female, 153 male, age: 22–64 years). Participants were recruited by a market research institute. Selection criteria included full-time employment (30 or more hours per week), professional education (university degree) and flexible working hours. For analysis, participants were divided into eight groups by gender and age to obtain four groups of men and four groups of women in the same age categories (see Table 1). To measure the popularity of the elements they were weighted on a linear scale and compared in terms of the average over all answers in a certain group. Table 1. Groups by age/gender. Group name Gender Age Number of participants A Female 20–29 35 B Female 30–39 47 C Female 40–49 37 D Female 50–64 28 E Male 20–29 26 F Male 30–39 58 G Male 40–49 39 H Male 50–64 30

6

S. Jent and M. Janneck

4 Results The combined results (all participants, see Table 2) show that the use of progress bars, collecting points, and the activation of new content is considered most motivating by the participants. Collecting badges and a leaderboard are rated significantly less motivating than the other elements.

Table 2. Survey results (all participants, no division in groups). Badges

Points

Leaderboard Progress bars Activation of new content very motivated 4,3% (13) 9,7% (29) 5,7% (17) 10,7% (32) 7,7% (23) motivated 27,7% (83) 39,7% (119) 23,0% (69) 41,3% (124) 37,3% (112) partly motivated 33,3% (100) 34,7% (104) 39,3% (118) 33,7% (101) 37,0% (111) hardly motivated 20,7% (62) 9,7% (29) 20,7% (62) 7,7% (23) 9,0% (27) not motivated 14,0% (42) 6,3% (19) 11,3 (34) 6,7% (20) 9,0% (27)

Gender seems not to affect the general tendency – women (groups A–D combined) as well as men (groups E–H combined) rated the use of progress bars, collecting points, and the activation of new content best. However, for each element women chose “very motivating” more often than men, but the ratio of persons choosing either “very motivating” or “motivating” is higher among men than among women (see Table 3). Table 3. Survey results of female (f, groups A–D) and male (m, groups E–H) participants.

very motivated

f m motivated f m partly motivated f m hardly motivated f m not motivated f m

Badges

Points

5,4% (8) 3,3% (5) 23,1% (34) 32,0% (49) 33,3% (49) 33,3% (51) 23,1% (34) 18,3% (28) 15,0% (22) 13,2% (20)

11,6% (17) 7,8% (12) 36,1% (53) 43,1% (66) 34,0% (50) 35,3% (54) 9,5% (14) 9,8% (15) 8,8% (13) 3,9% (6)

Leaderboard Progress bars Activation of new content 7,5% (11) 11,6% (17) 8,8% (13) 3,9% (6) 9,8% (15) 6,5% (10) 20,4% (30) 36,7% (54) 35,4% (52) 25,5% (39) 45,8% (70) 39,2% (60) 36,1% (53) 35,4% (52) 36,1% (53) 42,5% (65) 32% (49) 37,9% (58) 22,4% (33) 8,8% (13) 9,5% (14) 19,0% (29) 6,5% (10) 8,5% (13) 13,6% (20) 7,5% (11) 10,2% (15) 9,2% (14) 5,9% (9) 7,8% (12)

Using Gamification to Enhance User Motivation

7

Table 4. Survey results of all female groups (A: 20–29 years, B: 30–39 years, C: 40–49 years, D: 50–64 years).

very motivated

A B C D motivated A B C D partly motivated A B C D hardly motivated A B C D not motivated A B C D

Badges

Points

5,7% (2) 6,4% (3) 2,7% (1) 7,1% (2) 34,3% (12) 31,9% (15) 16,2% (6) 3,6% (1) 31,4% (11) 29,8% (14) 43,2% (16) 28,6% (8) 17,1% (6) 17,0% (8) 32,4% (12) 28,6% (8) 11,4% (4) 14,9% (7) 5,4% (2) 32,1% (9)

8,6% (3) 6,4% (3) 16,2% (6) 17,9% (5) 45,7% (16) 46,8% (22) 27,0% (10) 17,9% (5) 34,3% (12) 29,8% (14) 43,2% (16) 28,6% (8) 5,7% (2) 10,6% (5) 10,8% (4) 10,7% (3) 5,7% (2) 6,4% (3) 2,7% (1) 25,0% (7)

Leaderboard Progress bars Activation of new content 2,9% (1) 11,4% (4) 8,6% (3) 8,5% (4) 8,5% (4) 12,8% (6) 5,4% (2) 16,2% (6) 5,4% (2) 14,3% (4) 10,7% (3) 7,1% (2) 22,9% (8) 37,1% (13) 40,0% (14) 29,8% (14) 46,8% (22) 29,8% (14) 18,9% (7) 29,7% (11) 40,5% (15) 3,6% (1) 28,6% (8) 32,1% (9) 42,9% (15) 37,1% (13) 31,4% (11) 34,0% (16) 31,9% (15) 42,6% (20) 40,5% (15) 40,5% (15) 37,8% (14) 25,0% (7) 32,1% (9) 28,6% (8) 17,1% (6) 5,7% (2) 8,6% (3) 19,1% (9) 8,5% (4) 8,5% (4) 32,4% (12) 13,5% (5) 13,5% (5) 21,4% (6) 7,1% (2) 7,1% (2) 14,3% (5) 8,6% (3) 11,4% (4) 8,5% (4) 4,3% (2) 6,4% (3) 2,7% (1) 0,0% (0) 2,7% (1) 35,7% (10) 21,4% (6) 25,0% (7)

Moreover, an evaluation was made with regard to age. The younger participants (age 20–29 years, groups A and E) felt most motivated by collecting points. When considering the female groups (A–D) representing the different age categories, we observed that in all groups progress bars, collecting points, and the activation of new content were rated best. The women’s age group B (30–39 years) shows the highest levels of motivation, which decrease substantially with increasing age (see Table 4). Within the male groups (E–H), the best results were achieved for progress bars, collecting points, and the activation of new content. As for women, motivation through gamification decreases with age (see Table 5). The results also show that many respondents either like or dislike gamification elements – those who did not feel motivated by progress bars, points, and activation of new content also rated the other elements no higher than “hardly motivating”.

8

S. Jent and M. Janneck

Table 5. Survey results of all male groups (E: 20–29 years, F: 30–39 years, G: 40–49 years, H: 50–64 years).

very motivated

E F G H motivated E F G H partly motivated E F G H hardly motivated E F G H not motivated E F G H

Badges

Points

7,7% (2) 5,2% (3) 0,0% (0) 0,0% (0) 46,2% (12) 32,8% (19) 35,9% (14) 13,3% (4) 23,1% (6) 37,9% (22) 41,0% (16) 23,3% (7) 11,5% (3) 15,5% (9) 12,8% (5) 36,7% (11) 11,5% (3) 8,6% (5) 10,3% (4) 26,7% (8)

15,4% (4) 5,2% (3) 10,3% (4) 3,3% (1) 53,8% (14) 50,0% (29) 35,9% (14) 30,0% (9) 23,1% (6) 32,8% (19) 41,0% (16) 43,3% (13) 7,7% (2) 5,2% (3) 10,3% (4) 20,0% (6) 0,0% (0) 6,9% (4) 2,6% (1) 3,3% (1)

Leaderboard Progress bars Activation of new content 3,8% (1) 15,4% (4) 7,7% (2) 6,9% (4) 10,3% (6) 6,9% (4) 0,0% (0) 12,8% (5) 10,3% (4) 3,3% (1) 0,0% (0) 0,0% (0) 15,4% (4) 46,2% (12) 42,3% (11) 29,3% (17) 44,8% (26) 50,0% (29) 30,8% (12) 46,2% (18) 30,8% (12) 20,0% (6) 46,7% (14) 26,7% (8) 53,8% (14) 26,9% (7) 34,6% (9) 39,7% (23) 34,5% (20) 31,0% (18) 43,6% (17) 25,6% (10) 43,6% (17) 36,7% (11) 40,0% (12) 46,7% (14) 26,9% (7) 7,7% (2) 11,5% (3) 13,8% (8) 1,7% (1) 3,4% (2) 15,4% (6) 10,3% (4) 10,3% (4) 26,7% (8) 10,0% (3) 13,3% (4) 0,0% (0) 3,8% (1) 3,8% (1) 10,3% (6) 8,6% (5) 8,6% (5) 10,3% (4) 5,1% (2) 5,1% (2) 13,3% (4) 3,3% (1) 13,3% (4)

5 Conclusions In this study we investigated how persons with individualized working conditions could be motivated to use an online coaching application by means of gamification. Specifically, we focused on the influence of gender and age. In summary, gamification can be useful for each gender and age group to keep motivation high, especially elements like progress bars, points and the activation of new content. This confirms, with a larger sample, the results of the previous study [8] with respect to these elements. On the other hand, we notice a less motivating effect of leaderboards and badges in the current study. However, badges might have a positive effect in other contexts or for different target groups, since motivation strongly depends on the task to be performed [8]. It is hence important to consider in detail for which tasks badges are awarded. Although badges represent possible activities in an application [15], these should not be the only option for the user to obtain an overview. Generally, we observed that male users would feel more motivated by game-design elements than female participants. Moreover, motivation through gamification can be achieved in all age groups, although the effects decrease with age in our target group.

Using Gamification to Enhance User Motivation

9

The decreased motivation may be connected to the general reduction in ease of use at higher age, as observed by Koivisto and Hamari [20]. Consequently, it may be reasonable to introduce gamification elements step by step in order to not overstrain the user. In addition, one should give the users the possibility to regulate the presence of gamification, e.g., by the option to disable notification pop-ups informing the user about newly acquired badges or gained points. In particular, for our target group, consisting of people whose work life is characterized by high self-responsibility and autonomy, a flexible configuration seems appropriate. Although this survey shows less motivation caused by leaderboards, it remains to study the specific reasons. The competitive aspect could, for example be reduced through anonymization. Moreover, the comparison within an individually selected group of people, e.g., sharing similar age or profession, could be more interesting for the user than the classical leaderboard relating all users. In our future research we will analyse the influence of personality traits on the motivational assessment of different gamification elements. Furthermore, we plan to investigate more in detail how and for what kind of activities badges should be awarded in order to increase motivation. We also plan to integrate the studied concepts into our coaching application which will allow us to compare our results with a practical evaluation.

References 1. Lohmann, H., Luber, S.: Trends in self-employment in Germany: different types, different developments. In: Arumm, R., Müller, W. (eds.) The Reemergence of Self-employment: A Comparative Study of Self-employment Dynamics and Social Inequality, pp. 36–74. Princeton University Press, Princeton (2004) 2. Hill, E.J., Hawkins, A.J., Miller, B.C.: Work and family in the virtual office: perceived influences of mobile telework. Fam. Relat. 45, 293–301 (1996) 3. Parslow, R.A., Jorm, A.F., Christensen, H., Rodgers, B., Strazdins, L., D’Souza, R.M.: The associations between work stress and mental health: a comparison of organizationally employed and self-employed workers. Work Stress 18(3), 231–244 (2004) 4. Prottas, D.J., Thompson, C.A.: Stress, satisfaction, and the work-family interface: a comparison of self-employed business owners, independents, and organizational employees. J. Occup. Health Psychol. 11(4), 100–118 (2006) 5. Bakker, A.: Engagement and “job crafting”: engaged employees create their own great place to work. In: Albrecht, S.L. (ed.) Handbook of Employee Engagement: Perspectives, Issues, Research and Practice, pp. 229–244. Edward Elgar, Cheltenham (2010) 6. Tims, M., Bakker, A.B.: Job crafting: towards a new model of individual job redesign. J. Ind. Psychol. 36(2), 1–9 (2010) 7. Detering, S., Dixon, D., Khaled, R., Nacke, L.: From game design elements to gamefulness: defining gamification. In: Proceedings of the 15th International Academic MindTrek Conference: Envisioning Future Media Environments. ACM (2011) 8. Jent, S., Janneck, M.: Using gamification to enhance user motivation in an online-coaching application for flexible workers. In: Proceedings of the 12th International Conference on Web Information Systems and Technologies (2016)

10

S. Jent and M. Janneck

9. Hamari, J., Koivisto, J., Sarsa, H.: Does gamification work? - a literature review of empirical studies on gamification. In: 47th Hawaii International Conference on System Sciences (HICSS) (2014) 10. Corriero, N., Di Bitonto, P., Roselli, T., Rossano, V., Pesare, E.: Simulations of clinical cases for learning in e-health. Int. J. Inf. Educ. Technol. 4, 378–382 (2014) 11. Iosup, A., Epema, D.: An experience report on using gamification in technical higher education, In: Proceedings of the 45th ACM Technical Symposium on Computer Science Education. ACM (2014) 12. Zichermann, G., Cunningham, C.: Gamification by Design: Implementing Game Mechanics in Web and Mobile Apps. O’Reilly Media, Inc., Sebastopol (2011) 13. Werbach, K., Hunter, D.: For the Win: How Game Thinking Can Revolutionize Your Business. Wharton Digital Press, Philadelphia (2012) 14. Kumar, J.: Gamification at Work: Designing Engaging Business Software. Springer, Heidelberg (2013) 15. Montola, M., Nummenmaa, T., Lucero, A., Boberg, M., Korhonen, H.: Applying game achievement systems to enhance user experience in a photo sharing service. In: Proceedings of the 13th International MindTrek Conference: Everyday Life in the Ubiquitous Era, pp. 94–97. ACM, September 2009 16. Costa, J.P., Wehbe, R.R., Robb, J., Nacke, L.E.: Time’s up: studying leaderboards for engaging punctual behaviour. In: Proceedings of the First International Conference on Gameful Design, Research, and Applications, pp. 26–33. ACM, October 2013 17. Kapp, K.M.: The Gamification of Learning and Instruction: Game-Based Methods and Strategies for Training and Education. Wiley, San Francisco (2012) 18. Neeli, B.K.: A method to engage employees using gamification in BPO industry. In: Third International Conference on Services in Emerging Markets (ICSEM). IEEE (2012) 19. Hsu, S.H., Chang, J.W., Lee, C.C.: Designing attractive gamification features for collaborative storytelling websites. Cyberpsychol. Behav. Soc. Netw. 16(6), 428–435 (2013) 20. Koivisto, J., Hamari, J.: Demographic differences in perceived benefits from gamification. Comput. Hum. Behav. 35, 179–188 (2014) 21. Bittner, J.V., Schipper, J.: Motivational effects and age differences of gamification in product advertising. J. Consum. Mark. 31(5), 391–400 (2014)

The Interaction of Causal Attribution of Performance and Compliance with Decision Support Systems in Cyber-Physical Production Systems - An Empirical Study Using a Business Simulation Game Philipp Brauner(&), Ralf Philipsen, André Calero Valdez, and Martina Ziefle Human-Computer Interaction Center, RWTH Aachen University, Aachen, Germany {brauner,philipsen,calero-valdez, ziefle}@comm.rwth-aachen.de

Abstract. Supply Chains and production networks are complex sociotechnical systems whose performance is determined by system, interface, and human factors. While the influence of system factors (e.g., variances in delivery times and amount, queuing strategies) is increasingly well understood, the influence of the interface and human factors is currently insufficiently explored. Previous research has shown that decision support systems may help to enhance performance by improving the interface. In this work, we address the users’ trust in a decision support system. In a user study (n = 40), using a business simulation game, we investigated how four dimensions of attribution theory relate to trust in decision support systems and further to task performance. The results show that human factors, especially trust in automation and attribution theory relate to the performance in the business simulation game. We conclude that attribution relates to job compliance and performance in material disposition tasks and supply chain management. Keywords: Business simulation game
Industrial internet of things Automation
Trust in automation
Attribution theory
User modelling Human factors





1 Introduction In the era of digitalization, decisions are no longer made by humans alone. Algorithms filter and preselect options for later decision making or - with increasing pervasiveness decide on their own [1]. While there are many situations and contexts in which algorithms might have the better understanding of the situation and thus decide more competently and flawlessly, there are still many situations and contexts in which humans in the loop are needed due to their higher cognitive expertise and awareness for the respective situation. While both approaches might have benefits as well as drawbacks, it is pivotal to find a right balance between humans and cyber-physical systems. In © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_2

12

P. Brauner et al.

particular, countries with high wages must augment their production systems with smart algorithms to maintain their competitiveness. Despite the increasing automation, the human-in-the-loop has an irreplaceable role in these cyber-physical production systems and because of the automation, the human operators need to handle increasingly complex tasks in shorter time. The challenge in finding this balance is to harness the “individual” capabilities of both humans and computers to their maximal extent, in order to reap benefits beyond the individual contribution [2, 3]. Computers are very suitable for tasks that involve complex mathematical operations, handling large data, or e.g., optimizing unique fitness functions. However, they are (still) blind to information not encoded in data (e.g., larger context, ethical implications [4], human perception of decisions, black swan effects [5]). This blindness is directly related to tasks where humans excel. The challenge at hand is how to harmonize the effort of both entities to maximize benefits and utility. In a cyber-physical production network—often referred to as Industry 4.0—setting, the operator must evaluate algorithm and data output to optimize the production point. This is particularly important when contextual information, unavailable to the closed world of the machine, would tip the scale in favor of a different decision. The new linchpin in this scenario is obviously the interface between machine and human, algorithm and operator, digital and analog. This crucial part must be designed to optimize the efficient application of the human resource and the transformation between the digital and analog world is most constrained by human perception (mostly visual ergonomics) and evaluation (in terms of interpretation of situational issues, cognitive ergonomics). Addressing human perception in interface design is to a large extent mapping multidimensional, highly complex data to lower dimensional information that is both comprehensible and actionable. This mapping process is mostly assumed to be governed by laws of perception and screen real-estate. Still, a large part of the uniqueness of human capability is ignored. Capabilities such as risk perception, multi-criterion decision making, success attribution [1] and other cognitive and affective factors enable human decision making, yet they are unequally distributed across individuals. Here, human factors research can play a critical role in leveraging individual differences by incorporating them in the “interface equation”. A key question is whether a one-size-fits-all approach is adequate and justified, given the diversity of human attributes [6]. In this article, we look at how individual differences in success attribution affect judgement in a decision-making task. This task is assisted by a decision support system that, depending on the experimental condition, may be helpful or deflective. By investigating the associations of performance and human factors, success attribution and objective task criteria, we hope to find rules that allow allocating human resources to tasks that are both suitable and useful. The overall objective is to develop smart and targeted tools and methods for augmenting the abilities of the human-in-the-loop in these socio-technical, cyber-physical production systems. Decision Support Systems (DSS) can be a viable solution, however, we need to understand when these support systems provide a real, measurable, and sustainable benefit. Also, we need to understand if, when, and why these systems are blindly obeyed with possible disastrous consequences for the process, the companies, and the customers.

The Interaction of Causal Attribution

13

2 Related Work The following sections present a brief overview on Decision Support Systems and Human Factors in Decision Support Systems. The second section summarizes the theory of causal attribution and illustrates its relationship with the design of interactive systems.

2.1

Decision Support Systems

The development of decision support systems has its roots in the 1950’s and aimed at providing workers assistance in decision making tasks by using the capabilities of upcoming computer systems [7]. Industrial and military tasks were the primary fields of application where operational, tactical, or strategic decision problems were (and still are) subject to growing task and information complexity. The idea was to split decision tasks into a computational part that is solved by machine, by implementing knowledge, models, and decision rules into a computational form [8], and a part that is still carried out by human, who then can use the support in terms of recommendations, reports, or visualizations. Today, modern DSS cover a broad range of methods and deployment scenarios, e.g., data warehousing [9], online analytical processing (OLAP) [10], or data-mining [11]. The growing data complexity is accompanied with an increasing importance of data-centric approaches and artificial intelligence that is able to handle large amounts of fuzzy information [12]. However, as long as decision-making is not fully automated and the human is still in the loop due to responsibility, ethical, political, or organizational reasons, it is necessary to study the effects of human factors on the use of decision support systems [13, 28]. Next to user interface issues, the compliance in terms of accepting or neglecting the assistance of a DSS is crucial for successful decision making, especially if the DSS is defective and should be overruled by human. In this context, the perceived effectiveness, the perceived usefulness and the trust in the system are likely to influence the acceptance and thereby the compliance with DSS [14–16]. Two companion papers on this study have investigated the influence of the Decision Support System’s defectiveness on attained performance, compliance and trust in automation [16, 28]: In general, a correctly working DSS yields higher trust in automation, higher compliance with the system, and higher overall company profit than a defective system. A further finding is that the subjective evaluations are basically consistent with the measurements from the underlying simulation model, meaning that the participants have a good perception of their own performance in the game. However, to better understand humans’ perception of both flawless and defective decision support systems, we need to gain deeper insights into the attribution of success and failure to these systems.

14

P. Brauner et al.

2.2

Theory of Causal Attribution

In the social sciences, the theory of causal attribution is defined as “[a] theory designed to explain how people perceive, infer, or ascribe causes of their own and other people’s behavior” [17]. Weiner, as a key scientist in this field, argued that attribution is subdivided into the four dimensions Locus Of Control, Stability, Controllability and Globality [18]: Locus of control addresses if an individual attributes the cause for the outcome of their actions to their own abilities (internal) or external factors (external). For example, the same good performance in a math exam might be attributed to oneself (“I am good at math”) or to external factors (“The exam was easy”). Stability as a dimension captures how much the cause for the outcome of one’s actions is considered to be just temporary (unstable) or rather constant (stable). For example, the exam performance can be attributed to luck (temporary) or as stable (stable). Controllability addresses whether the cause of the outcome is modifiable (controllable vs. not controllable). For example, the good performance in the aforementioned math exam can be attributed as controllable (“I am good at math because I have practiced a lot”) or as stable (“I can’t help it. I am just good at math”). Globality captures how much the cause for outcomes of one’s actions are specific to a certain domain or how much they also affect other domains (domain specific vs. generic; “I am good at math” vs. “I am good at school”). Figure 1 shows the theory of causal attribution and its dimensions. Attribution Theory—though well-known in the social science as a major driver of human well-being [17, 19]—is currently gaining importance in human-computer interaction research. For example, Niels et al. [20] successfully linked the four dimensions locus, stability, controllability, and globality from the attribution theory to the evaluation of a product on the User Experience Questionnaire (UEQ), or used attribution theory to study gamified supermarket checkouts [21]. Also, sub-dimensions of attribution theory and related constructs have been identified as a key player in human-computer interaction: Bandura’s self-efficacy theory is strongly related to locus of control and locus of control has been identified as a key predictor of efficiency, effectivity, and learning to use interactive technology in variety of different domains [22–24].

Fig. 1. Illustration of the four considered dimensions of the theory of causal attribution.

The Interaction of Causal Attribution

15

3 Research Design To evaluate the relationship between human factors in a scenario with both correct and defective decision support systems in cyber-physical production systems, we applied an exploratory experimental approach. We looked at four larger constructs in both human performance and attitudes: (1) trust in automation, (2) compliance, (3) performance, and (4) attributional factors. The participants interacted for two rounds with the “Quality Intelligence Game” business simulation game [25] that combines material disposition tasks from Sterman’s Beer Distribution Game [26] and quality management aspects from Goldratt’s game [27]. Here, the objective of the participants is to carefully balance costs of purchasing and stock-keeping supplies on one side with the investments into inspection of incoming goods and production quality on the other side—all in the setting of a simulated company. After each of the two rounds, a post-questionnaire captured the participant’s evaluation of the decision support system. A more verbose elucidation of the method can be found in our companion paper [16]. Within-subject factor: The Correctness of the Decision Support System (DSS) for ordering supplies is modified as a within-subject factor. This means that we adjust whether the DSS is helpful or leads the user astray. In the case of a correct DSS, the suggested number of supplies that should be ordered is near the optimum (only very experienced players might find slightly better order levels). In the case of the defect DSS, the suggested orders are correct for the first six turns of the game, then the system turns defective and suggests values 50% below the optimum; the defectiveness of this value is directly sensible from within the user interface (lower than the number of requested orders by the customer) and indirectly through dramatically increasing penalty costs in the subsequent turns of the game. The other tasks - investments in production quality - are not supported by a decision support system. Each round of the game consists of 18 turns (i.e., 18 months of the simulated company). Dependent variables: After each of the two rounds, the following dependent variables were measured either with log files in the game or by a survey. Compliance: The compliance with the Decision Support System is measured with the item “How often did you follow the suggestion of the DSS during the game?”. Performance: Following Goldratt and Cox the Company’s Profit is calculated as the overall objective performance metric as the cumulated net profit for each turn of the game [27]. In addition, the participants also reported their Perceived Performance. Attribution styles: After each round the participants reported on the perceived causes of their performance. The four attribution dimensions Locus of Control, Stability, Controllability, and Globality were captured with one item each (see Table 3). The subjective measures are captured on 6-point Likert scales and rescaled to 0–100% for reasons of legibility. Subjective Compliance was directly captured on a scale from 0–100%. Figure 2 illustrates the experimental setup of this study.

16

P. Brauner et al.

Fig. 2. Illustration of the experimental setup.

3.1

Methods

We analyzed the data using parametrical and non-parametrical methods, such as bivariate correlations (Pearson’s r or Spearman’s q), Wilcoxon tests, single and repeated multi- and univariate analyses of variance (M/ANOVA), and multiple linear regressions. We set the type I error rate (level of significance) to a = .05 (findings .05 < p < .1 are reported as marginally significant). Pillai’s value is considered for the multivariate tests and effect sizes are reported as η2. If the assumption of sphericity is not met, Greenhouse-Geisser–corrected values are used, but uncorrected dfs are reported for legibility. The whiskers in diagrams represent the standard error (SE) of the point estimate, arithmetic means are reported with standard deviations (denoted ±).

3.2

Description of the Sample

In total, 40 people in the age range from 20–56 years (M = 28.5 ± 8.6, Md = 25) participated voluntarily in the study (23 male, 17 female). Age and gender were not correlated in our sample (q = .200, p = .215 > .05, n.s.).

4 Results The results section is structured as follows. First, we look at the effect of playing the business simulation game twice for each player, guided by the question, if players improved their success in a second run of the game. We then look on the influence of the DSS. Here, the crucial question is whether a defective DSS changes the compliance and the performance for each player and how this changes the attribution of success for the player.

4.1

Effect of Repetition

Our study revealed no significant increase on overall attainted profit increased from the first (M = −2,622.5 ± 25,672, Md = 12,275) to the second round of the game (M = 3,557.69 ± 20,789, Md = 11,650) (Z = −.132, p = .895 > .05).Yet, there is a

The Interaction of Causal Attribution

17

strong correlation between the performance attained in the first and the second round of the game (q = .751, p < .001). Both these findings indicate that some players consistently attain higher performances than others. As previous studies have reported an increase in performance [25], we conclude that the absence of this effect in this study is based on the influence of the correctness of the decision support system. For both rounds of the game, there is a strong correlation between the actual Company Profit attained in the game and the Perceived Performance (q = .470, p = .002 < .05 for round 1 and q = .577, p < .001 for round 2). Thus, the participants from the study are able to judge their performance in line with their actual performance. The reported Compliance with the Decision Support System was 36.9 ± 30.0% in the first round of the game and 42.9 ± 28.7% in the second round. However, a repeated measures ANOVA with Round as within-subject variable and Compliance in round 1 and 2 as dependent variable revealed no significant differences (V = .029, F1,33 = .977, p = .330, η2 = .029). This means that players at least report to follow the suggestions by the decision support system to a similar extend in both rounds. Thus, further investigation on the evaluation of attribution regarding the DSS seems reasonable. If players would not have followed the DSS, no such conclusions could be drawn. Regarding the performance attribution, a repeated measures MANOVA with the four dimensions Locus of Control, Stability, Controllability, and Globality revealed no significant overall effect of repetition across both rounds of the game (V = .146, F4,35 = 1.492, p = .226 > .05, η2 = .146). Hence, the participant’s performance attribution does not change significantly during both rounds. Merely for the dimension Stability, a small and marginally significant difference emerges (F1,38 = 3.755, p = .060 > .05, η2 = .090) and the participants’ perceived stability of the performance increases from 56.9 ± 29.2% to 68.2 ± 19.9%). This means that the attribution of success—or lack thereof—becomes more stable after playing a second round of the game. A correlation analysis revealed that the attributed Locus of Control (q = .037, p = .819 > .05) and the Stability (q = −.007, p = .968 > .05) of the performance is not stable over the two repetitions of the game, whereas the perceived Controllability (q = .692, p < .001) and the perceived Globality (q = .508, p = .001 < .05) of the performance remains stable (see Table 2, left).

4.2

Effect of the Decision Support System

The overall attained Company Profit as the key performance metric for the defect and correct DSS are strongly correlated (q = .777, p < .001), which indicates that some players are consistently more successful than others. On average, the Company Profit gained with a correct DSS is higher (M = 7,110.3 ± 12,599.2, Md = 13,100) than the Company Profit gained with a defect DSS (M = −6,086.3 ± 29,269.2, Md = 11,350). A Wilcoxon test attests that this difference is significant (Z = −2.647, p = .008 < .05). Again, the Perceived Performance is related to the objectively measured Company Profit for both the defect (q = .442, p < .05) and the correct DSS (q = .389, p < .05).

18

P. Brauner et al.

Next, the relationship between the attained performance and the compliance with the support system is evaluated. First, for the correct DSS, then for the defective DSS: For the correct DSS, there is a marginally significant positive correlation between the Compliance with the DSS’s suggestions and the Perceived Performance (q = .284, p = .08 > .05) but no significant relationship between the objective Company Profit and the Compliance (q = −.174, p = .295 > .05). However, for the case of the defective support system, there is a strong negative relationship between Compliance and measured Company Profit (q = −.668, p < .001) as well as between Compliance and Perceived Profit (q = −.339, p < .05). Therefore, the compliance with the support system is a strong determinant for performance, but the influence of compliance can either be positive or negative, depending on the correctness of the support system. For the first round of the game, a MANOVA with the Correctness of the DSS as independent variable and the four attribution items revealedno significant overall effect of Correctness on Attribution (V = .056, F4,34 = .509, p = .730 > .05). As argued in the companion paper [16], the results from the first round of the game might not provide a clear view on differences in attribution, as the participant’s internal reference frame is not yet established after the first round of playing. However, significant overall differences in attribution based on Correctness emerge for the second round of the game (V = .272, F4,35 = 3.271, p = .022 < .05, η2 = .272). Specifically, significant differences are found for the dimensions Locus of Control (p = .027 < .05, η2 = .122) and Controllability (p = .030 < .05, η2 = .118), whereas Stability (p = .329 > .05) and Globality (p = .295 > .05) are not affected. This means that those aspects of attribution that are innate to the player are—as expected—unaffected by the experiment, while the two dimensions depending on the immediate context do change in our scenario. This is expected as the conditions between rounds change. As Table 1 and Fig. 3 illustrate, the participants with the correct decision support system experienced a significantly higher external locus of control than participants with the defect system (54.3 ± 29.1 vs. 33.7 ± 27.5%). However, participants with the correct decision support system perceive their performance as much more controllable than the participants with the defective system (68.6 ± 28.7 vs. 47.4 ± 30.7%). A correlation analysis shows that the performance attribution of defective and correct support systems is not related for the dimensions Locus of Control (q = .033, p = .841 > .05) and Stability (q = −.002, p = .992 > .05). But again, the dimensions Controllability (q = .718, p < .001) and Globality (q = .526, p < .001) are strongly related (see Table 2, right).

Table 1. Effect of the DSS’s correctness on the four dimensions of causal attribution for the first and second round of the game. Correctness External Locus Round 1 Correct Defect

Round 2

High Stability Round 1

High Control Round 2

Round 1

High Globality Round 2

Round 1

Round 2

44.2 ± 19.5 54.3 ± 29.1 63.3 ± 24.0 71.4 ± 24.1 53.7 ± 27.5 68.6 ± 28.7 47.4 ± 27.7 52.4 ± 27.2 45.7 ± 27.7 33.7 ± 27.5 51.4 ± 32.6 65.3 ± 13.1 61.9 ± 26.8 47.4 ± 30.7 46.7 ± 23.9 44.2 ± 20.6

The Interaction of Causal Attribution

19

100%

Level of AƩribuƟon

Defect

Correct

75%

50%

25%

0% External Control

High Stability

High Controllability

High Globality

Round 1

External Control

High Stability

High Controllability

High Globality

Round 2

Fig. 3. Attribution of the attained performance for both rounds for the correct and defective Decision Support System. Differences emerge for the second round (sig. for External Locus of Control and Controllability), as no reference frame has been established in the first round.

Table 2. Autocorrelations of the dimensions Locus of Control (Loc), Stability (Stab), Controllability (Cont), and Globality (Glob) from attribution theory based on repetition (left) and DSS correctness (right) (** p < .001, * p < .05, + p < .1, () n.s.). Repetition (R1 – R2) StabR2 ContR2 GlobR2 LocR2 LocR1 −0.037 (−.200) (−.147) −.285+ StabR1 (−.136) (−.007) (−.052) (−.157) ContR1 −0.01 −0.045 .692** .302+ GlobR1 −0.104 −0.175 .266+ .508**

Correctness (Correct – Defect) LocD StabD ContD LocC −0.033 (−.018) (−.117) StabC (−.183) (−.002) .279+ ContC −0.017 (−.244) .718** GlobC (−.225) (−.218) .355*

GlobD −0.207 −0.259 −0.233 .526**

5 Discussion and Conclusion The digitalization that is transforming manufacturing in high-wage countries and cyber-physical production systems benefits from increased automation. At the level of cross-company cooperation, supply chain disruptions are a major threat for manufacturing companies and suitable decision support systems are a viable method to mitigate these disruptions. However, previous studies found that operators are easily deflected by defective decision support systems and are misguided by insufficient or wrong information [16, 27]. The present study addressed the root of this issue and tried to understand if, when, and why people comply with support systems that are working correct, provide guidance, and offer support, and if, when, and why they are obeyed, even when they are misleading, defective, or deflective. The study found evidence that compliance with a correctly working system is beneficial for subjective performance. However, obedience with a defective DSS is thoroughly linked to a decrease in subjective performance as well as a significant decrease in the objective company performance.

20

P. Brauner et al.

The key question addressed in this study was to identify the causes for compliance and system obedience based on Weiner’s attribution theory. In general, the presented results show that attribution theory can indeed be applied in this context, as some of the captured constructs could successfully be linked to the experimental conditions from the study. Specifically, the dimensions locus of control and controllability were influenced by the experimentally varied correctness of the DSS, whereas the dimensions stability and globality remained unaffected. Obviously, a study focused on understanding the applicability of attribution theory in the context of decision support in cyber-physical production systems cannot provide definite and conclusive guidelines on how this theory can be harnessed in the design of support systems. Yet, even these focused findings presented here hint at locus of control and controllability as the key determinants for compliance with correct systems and obedience of faulty systems. Hence, shifting the locus of control to an internal attribution and increasing the perceived controllability of the situation, despite a defective support system, is crucial. This can be addressed either by adequately designed and implemented support systems that not only make the required tasks easier but also provide the rational for their suggestions and thereby increasing the operators’ understanding, ability, and confidence in their own capability. This will likely increase the compliance with correctly working systems and likewise will reduce the blind obedience with defective systems. In addition, an increased understanding of the underlying principles might be facilitated through trainings specifically addressing the control perception. Serious business simulation games—such as the one used here in the study—might be a valuable component of these knowledge and ability dissemination strategies. In summary, this study has shown that attribution theory can offer valuable insights on compliance with and obedience of correct and defect decision support in cyber-physical production systems. A profound understanding of how attribution theory relates to compliance, obedience, and performance will enable us to find the right balance between automation in cyber-physical production systems on one side and inclusion of the human-in-the-loop with their unique capabilities on the other side.

6 Limitations The present study allows some valuable insights on the link between attribution theory and the compliance with correct and defect decision support systems in the context of supply chain and quality management. However, the generalizability of the findings from this study is limited because of the small sample, the missing calibration of subjective evaluations in the first round of the game, and the possible confounding effects of practice and correctness. The experiment is based on a sample of just over 40 participants, which limits the analysis of between-subject effects. Also, the companion study found that the participants’ evaluations are not calibrated at the beginning of the study, which limits the permissibility of repeated measures analysis. Furthermore, the effect of the DSS’s correctness reported for the second round of the game is tainted by the effect of the previous round: People’s trust in automation and performance attribution may vary depending on whether they experienced a correct or defective DSS in

The Interaction of Causal Attribution

21

the beginning of the game. Thus, future studies should build on a larger sample and must more clearly separate the factors of repetition and correctness of the Decision Support System. Acknowledgements. We thank all participants of the presented study for their willingness to contribute to our research and our colleagues Anne Kathrin Schaar, Felix Dietze, Lisa Schwier, Marco Fuhrmann, Sebastian Stiller, Hao Ngo, and Robert Schmitt for support and in-depth discussions on this work. Furthermore, we like to thank Sabrina Schulte for her research support. The German Research Foundation (DFG) founded this project within the Cluster of Excellence “Integrative Production Technology for High-Wage Countries” and the Integrated Cluster Domain “Self-Optimizing Production Networks” [29].

Appendix The item texts of the constructs can be seen in Table 3. Table 3. Items for capturing the participant’s performance attribution (based on [20]). Dimension Locus of control Stability Controllability Globality

Text “The performance was mostly determined by external factors (such as the game and the support system)” “I will achieve a similar performance the next time I play this game” “I think the degree of my success is adjustable (for example, by putting more effort into solving the problem)” “For different games, I would have attained a similar performance”

References 1. Nguyen, T.T., Maxwell, P.H.F., Loren, H., Joseph, T.: Exploring the filter bubble: the effect of using recommender systems on content diversity. In: Proceedings of the 23rd International Conference on World Wide Web, pp. 677–686. ACM (2014) 2. Calero Valdez, A., Brauner, P., Schaar, A.K., Holzinger, A., Ziefle, M.: Reducing complexity with simplicity - usability methods for industry 4.0. In: 9th Triennial Congress of the International Ergonomics Association (IEA 2015), Melbourne, Australia (2015) 3. Holzinger, A.: Interactive machine learning for health informatics: when do we need the human-in-the-loop? Brain Inform. 3, 119–131 (2016) 4. Moor, J.H.: The nature, importance, and difficulty of machine ethics. IEEE Intell. Syst. 21, 18–21 (2006) 5. Aven, T.: On the meaning of a black swan in a risk context. Saf. Sci. 57, 44–51 (2013) 6. Calero Valdez, A., Brauner, P., Ziefle, M., Kuhlen, T.W., Sedlmair, M.: Human factors in information visualization and decision support systems. In: Workshop Human Factors in Information Visualization and Decision Support Systems Held as Part of the Mensch und Computer 2016. Gesellschaft für Informatik (2016) 7. Shim, J.P., Warkentin, M., Courtney, J.F., Power, D.J., Sharda, R., Carlsson, C.: Past, present, and future of decision support technology. Decis. Support Syst. 33(2) 111–126 (2002)

22

P. Brauner et al.

8. Gorry, G.A., Morton, M.S.S.: A framework for management information systems. Sloan Manag. Rev. 13, 50–70 (1971) 9. Kimball, R., Ross, M.: The Data Warehouse Toolkit: The Complete Guide to Dimensional Modelling. Wiley, New York (1996) 10. Codd, E., Codd, S., Salley, C.: Providing OLAP to User-Analysts: An IT Mandate (1993) 11. Bra, A., Lungu, I.: Improving decision support systems with data mining techniques. In: Advances in Data Mining Knowledge Discovery and Applications. InTech (2012) 12. Phillips-Wren, G.: Ai tools in decision making support systems: a review. Int. J. Artif. Intell. Tools 21(02), 13 pages (2012) 13. Shibl, R., Lawley, M., Debuse, J.: Factors influencing decision support system acceptance. Decis. Support Syst. 54, 953–961 (2013) 14. Althuizen, N., Reichel, A., Wierenga, B.: Help that is not recognized: harmful neglect of decision support systems. Decis. Support Syst. 54, 719–728 (2012) 15. Ben-Zvi, T.: Measuring the perceived effectiveness of decision support systems and their impact on performance. Decis. Support Syst. 54, 248–256 (2012) 16. Brauner, P., Calero Valdez, A., Philipsen, R., Ziefle, M.: How correct and defect decision support systems influence trust, compliance, and performance in supply chain and quality management – a behavioral study using business simulation games. In: HCI in Business, Government, and Organizations (HCIGO), Held as Part of HCI International. Springer (2017, in press). doi:10.1007/978-3-319-58484-3_26 17. Colman, A.M.: Oxford Dictionary of Psychology. Oxford University Press, Oxford (2015) 18. Weiner, B.: An attributional theory of achievement motivation and emotion. Psychol. Rev. 92, 548–573 (1985) 19. Graham, S., Folkes, V.S.: Attribution Theory: Applications to Achievement, Mental Health, and Interpersonal Conflict. Lawrence Erlbaum Associates, Hillsdale (1990) 20. Niels, A., Guczka, S.R., Janneck, M.: The impact of causal attributions on system evaluation in usability tests. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 3115–3125. ACM (2016) 21. Niels, A., Zagel, C.: Gamification: Der Einfluss von Attributionen auf die Motivation [The Influence of Attributions on Motivation]. In: Prinz, W., Borchers, J., Jarke, M. (eds.) Mensch und Computer 2016 - Tagungsband. Gesellschaft für Informatik e.V. (2016) 22. Arning, K., Ziefle, M.: Understanding age differences in PDA acceptance and performance. Comput. Hum. Behav. 23, 2904–2927 (2007) 23. Brauner, P., Leonhardt, T., Ziefle, M., Schroeder, U.: The effect of tangible artifacts, gender and subjective technical competence on teaching programming to seventh graders. In: Hromkovic, J., Královiè, R., Vahrenhold, J. (eds.) Proceedings of the 4th International Conference on Informatics in Secondary Schools (ISSEP 2010), Zurich, Switzerland. LNCS, vol. 5941, pp. 61–71. Springer, Heidelberg (2010) 24. Wittland, J., Brauner, P., Ziefle, M.: Serious games for cognitive training in ambient assisted living environments – a technology acceptance perspective. In: Abascal, J., Barbosa, S., Fetter, M., Gross, T., Palanque, P., Winckler, M. (eds.) Proceedings of the 15th INTERACT 2015 Conference. LNCS, vol. 9296, pp. 453–471. Springer, Cham (2015) 25. Stiller, S., Falk, B., Philipsen, R., Brauner, P., Schmitt, R., Ziefle, M.: A game-based approach to understand human factors in supply chains and quality management. Procedia CIRP 20, 67–73 (2014) 26. Sterman, J.D.: Modeling managerial behavior: misperceptions of feedback in a dynamic decision making experiment. Manag. Sci. 35, 321–339 (1989) 27. Goldratt, E.M., Cox, J.: The Goal: A Process of Ongoing Improvement. North River Press, Great Barringtons (1992)

The Interaction of Causal Attribution

23

28. Brauner, P., Calero Valdez, A., Philipsen, R., Ziefle, M.: Defective still deflective – how correctness of decision support systems influences user’s performance in production environments. In: Nah, F.F.-H., Tan, C.-H. (eds.) HCI in Business, Government, and Organizations (HCIGO), Held as Part of HCI International 2016, pp. 16–27. Springer, Cham (2016) 29. Schlick, C., et al.: Cognition-enhanced, self-optimizing production networks. In: Brecher, C., Özdemir, D. (eds.) Integrative Production Technology - Theory and Applications, pp. 645–743. Springer, Heidelberg (2017)

Gamified Self-service Checkouts: The Influence of Computer-Related Causal Attributions on User Experience and Motivation Adelka Niels1(&) and Christian Zagel2 1

2

Luebeck University of Applied Sciences, Mönkhofer Weg 239, 23562 Luebeck, Germany [email protected] Coburg University of Applied Sciences, Friedrich-Streib-Strasse 2, 96450 Coburg, Germany [email protected]

Abstract. This paper presents the concept and prototype of a smartphone application that gamifies the payment process at supermarket self-service checkouts. It is evaluated within a laboratory experiment conducted with 65 participants. As its primary goal, the study investigates the impact of computer-related causal attributions on the users’ evaluation of the gamified service. Attributions can be understood as individuals’ causal explanations for events influencing their emotions, motivation, and behavior. Results indicate that especially attributions in situations of success influence the level of motivation and user experience. The findings indicate that evoking specific attributions through the design of gamified applications can lead to increased user engagement. Keywords: Self-service checkouts
Gamification
User motivation
User experience
Service fascination
Personality traits
Attribution theory

1 Introduction According to a study conducted amongst 5,000 retailers and customers, 29% of the customers left a store without buying because of long queues at checkout stations [1]. Independent of the payment method used, these queues particularly originate from slow processing of the goods by the cashier. The study also found that 25% leave the store immediately after entering if the queue is perceived as too long. Several scientific studies show a direct correlation between waiting time at cash desks and the overall satisfaction with the shops. It also has been shown that shortening the checkout waiting time, which is usually perceived as boring and annoying, has a positive influence on customer loyalty [2]. Several endeavors have been undertaken to optimize the payment process at cash desks. A method to reduce the actual, and also the perceived waiting time is the implementation of self-service checkouts. By doing so, the customers themselves act as cashiers. Independent of human employees, customers can scan their products and pay cashless, e.g., with a credit card. These kinds of systems are currently used, for © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_3

Gamified Self-service Checkouts

25

example, in supermarkets and furniture stores. In contrast to the traditional checkout process, the customers themselves become active and the chance of boredom is largely reduced [3]. A recent example is the Swedish supermarket “Näraffär” that operates without human personnel – solely the customers complete the payment process through a smartphone application [4]. However, this only one of many supermarkets that allow consumers to pay via self-service checkouts or mobile payment applications, without the need of interacting with human cashiers [5–8]. Many of these processes are even realized without assistance personnel. Although the introduction of self-service checkouts or mobile payment solutions opens up benefits (e.g., reduced waiting time), some customers respond with resistance to adopting them. Reasons are, for example, a lack of users’ technology readiness [9], usability and user experience [10], a lack of trust towards the system [11], as well as missing enjoyment [12]. Thus, the use of technology-based services is highly influenced by the users’ personal characteristics. Experience, expertise, but also self-confidence and technology readiness play an important role in how people interact with such systems. This especially applies when problems arise [13]. In this regard, Attribution Theory could be a new and promising approach to explaining user behavior and to optimize self-service checkouts for meeting the users’ needs. Attribution theory deals with the explanations people find in situations of success and failure and correlates those beliefs to subsequent emotions, motivation, and behavior. Based on a newly developed smartphone application to gamify the shopping experience, this exploratory study investigates the influence of users’ causal attributions on the perception and evaluation of gamified self-service checkout applications. The study contributes to better understanding, whether the implementation of gamified shopping checkout applications is an appropriate method to increase customer motivation, as well as on how to design gamified checkout applications for users with different personality traits.

2 Related Work and Theoretical Background 2.1

Gamification of Shopping Experiences

A new method gaining attention, especially in relation to its positive effects on user motivation and efficiency, is gamification [14]. Gamification describes the application of game-type elements in non-game contexts with the goal of increasing user motivation and the creation of an enhanced user experience [15]. Hence, applying the concept promises an increase in productivity while simultaneously improving user satisfaction [16]. The method is applied in many ways in organizational contexts. Nevertheless, literature and practice still demand additional practical examples that prove its efficiency [16, 17]. While previous studies were able to prove a positive influence of gamification towards motivation and engagement in everyday processes [18], the payment process at supermarket self-service checkouts has not yet been investigated. Most attempts to gamify the shopping experience have been done in online shopping or during the shopping process inside shops. The website Groupon, for example,

26

A. Niels and C. Zagel

implemented the so-called “SOS mechanic”, where a certain amount of people needed to order a service in order to get a cheaper price [19]. This way the company motivated their customers to additionally market their promotions and to motivate others in also ordering them. Lounis et al. [20] examined the use of gamification to promote a sustainable buying behavior for fast moving consumer goods. They found that customers, in general, are willing to participate in gamified services. Nevertheless, the results cannot be generalized, as different personality traits demand different ways of implementing gamification elements. The authors even state that a gamification scheme needs to be customizable and personalized to become efficient. Gamification is also to be found to enhance the retail experience in online shopping in general. Insley and Nunan [21] propose to adjust the way of integrating game elements into the shopping experience in dependence of the customers’ shopping task (e.g., recreational vs. functional shopping). In addition, they found that gamification is able to influence consumer behavior, e.g., reducing undesirable actions as the misuse of postage policies. An example of gamification in brick and mortar comes from Starbucks [22]. They managed to increase customer engagement and sales by implementing a rewards loyalty program. Using a Starbucks card, consumers can track their expenses and level up through additional purchases. As a reward, they, for example, get free drinks. Already before the introduction of gamification as a specific method authors looked into the importance of consumer’s personality traits with a specific focus on user experience and enjoyment in the shopping context. Dabholkar and Bagozzi [23] found that the aspiration of fun when using a technology-based self-service might be seen as worth waiting for. In addition, these positive emotional attributes are especially important in situations where situational pressures are to be expected. While there are several examples of using gamification in loyalty programs and with the goal of driving additional sales in online and brick and mortar environments its application in the checkout process has not been researched, yet. The idea of this work is to increase customer motivation and to gain positive customer experience by integrating a challenge into the process, making it more attractive and less boring.

2.2

Attribution Theory

An attribution is a subjective causal explanation for an event or behavior that enables persons to orient and to control their environment [24]. Attributions are highly relevant for people’s behavior, emotions, and motivation [25, 26] and determine reactions to success and failure. Attributions can be classified along the four dimensions Locus, Stability, Controllability, and Globality [24]. Heider [27] differentiates causes with respect to the dimensions of Locus and Stability. The Locus dimension indicates whether the cause is located within (internal) or outside a person (external), influencing emotional reactions. For instance, in situations of failure, internal attributions are often associated with self-focused negative emotions, such as guilt and shame. Contrary, external attributions could be associated with external-focused negative emotions like anger or even aggression.

Gamified Self-service Checkouts

27

The Stability dimension describes the temporal constancy of a cause and distinguishes between recurring (temporally stable) and singular, nonrecurring (temporally unstable) causes, thus affecting individuals’ future expectations. For example, perceived causes such as intelligence or luck are generally considered as stable because they are difficult or impossible to change. Unstable causes, such as the effort invested, are comparatively easy to change [27]. A third level to classify attributions is known as the Controllability dimension: To what extent does a person have the ability to control his/her actions and surrounding conditions? For instance, in the above-mentioned scheme, intelligence stands for an internal-stable power factor. In addition, Controllability can be classified as being either a pre-existing, uncontrollable aspect (which cannot be influenced much) or as aneffort or working attitude, which constitutes a person’s conscious decision and therefore a controllable cause [28]. Finally, a fourth dimension refers to Globality. A cause is considered as global if it has similar effects in different situations (e.g., different applications), while it is specific if the cause is limited to a particular situation (e.g., specific application). A study conducted by Niels et al. [29] showed that users’ attribution patterns influence their evaluation and experience of computer systems. For instance, users who see mainly external reasons for success rated Quality of Use aspects, as well as Design Quality aspects, more positive than users with internal attribution patterns. Thus, they explain their success with good system qualities. Furthermore, users who perceived the reason for failure to be stable and global rated the tested systems more positive regarding the Quality of Use. Based on these results we expect that individual attribution patterns also have an influence on the perception, the evaluation, and the experience of gamified and technology-based self-service checkouts.

3 Research Questions and Methodology As a practical example, this paper presents a smartphone application that gamifies the payment process at supermarket self-service checkouts. It is evaluated within a laboratory experiment addressing the acceptance criteria described above. Hence we assess the gamified process in regards to User Experience, and enjoyment (Motivation, Service Fascination). To get a deeper insight about how the user’s personal characteristics influence the acceptance of gamified self-services, we link the results (User Experience, Motivation, Service Fascination) to the user’s Causal Attributions. The findings should provide implications for optimization of gamified services as well as for future research on self-service systems.

3.1

Process’ Suitability for Gamification

In order to identify the applicability of gamification methods to the checkout process in supermarkets in general, the classification method described in [30] is applied. It allows grading processes according to: (1) process-oriented (number of iterations, number of executors, standardizability, seclusiveness, compactness) and (2) emotional (motivation,

28

A. Niels and C. Zagel

emotion, feedback) criteria to evaluate a process’ suitability for gamification. The classification was done on the basis of expert interviews and observations in various supermarkets [31]. The results of the expert interviews indicate a good suitability, especially due to the high number of process iterations and executors as well as the lack of customer motivation.

3.2

Software Prototype

For realizing the gamification concept, a smartphone application in the form of a mobile-optimized website was implemented. The application offers the user the possibility to create and manage user accounts. Via the start screen, the user has the possibility to enable the scan function and employ an integrated QR code reader to identify the products captured. The game begins with scanning the first article and is terminated by clicking on the “checkout” button. The system displays the products’ name and price each time an article is detected. As already stated above, there are several possible ways to gamify the shopping process. However, most of them are not practicable for a laboratory context. Therefore, we decided to use a contest between the customers to realize the gamification aspect. A score is calculated based on the number of articles scanned and time needed. The goal of the game is to scan the articles in the shopping basket as quickly as possible. For a real world scenario this would mean: The faster a shopping basket is processed, the higher is the number of customers handled per line, resulting in shorter queues. After finishing the payment process, the user’s score is displayed in a leaderboard, providing insight on one’s performance in relation to other customers. The clear task, the ranking list, the direct feedback when scanning items, and the result transparency focus on increasing user motivation [32].

3.3

Measurement

Next to the assessment of general demographic aspects (age and gender), four validated questionnaires were used covering the aspects of User Experience, Motivation, Service Fascination, and Computer-related Causal Attributions. The User Experience Questionnaire (UEQ) developed by Laugwitz et al. [33, 34] is used to measure user experience, focusing on perceived Attractiveness, Quality of Use, and Design Quality of software systems. It consists of 6 scales with 26 items to measure the factors Attractiveness, Perspicuity, Efficiency, Dependability, Stimulation, and Novelty. While Attractiveness is a pure valence aspect, Perspicuity, Efficiency, and Dependability mark pragmatic, goal-oriented quality aspects (Quality of Use). Stimulation and Novelty, in contrast, describe hedonic, non-goal-oriented, quality aspects (Design Quality). The questionnaire consists of word pairs of contrasting attributes that may apply to the tested process, system, or software. The items are arranged in the form of a seven-stage semantic differential. The Motivation Questionnaire developed by Zagel and Bodendorf [30, 31], is used to measure the motivational effects of gamification. It is applied to rate processes and

Gamified Self-service Checkouts

29

tasks based on the user’s intrinsic motivation to actively perform, covering aspects of competition, intuitiveness, feedback, repetition, and the level of boredom mentioned in the gamification literature [15]. Since the UEQ is already capturing items to measure pragmatic quality, only items assessing the willingness to repeat the process, as well as the willingness to improve the own performance were integrated. Service Fascination is described as an extraordinary positive emotional state arising through conscious and subconscious effects of technological self-service use (enjoyment). The Service Fascination Questionnaire developed by Zagel [35] focuses on assessing these effects, manifested in the form of excitement, active word-of-mouth, and the willingness to repeatedly use a technology in the future. The measures for Motivation and Service Fascination are shown in Table 1. All items were rated on a seven-point Likert-type scale labeled at the end points (1 = I strongly agree, 7 = I strongly disagree). Table 1. Items to measure Motivation (M) [30] and Service Fascination (SF) [35]. Item M1 M2 M3 M4 SF1 SF2 SF3 SF4 SF5

Description The system/the process motivates me in solving my task I am motivated to repeat my task I am motivated to improve my own performance when repeating the task I feel allured in challenging myself with others when solving the task I would share my good experience about using the system I would recommend using the system Using the system is exciting Given that I have access to the system, I predict that I would use it I will frequently use the system in the future

The standardized Attribution Questionnaire developed by Guczka and Janneck [36] was used to measure computer-related attributions. The questionnaire distinguishes between success and failure situations, as usually done in attribution research, and contains four items to measure the attributional dimensions of Locus, Stability, Controllability and Globality. Table 2 shows the questionnaire relating to situations of failure. Items measuring attributions of success are worded analogously. Table 2. Excerpt from the Attribution Questionnaire for failure situations [36].

30

3.4

A. Niels and C. Zagel

Procedure

First, participants were asked to provide demographic data and to perform a payment process using the gamified application. After a short introduction of the applications’ functionality, the participants were provided with a shopping bag consisting of ten articles (e.g., a bar of chocolate, rice, wine). They were told that the purpose of the game is to scan the articles as fast as possible and their goal is to reach the first rank in the leaderboard. In order to drive motivation, participants were told that there is the possibility to win a 50€ voucher for the participant reaching the first place in the competition. The test scenarios, as well as the laboratory and light conditions (Samsung Galaxy 5 Neo smartphone, 249 l illumination), were identical for each participant. They were allowed to arrange the articles in their preferred way on a given table/cash desk. Each of the items carried a QR code that can be read by the application and that is used to identify the products. With regard to the attributions, we purposely used different product packages with varying recognition quality to ensure that situations of success and failure will occur during the process. For example, a code on a flat box was easier to detect than one that was attached to the curved surface of a bottle or the uneven surface of a rice bag. After finishing the process, participants were asked to provide feedback by means of the User Experience Questionnaire, as well as the Motivation and Service Fascination Questionnaire. The respondents’ attributions were then assessed using the Attribution Questionnaire for both, success and failure situations occurring during the gamified task. Since attribution theory is about individuals’ subjectively perceived situations, it was important that the participants decided by themselves what they considered as a success or failure (e.g., high ranking at the leaderboard, an article was hard to scan).

3.5

Sample

The laboratory experiment was conducted in February 2016 at the business faculty of a university in Bavaria, Germany. In total, 65 persons participated (23 females, 42 male), aged between 19 and 55 years (M = 25.65; SD = 6.62). The participants on average reached a score of 5.92, which reflects the average time in seconds needed to scan anarticle (min = 3.4, max = 15.4; SD = 2.553). The participation was voluntary and the participants were not paid for taking part in the study. They always had the possibility to drop out if they did not want to finish the experiment.

4 Data Analysis and Results 4.1

User Experience, Motivation and Service Fascination

The User Experience Questionnaire data was analyzed using the Excel Data Analysis Tool (www.ueq-online.org) provided by Schrepp and colleagues [34]. The first step of the analysis is to transform the item values as the order of positive and negative terms is randomized in the questionnaire. The resulting unified scale ranges from +3 to −3. Mean values were calculated for each scale. Values 0.8

Gamified Self-service Checkouts

31

Table 3. Results user experience, motivation and service Fascination Questionnaire. Aspect Mean Motivation and service fascination 0.97 UEQ attractiveness 1.14 UEQ quality of use 1.12 UEQ design quality 1.15

SD 1.21 0.90 0.76 0.74

a .908 .864 .828 .766

represent a positive evaluation [37]. In order to simplify the evaluation, the six UEQ-Scales were summarized to the three aspects Attractiveness, Quality of Use, and Design Quality. Moreover, the mean values for Motivation and Service Fascination were added. The results show a positive evaluation for all aspects (Means, standard deviations, and Cronbach’s Alpha scores are shown in Table 3).

4.2

Computer-Related Causal Attributions

In our analysis, we distinguished between situations of success and failure, as it is usually done in attribution research. Mean value for each attributional dimension was calculated (Table 4). Results show that the participants predominantly see internal reasons for their success and failures. Furthermore, the cause is perceived to be stable over time, persists in different situations (global), and is perceived as controllable. Overall, this represents a more positive attributional pattern.

Table 4. Results Attribution Questionnaire. Attributional dimension Success Mean SD Locus 3.14 1.54 Stability 5.26 1.35 Controllability 3.03 1.47 Globality 4.52 1.39

4.3

Failure Mean SD 3.75 1.79 5.05 1.45 3.14 1.57 4.03 1.53

Influence of Computer-Related Causal Attributions

To analyze the relations between Causal Attributions and User Experience, as well as Motivation and Service Fascination towards the gamified self-service process, correlations (Spearman’s Rho) were calculated. Results are shown in Table 5. In situations of success, data analyses revealed significant correlations between the attributional dimensions of Stability, Controllability, and Globality and the evaluation criteria. Merely for Locus, no differences were found. With respect to the Stability dimension, positive correlations for all evaluation criteria were found. Thus, persons who perceive causes for success as stable over time rated the evaluation criteria as more positive than users with low stability values.

32

A. Niels and C. Zagel

Table 5. Results correlation analysis (MSF = Motivation and Service Fascination; ATT = UEQ Attractiveness; QU = UEQ Quality of Use; DQ = UEQ Design Quality; LOC = Locus, STA = Stability; CON = Controllability; GLO = Globality). Bold font mean significant difference.

The Controllability dimension correlates with all evaluation criteria too. In this case, correlations are negative since high values on the Controllability dimension indicate a low control perception. This means that users with low control perception rated all evaluation criteria as more negative than users with high control perception. Finally, the Globality dimension correlates significantly with the three UEQ-Aspects of Attractiveness, Quality of Use, and Design Quality. Thus, users who see global causes for their successes rated the process more positive than users that attribute success to a specific situation. In situations of failure, significant correlations for the attributional dimensions of Locus and Controllability were found. For Stability and Globality data analyses revealed no significant correlations. For the Locus dimension, results showed a significant negative correlation for Quality of Use. Users who blame themselves for the failure rated this aspect more positive than users who ascribe the failure to external reasons. Just like in situations of success, the Controllability dimension showed significant negative correlations for all evaluation criteria. Users with low control perception rated the evaluation criteria as more negative than users with high control perception.

5 Discussion The aim of this study was to investigate the impact of user’s Causal Attributions on the User Experience, as well as the Motivation and Service Fascination within gamified self-service checkout processes. The results suggest that this is mainly true. Regarding situations of success, correlation analyses showed significant differences for three of the four dimensions, namely Stability, Controllability, and Globality. Merely for the Locus, dimension no differences were observed. Consequently, for the users, it makes no difference if they see the reason for the success in themselves or the system. A better user experience is generated if the users

Gamified Self-service Checkouts

33

expect to succeed again in the future (stable) and believe having enough control to influence the cause of success. This will also increase the willingness to use the system again and lead to a positive attitude towards sharing the good experience with others. Moreover, high values for Globality indicate that a person believes that success is present in different contexts (e.g., other applications, devices or tasks). Thus, it reflects an overall very confident handling of technology-based systems (“I can get on well with technology”). Interestingly, this has an impact on the user experience in general (perceived attractiveness, hedonic quality, pragmatic quality), but will not affect the motivation to use the system in the future again. Thus, the sense of competence has no effect on the willingness to use such systems or to improve the own performance. So if the user finally attributes a successful interaction outcome to stable, global, or controllable causes this leads to better user ratings. However, for situations of failure merely the influence of the Controllability dimension could be successfully proven in all aspects. A high control perception leads to a positive user experience. Furthermore, a negative correlation between Locus and Quality of Use was found. This means persons who blamed themselves for the failure rated this aspect more positively than persons who blamed the system. Thus, if the user believes that he/she can improve his/her performance (e.g., “If I learn how to scan the articles faster, I will get a better ranking on the leaderboard”), this leads to more enjoyment and an increase in motivation. Further, this also increases the perceived intuitiveness, efficiency and dependability of the system.

5.1

Implications and Recommendations

From an academic point of view, this study contributes to a better understanding of how causal attributions as personality traits affect the (initial) adoption of gamified self-service checkouts. In particular, this study assesses the suitability of using Attribution Theory as a new approach to explain user experience and behavior. Since this study could largely confirm the influence of causal attributions, it motivates for future research in this area and shows potential worth exploring by testing these effects e.g., with other gamification approaches in the context of self-service checkouts. There are also implications for practitioners. First, the results suggest that it is important to keep the users’ personality traits in mind when developing or designing such systems. In this regard, Attribution Theory could be a promising approach to provide a positive User Experience to costumers with unfavorable attributions towards gamified self-service checkouts. For instance, manipulating a user’s causal attributions by providing positive attribution-relevant information could help to increase their intention to use such services. This could be done, for example, by including emotionally positive and informative feedback. This could include comments that contain the desired attributions. A good score (success) could, for example, be commented with “Once again (stable) you have (internal) achieved good results. You can improve your score by arranging the articles before starting to scan (controllability)”. A poor score (failure), in contrast with “The QR code reader (external) has problems in low light conditions (controllable)”. Designers should bear this in mind and future research should take this into consideration.

34

5.2

A. Niels and C. Zagel

Limitations and Future Research

On one hand, a laboratory approach may be viewed as making the findings less applicable to the real world. On the other hand, situational factors in a field setting are usually uncontrollable. Using a realistic scenario that the respondents can easily imagine themselves in allows the researcher to control treatments and test outcomes effectively. Nevertheless, tests in a real-life scenario need to be done in order to prove these findings. In turn, a real-life scenario would enable to examine the effects of gamification on the actual waiting time. Further, it would be interesting to investigate the level of stress perceived during such a process. Next to its laboratory setting, the study is limited by the relatively small number of 65 participants and the focus on only one technological solution. Hence, additional versions of the concept will be implemented, compared and tested. One aspect excluded from the analyses in this study was evaluating the effects of user attributions and gamification methods towards the score achieved. This issue will be suspect of future studies. Next to a simple time score, as presented in this paper, other targets (e.g., healthiest shopping basket) could further improve user engagement and the willingness to use the self-service.

References 1. Epson. http://www.epson.de 2. Hinterhuber, H., Stahl, H: Erfolg durch Dienen? Beiträge zur wertsteigernden Führung von Dienstleistungsunternehmen. Expert Verlag, Renningen (2002) 3. Boslau, M.: Kundenzufriedenheit mit Selbstbedienungskasse im Handel: Der Erklärungsbeitrag ausgewählter verhaltenswissenschaftlicher Theorien. Springer Gabler, Wiesbaden (2009) 4. Heise. http://www.heise.de 5. Dabholkar, P., Bobbitt, M., Lee, E.J.: Understanding consumer motivation and behavior related to self-scanning in retailing: implications for strategy and research on technology-based self-service. Int. J. Serv. Ind. Manag. 14, 59–95 (2003) 6. Orel, F.D., Kara, A.: Supermarket self-checkout service quality, customer satisfaction, and loyalty: empirical evidence from an emerging market. J. Retail. Consum. Serv. 21, 118–129 (2014) 7. Lowe, M., Wrigley, N.: Innovation in the retail internationalization: Tesco in the USA. Int. Rev. Retail Distrib. Consum. Res. 19, 331–347 (2009) 8. Strudthoff, M.: Die mobile Revolution im Handel definiert die Zukunft des Bezahlens neu. In: Linnhoff-Popien, C., Zaddach, M., Grahl, A. (eds.) Marktplätze im Umbruch, pp. 399– 407. Springer, Heidelberg (2015) 9. Lee, H.J., Jeong, H., Wenwen, C., Fairhurst, X.A.: The influence of consumer traits and demographics on intention to use retail self-service checkouts. Mark. Intell. Plan. 28, 46–58 (2010) 10. Dabholkar, P.A.: Consumer evaluations of new technology-based self-service options: an investigation of alternative models of service quality. Int. J. Res. Mark. 13, 29–51 (1996) 11. Wang, C., Harris, J., Patterson, P.: Customer choice of self-service technology: the roles of situational influences and past experience. J. Serv. Manag. 23, 54–78 (2012)

Gamified Self-service Checkouts

35

12. Elliott, K., Hall, M., Meng, J.: Consumers’ intention to use self-scanning technology: the role of technology readiness and perceptions toward self-service technology. Acad. Mark. Stud. J. 17, 129–143 (2013) 13. Janneck, M., Vincent-Höper, S., Ehrhardt, J.: The computer-related self concept: a gender-sensitive study. Int. J. Soc. Organ. Dyn. IT (IJSODIT) 3, 1–16 (2013) 14. Gamification. http://www.gamification.co 15. Deterding, S., Sicart, M., Nacke, L., O’Hara, K., Dixon, D.: Gamification: using game-design elements in non-game contexts. In: CHI 2011 Extended Abstracts on Human Factors in Computing Systems, pp. 2425–2428. ACM, New York (2011) 16. Gonzales-Schaller, P.: Trendthema Gamification: Was steckt hinter diesem Begriff? In: Diercks, J., Kupka, K. (eds.) Recrutainment, pp. 33–51. Springer, Wiesbaden (2013) 17. Hamari, J., Koivisto, J., Sarsa, H.: Does gamification work? – A literature review of empirical studies on gamification. In: Proceedings of the 47th Hawaii International Conference on System Science, pp. 3025–3034 (2014) 18. Neeli, B.K.: A method to engage employees using gamification in BPO industry. In: Third International Conference on Services in Emerging Markets (ICSEM), pp. 142–146. IEEE (2012) 19. Zicherman, G., Cunningham, C.: Gamification by Design: Implementing Game Mechanics in Web and Mobile Apps. O’Reilly, Cambridge (2011) 20. Lounis, S., Neratzouli, X., Pramatari, K.: Can gamification increase consumer engagement? A qualitative approach on a green case. In: Douligeris, C., Polemi, N., Karantjias, A., Lamersdord, W. (eds.) Collaborative, Trusted and Privacy-Aware E/M-Services, pp. 200– 212. Springer, Heidelberg (2013) 21. Insley, V., Nunan, D.: Gamification and the online retail experience. Int. J. Retail Distrib. Manag. 42, 340–351 (2014) 22. Conaway, R., Garay, M.C.: Gamification and service marketing. SpringerPlus 3, 1–11 (2014) 23. Dabholkar, P., Bagozzi, R.: An attitudinal model of technology-based self-service: moderating effects of consumer traits and situational factors. J. Acad. Mark. Sci. 30, 184– 201 (2002) 24. Försterling, F.: Attribution: An Introduction to Theories, Research, and Applications. Psychology Press, Hove (2001) 25. Abramson, L.Y., Seligman, M.E., Teasdale, J.D.: Learned helplessness in humans: critique and reformulation. J. Abnorm. Psychol. 87, 49–74 (1978) 26. Weiner, B.: Achievement Motivation and Attribution Theory. General Learning Press, New York (1974) 27. Heider, F.: The Psychology of Interpersonal Relations. Wiley, New York (1958) 28. Weiner, B.: An Attributional Theory of Motivation and Emotion. Springer, New York (1986) 29. Niels, A., Guczka, S.R., Janneck, M.: The impact of causal attributions on system evaluation in usability tests. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI 2016), pp. 3115–3125 (2016) 30. Zagel, C., Bodendorf, F.: Gamification: Auswirkungen auf Usability, Datenqualität und Motivation. In: Mensch und Computer Tagungsband 2014. Oldenbourg, München, pp. 15– 24 (2014) 31. Zagel, C., Niels, A.: Einsatz von Gamification zur Steigerung der Kundenmotivation beim Bezahlprozess an Supermarktkassen. Medienproduktion - Online Zeitschrift für Wissenschaft und Praxis 8, 8–12 (2015)

36

A. Niels and C. Zagel

32. Baumann, K., Klein, P., Carl, A., Bender, D.: Gamification in the inDAgo HelpMe application. In: Mensch und Computer Tagungsband 2014. Oldenbourg, München, pp. 25– 34 (2014) 33. Laugwitz, B., Held, T., Schrepp, M.: Construction and evaluation of a user experience questionnaire. In: Proceedings of the HCI and Usability for Education and Work, pp. 63–76 (2008) 34. Schrepp, M., Hinderks, A., Thomaschewski, J.: Applying the User Experience Questionnaire (UEQ) in different evaluation scenarios. In: Marcus, A. (ed.) Design, User Experience, and Usability. Theories, Methods, and Tools for Designing the User Experience, pp. 383–392. Springer, Cham (2014) 35. Zagel, C.: Service Fascination – Gaining Competitive Advantage Through Experiential Self-Service Systems. Springer Fachmedien, Wiesbaden (2016) 36. Guczka, S.R., Janneck, M.: Erfassung von Attributionsstilen in der MCI – eine empirische Annäherung. In:Mensch und Computer 2012: interaktiv informiert – allgegenwärtig und allumfassend? Oldenbourg, München, pp. 223–232 (2012) 37. Analyzing the user experience questionnaire. http://www.ueq-online.org

Defining Recrutainment: A Model and a Survey on the Gamification of Recruiting and Human Resources Oliver Korn(&), Florian Brenner, Julian Börsig, Fabio Lalli, Maik Mattmüller, and Andrea Müller Offenburg University, Badstr. 24, 77652 Offenburg, Germany [email protected], {fbrenner,jboersig,flalli, mmattmue}@stud.hs-offenburg.de, [email protected] Abstract. Recrutainment, is a hybrid word combining recruiting and entertainment. It describes the combination of activities in human resources and gamification. Concepts and methods from game design are now used to assess and select future employees. Beyond this area, recrutainment is also applied for internal processes like professional development or even marketing campaigns. This paper’s contribution has four components: (1) we provide a conceptual background, leading to a more precise definition of recrutainment; (2) we develop a new model for analyzing solutions in recrutainment; (3) we present a corpus of 42 applications and use the new model to assess their strengths and potentials; (4) we provide a bird’s eye view on the state of the art in recrutainment and show the current weighting of gamification and recruiting aspects. Keywords: Recrutainment
Gamification Human resources
Human factors


Recruiting
Applicant selection


1 Introduction In today’s competitive job market, companies face many challenges concerning the recruiting process. The costs for hiring the “wrong” person for a specific job are very high. Thus, innovative companies are searching for possibilities to improve the recruiting process to attract and identify ideal candidates [1]. This task is even more difficult with job beginners, i.e. young persons without work experience. The number of potential young candidates is lower than it used to be. This is a natural result of the demographic transition, especially in Europe and Japan but also in the United States and many other industrialized countries. To attract young high potentials, either companies can use traditional concepts, spending large budgets on image campaigns and advertising for employer branding, or they break new ground [2]. Special events like recruiting fairs at high schools and universities or using the power of social networks [3] have almost become standards. On the other hand, too many applicants can become a time-consuming problem, as the applications need to be screened, assessed and finally job interviews are to be conducted. © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_4

38

O. Korn et al.

Especially the “Generation Y” tends to apply at multiple companies. At the same time, the applicants are unsure, which job provides the best fit. To reduce this insecurity it would be very beneficial for the applicants to get the right information on a job before they apply [4]. However, what motivates potential applicants to engage for a yet unknown company and to deal intensely with job details? Can solutions from other domains provide assistance? If so, how can they be adapted to be effective in the human resources business? “Recrutainment” may well contribute to address this dilemma. The term is a hybrid word combining the elements recruiting and entertainment, probably coined by the German company Cyquest [5]. It implies a combination of concepts and methods from the areas human resources and gamification, with gamification being defined as the use of elements from game design in real-world scenarios [6]. This work aims to provide a pioneering overview of recrutainment. The guiding research questions were: (1) how can recrutainment applications be described and analyzed and (2) what defines a good recrutainment solution, suited for the human resources business? Based on several trends and observations (Sect. 2) we will develop a model (Sect. 3) to assess recrutainment applications. We will then present a corpus of recrutainment applications, which are assessed using the model (Sect. 4), and conclude with a bird’s eye view on the state of the art in recrutainment (Sect. 5).

2 Background Since recrutainment is a relatively young method, it has hardly been described specifically in scientific literature. However, it is linked to several other concepts and societal trends, which provide a background. In this section, we will briefly outline the most important aspects: gamification, assessment, gaming and the “Generation Y”.

2.1

Gamification

As pointed out in the Introduction, recrutainment strongly builds on the concept of gamification, the use of game mechanics in a non-gaming context. The term has been described in detail by Deterding [6, 7]. However, gamification also is a contemporary umbrella term, which builds on ideas and concepts previously described as “serious games” when gamifying education [8], exergames when gamifying health [9, 10] or simply as “applied games” or “games with a purpose” [11]. Although these terms are not synonyms but relate to different research cultures and domains, they all share the aspect of “gamefulness” [12] which is the essence of the modern term gamification.

2.2

Assessment

Software tools for assessment are computerized tests, which evaluate the abilities or characteristics of persons. There is a large set of both standardized test batteries for motor [13] and cognitive functions. However, the cognitive assessment tools for unimpaired persons, usually are called “intelligence tests” [14].

Defining Recrutainment: A Model and a Survey

39

Furthermore there is a wide range of “attitude” or “aptitude” tests, which intend to identify the employees with the best person-environment fit, the latter being defined by Kristof-Brown et al. as “the compatibility between an individual and work environment that occurs when their characteristics are well matched” [15]. As pointed out in the Introduction, hiring the wrong candidate (or in Kristof-Brown’s terms: a badly matched candidate) is expensive: it wastes time, money and potentially damages the person’s self-confidence. Assessment tests try to minimize this risk. Classic assessment, for example at an assessment center, is costly – however, online assessment or assessment by mobile applications (apps) is much cheaper. While these modern versions of assessment may not always measure the same constructs as their traditional antecedents [16], they surely can serve as a first gate to streamline the flow of applicants.

2.3

Gaming

In 2016, the global game market generated an amount of 99.6 Bn USD with an estimated year-on-year growth rate of 8.5% [17]. Of this total, 27.1 Bn (27%) are generated on mobile phones and 9.8 Bn (10%) on tablets. In the US over 90% of the population has online access and about 55% play games. In Europe, the numbers are similar (86% have online access, 45% play games). Clearly, gaming has become a megatrend in our society. The market has expanded to such a size that non-gaming domains try to exploit it, too. Gamification is not only a good concept to motivate users – it also brings areas like business, health, or education closer to the attractive world of gaming and the large number of gamers. This can have a positive impact on the perceived attractiveness of the companies and institutions using it. Game elements like points, rankings, rewards or techniques like storytelling are common knowledge. Integrating them into assessment solutions(ideally called “job game” or similarly) increases their attractiveness and thus the potential number of participants.

2.4

Generation Y

The term “Generation Y” (also “millennials”) was coined in 2012 by Horovitz [18]. It is used to describe the cohort of people born from the early 1980s to early 2000s. Most of them have grown up with the internet and mobile phones. Now they are entering employment in large numbers. As Koudal and Chaudhuri point out, members of Generation Y are “entrepreneurial by nature, enjoy electronic games, place a high value on innovation, and are comfortable working in teams”. They appreciate flexible working times and regular feedback. If the conditions do not fit their desire for self-actualization, they will quickly leave to fulfil their ambitions elsewhere. This mind-set puts additional emphasis on finding the right candidate for the job, as the millennials are less willing to “adapt to fit in” than most previous cohorts [19].

40

2.5

O. Korn et al.

Synthesis: A Definition of Recrutainment

Based on the discussion of the background, we provide a definition of recrutainment: In Recrutainment, methods of cognitive assessment, aptitude assessment and gamification elements are brought together and embedded into a company’s recruiting process. The intentions are to attract more and younger candidates, and to gain better assessments results. This allows pre-selecting the candidates with the best personenvironment fit.

3 A Model for Recrutainment Applications Our leading research questions are to describe and analyze recrutainment applications as well as assess their quality. Thus, a theory-based model is required. Existing models to evaluate gamified applications lack the recruiting aspects while recruiting models fail to address the entertainment aspect. In this section, we present a new model, which combines elements from both fields. As a foundation we will introduce two reference models to build from: the gamification-related Octalysis Framework by Chou [20] and an approach to recruiting by Kristof-Brown and Guay [21], the person-environment fit model.

3.1

Foundations: Octalysis Framework and Person-Environment Fit Model

The Octalysis Framework is a representative for existing gamification frameworks. On its first level, it defines eight “core drivers” indicating whether a gamified application employs the engaging elements from the following list: 1. Epic Meaning and Calling: the feeling of doing something great. 2. Development and Accomplishment: the will to progress, improve and ultimately overcome a challenge. Examples design elements are leader boards or badges. 3. Empowerment of Creativity and Feedback: engages users in creative processes, where they can improve or alter their creations and see the results. 4. Ownership and Possession: designates the feeling of controlling something, resulting in the will to increase or improve it. 5. Social Influence and Relatedness: uses social elements to engage the user, for example companionship and competition. 6. Scarcity and Impatience: points at the desire to possess something rare and exclusive. 7. Unpredictability and Curiosity: create engagement through not knowing what is coming next. 8. Loss and Avoidance: engage the user to avoid something bad happening. This short overview of the Octalysis framework shows that it lacks “serious” business-related components, which are important when evaluating recrutainment applications. The focus of the framework lies on engagement: defining the core drivers when designing a gameful application. Chou states that such applications must not

Defining Recrutainment: A Model and a Survey

41

excel in every driver to engage the users. In contrast, it might be possible to excel in one and still be motivating. Indeed, for creating a model of recrutainment applications this is helpful: several of the drivers require complex game mechanics to be implemented and these are usually not a part of current recrutainment applications. The person-environment fit model is used as a basis for the recruiting aspects. It is compatible with self-assessments, which play a major role in most recrutainment applications. Similar to the “drivers” in the Octalysis framework, Kristof-Brown & Guay divided the model into several more specific “fits”: person-organization fit, person-job fit, person-group fit and person-person fit. In the following, we will only look at the “fits” relevant for recrutainment: 1. Person-vocation. A recrutainment application should let the user assess, if the vocation he or she is interested in is suitable to the personal needs, abilities and interests. 2. Person-job describes the compatibility between a person’s characteristics and those of a specific job. It goes hand in hand with comparing the job requirements with the qualifications of a person. If an assessment is designed for a specific job, it must help the user to determine if the job’s typical tasks fit the personal skillset. 3. Person-organization. The authors suggest that persons sharing the same values as companies perform best. Both ethical aspects and organizational structures and processes affect this fit.

3.2

Recrutainment Model

The new model for describing and analyzing recrutainment applications combines criteria from gamification models, with the goal of creating engagement, with criteria from recruiting models, with the goal of finding the candidate with the “best fit”. There are eight criteria in the Recrutainment model (Fig. 1):

Interactivity

Challenge

Recrutainment

Competition

Visual Attractiveness Fig. 1. Recrutainment model

42

O. Korn et al.

1. Interactivity (related to basic criteria 3) describes the extent of possible interactions within the recrutainment application. It ranges from no interaction at all to simple interactions like point and click to complex interactions like simulations. Interactivity relates to elements from gamification models; it improves users’ engagement. We assume that engagement correlates with the size of the solution space: a larger solution space with complex forms of interactions creates a feeling of freedom and suggests the potential to achieve something great. At the same time, complex interaction can overburden users and reduce usability and thus acceptance. 2. Narration (related to basic criteria 1) evaluates the level of storytelling in an application. It ranges from no story to complex stories, involving role playing and the appearance of different characters. While a good narration can be highly engaging and immersive, a lack of storytelling does not equal a bad application. As pointed out above with respect to the “core drivers”, not every driver is required to engage the users. However, this element is interesting as it connects to both the gamification and the recruiting model: an involving job-related story that enlivens the potential future work tasks surely helps to assess the person-job, person-organization and person-vocation fits. 3. Competition (related to basic criteria 2 and 8) evaluates the level of competition amongst users from “no competition” to real-time challenges. Under the label “social influence and relatedness” it is also one of Chou’s core drivers. Just like narration, this element gains importance by the fact that it is relevant for both gamification and recruiting. Competition can be abstract (for example in form of pre-generated leader boards) or very direct, if users are competing against each other. While such social competition clearly has the strongest effect on engagement [22], it is a demanding implementation task. However, it anonymity is guaranteed, social competition will improve the quality of the assessment results. Especially the examination of the person-job fit can benefit, as it incorporates data on how a candidate is performing in comparison to others. 4. Real-world Connection (related to basic criterion 6) evaluates how well the recrutainment solution connects to the real word. In recrutainment, “real world” means: does the candidate get useful feedback with regard to the job? The lowest score indicates no feedback at all, so the user receives neither a skill assessment nor an advantage in the application process. For the highest score, the successful completion of tasks in the recrutainment application results in an immediate invitation to a job interview. The importance of the real world connection currently marks one of the biggest differences to gamification applications, where this connection is helpful but optional. Without feedback, playing a recrutainment application would be senseless, while getting points and achievements for cleaning activities (to name one of many funny examples of personal gamification) might still be fun or at least funny. This criterion is also crucial for the person-environment fit, as the feedback can correct or specify the self-assessment and thus the user’s decision whether the job, vocation or organization is suitable. 5. Visual Attractiveness (related to basic criterion 9) evaluates the visual quality of the application’s user interface, its “hedonic” quality [23]. This is one of the more subjective categories and not directly connected to the recruiting aspect. However, it has a big influence on the acceptance of the recrutainment application. Furthermore,

Defining Recrutainment: A Model and a Survey

43

the visual appearance of applications directly related to a specific company, also affects this company’s perception or image. A low score shows that the design does not fit the company and the target group’s expectations. However, a high score does not automatically mean that high-end 3D graphics are applied: a recrutainment application can be very conservative and minimalistic – and still visually attractive, as it this might reflect the company’s image or the target group’s preferences. 6. Challenge (related to basic criteria 5 and 7) evaluates how well application address the target group’s abilities and expectations. Challenge and engagement go hand in hand, as only the mastery of an adequate challenge creates a feeling of accomplishment. This element is also important for recruitment, as the difficulty should map the skills of the users. Like visual attractiveness, a high challenge does not equal a good recrutainment application. The adequate difficulty depends on the target group and the objective of the application. An assessment in a late stage of the application process should be much more demanding than a first self-assessment. The scores range from no adequate challenge to a challenge that exactly fits the target group and allows direct comparisons. An adequate level of challenge will promote the person-environment fit. Especially if the application is designed for a demanding job profile, unsuitable candidates will have a negative feedback that is less embarrassing and costly than failure in a social setting. 7. Personalization (related to basic criteria 4, 6 and 10) evaluates the degree in which a user can individualize his or her representation within the application. The score ranges from no possibilities of personalization to being able to adapt an avatar in many respects, e.g. facial regions. A personalized experience will increase intrinsic motivation and engagement [24] and is well-connected to the gamification aspects. However, real engagement will also increase the reliability of the results from challenges and assessments. Furthermore, seeing oneself involved in potential future tasks will intensify the person-environment fit in both directions (agreement/denial), so they can better decide if the organization, job or vocation fits them. 8. Job-relatedness (related to basic criteria 5 and 11) evaluates how well the application is connected to a certain job or domain. As the name implies, it strongly connects with the recruiting aspect. A low score shows that the application has almost no connection to a job or branch. This can be the case if application primarily want to attract interest of candidates or even of consumer. The highest score indicates that the user is deeply immersed in the perspective of an employee, getting to know both the profession and the company. This criterion is crucial for the person-environment fit and contributes to the ability to assess the person-vocation, person-job and person-organization fit. Job specification mirrors personalization: if both elements are highly developed, the user will feel immersed and behave authentically. Naturally, this will create the most reliable assessment results.

44

O. Korn et al.

4 Study on Recrutainment Applications In this study, we present a large corpus of 42 recrutainment applications and use the Recrutainment Model (described in the previous section) to evaluate most of them. We could not evaluate 17 of the 42 applications (40%), as some are no longer accessible or require a formal registration. Under the abbreviation “Eval.” we indicate the ones, which were evaluated. In Table 1, we list the organizations related to the application and the URL where we found it (in early 2017). We do not provide application titles, as for several applications these are not explicitly mentioned.

Table 1. Corpus of 42 recrutainment applications. Organization Aldi Allianz Allianz Ausbildung.de BNP BP GCHQ Commerzbank Covestro Deloitte ElVi EY Formaposte Fresenius Gruner+Jahr Haniel Heineken hkp Knack Krones AG L’Oreal Lidl Lidl

URL http://www.aldicareers.com.au/The-ALDI-in-You https://karriere.allianz.de/oneweb/ajax/aspro-stateful/rp0/staticresources/JobMatcher/auswahl_schulabschluss.html https://www.allianz.com/de/karriere/ihr-einstieg/studenten/jobmatcher.html/ http://www.ausbildung.de/ http://acemanager.bnpparibas.com/ http://www.bp-dateiserver.de/videoquiz/ http://www.canyoucrackit.co.uk/ http://www.probier-dich-aus.de/ https://karriere.covestro.de/de/schueler/bewerbungsverfahren/ http://yourfuture.deloitte.de/online-praktika/ http://m.elementare-vielfalt.de/ausbildungsfinder http://www.young-targets.com/2016/10/17/ey-escape-game/ http://formaposte-iledefrance.fr/jeu-facteur-academy/ https://karriere.fresenius.de/de/services/fresenius-navigator http://recruitingportal.guj.de/selfAC/preloader.html https://www.haniel.de/fileadmin/content/01_unternehmen/03_ werte/Werte-Kompass/_de/index.html http://goplaces.theheinekencompany.com/en/age-gate http://www.hkp.com/de/karriere/der-hkp-recruitomat/ https://www.knack.it/ https://apps.facebook.com/kronese/index_ohne.php http://www.brandstorm.loreal.com/en/challenges/playexperiment-innovate http://lidl.cyquest.net/interessentest/ http://jobs.lidl.de/cps/rde/career_lidl_de/hs.xsl/ausbildung-beilidl-3152.htm

Eval. Yes Yes Yes Yes No Yes No Yes No Yes Yes No No Yes Yes Yes Yes Yes Yes Yes No Yes Yes (continued)

Defining Recrutainment: A Model and a Survey

45

Table 1. (continued) Organization Lidl Lufthansa OMV OTIB OTIB Personalamt Hamburg Playgen PWC Revelian Revelian RWE Shell SNCF Telekom Telstra Unilever US army Young targets, EXIT ZDH

URL http://lidl.cyquest.net/berufs-check/duales-studiumwirtschaftsinformatik/ https://www.be-lufthansa.com/jobs-und-ausbildung/ausbildung/ spiel-zur-berufsorientierung/ https://technikqueen.at/ http://www.professionals-thegame.nl/ http://www.arbomasters.nl/ http://cyou-startlearning.hamburg.de/zaf/index.php

Eval. Yes Yes No No No Yes

http://playgen.com/play/nanomission/ http://www.multipoly.hu/en/how-to-play/how-to-play.html http://www.revelian.com/theme-park-hero/ http://www.revelian.com/cognify/ http://www.rwe.com/web/cms/de/1062676/rwe/karriere/ einstiegslevel/schueler/ http://explorergame.shell.com/index-game.html https://defi-ingenieurs.sncf.com/end/end.html http://meine-entwicklung.telekom.com/#/ http://www.jobjam.com.au/ https://www.facebook.com/unilever/app/274419425907645/ https://www.americasarmy.com/ http://www.crackthecode.de/

No No No No Yes Yes No Yes Yes No No No

http://handwerk.de/berufe-checker/index.html

Yes

At least three experts evaluated each of the 25 applications we could access. The process builds on the established method of peer reviewing: first, each expert evaluated the application independently. Based on the model’s structure, the expert rated each of the eight criteria with a number between 1 (minimal) and 5 (maximal). In the second step, the experts came together in several meetings and discussed the results for each application – similar to the meeting of a program committee. If the standard deviation of the mean score of a criterion was above 1.0, the expert panel re-evaluated the application at the meeting to clarify open points or misunderstandings. The results are shown in Table 2, also providing the overall mean of all eight criteria. As pointed out before, a “low” score does not necessarily imply that the application is “not good”. It can reflect a conscious decision of the company to specify few job details, to refrain from implementing challenge or to provide no personalization. Nevertheless, the results offer a very good overview of the applications’ individual focal points.

46

O. Korn et al.

Table 2. Results for the 25 evaluated applications based on the expert panel’s agreed mean scores. Organization

Interactivity Narration Competition Real-world con.

Visual attr.

Challenge Personaliz. Job-related. Overall mean

Aldi

3

1

1

2

2

1

1

4

1.9

Allianz 1

2

1

1

2

2

1

2

4

1.9

Allianz 2

2

1

1

2

1

1

1

4

1.6

Ausbildung.de 2

1

1

1

2

1

1

2

1.4

BP

1

3

2

2

1

1

4

2.0

2

Commerzbank 3

4

1

2

3

3

1

5

2.8

Deloitte

3

1

2

3

3

1

4

2.5

3

ElVi

2

1

1

2

3

1

1

2

1.6

Fresenius

4

2

1

2

3

3

1

4

2.5

Gruner + Jahr

2

2

1

2

2

3

1

5

2.3

Haniel

2

1

1

2

3

2

1

4

2.0

Heineken

2

4

1

2

4

2

1

4

2.5

hkp

4

1

1

3

2

1

1

4

2.1

Knack

4

3

1

2

4

3

1

1

2.4

Krones AG

2

1

1

2

1

1

1

3

1.5

Lidl (dual study)

3

2

1

2

3

4

2

4

2.6

Lidl (interests) 3

1

3

3

2

1

1

4

2.3

Lidl (job check)

4

1

2

3

3

1

5

2.9

4

Lufthansa

2

1

1

3

2

2

1

4

2.0

Pers. Hamburg

3

4

3

3

3

3

1

5

3.1

RWE

2

1

1

3

2

1

1

2

1.6

Shell

4

3

2

2

3

3

1

2

2.5

Telekom

1

1

1

1

3

1

2

5

1.9

Telstra

4

3

3

2

3

2

1

4

2.8

ZDH

3

2

1

1

4

1

3

2

2.1

The applications’ individual profile can easily be visualized in net diagrams. We illustrate this exemplarily for two applications (Fig. 2): the one with the lowest and the one with the highest overall mean scores. The high ratings of the better solution (Fig. 2, right) are mainly due to the successful creation of an employee perspective (job-relatedness) and the good use of narration to guide the user through the tasks. In terms of recruiting, this is a very good application as the user can try different real-life job-tasks, and receives a detailed skill profile. However, it lacks in gaming characteristics and personalization. Indeed, if we compare the eight criteria within the 25 evaluated applications (Fig. 3), the means still show a large potential in most criteria. The only criterion, which currently seems to be a focal point of most applications, is job-relatedness. An ANOVA shows that the eight criteria are clearly distinct (high significance, p < .01). However, they mostly are not strongly developed. Especially personalization (mean score: 1.2) and competition (1.4) are rarely applied. Three other areas (narration, real-world connection, and challenge) score around 2.0. Interactivity (2.7) and visual attractiveness (2.6) are better integrated, with job-relatedness (3.6) being the most-developed criterion in the evaluated part of the corpus.

Defining Recrutainment: A Model and a Survey

ausbildung.de

Personalamt Hamburg

interacƟvity

interacƟvity job-relatedness

narraƟon

job-relatedness

compeƟƟon

personalizaƟon

narraƟon

personalizaƟon

real-world specificaƟon

challenge

47

compeƟƟon

real-world specificaƟon

challenge

visual aƩracƟveness

visual aƩracƟveness

Fig. 2. Comparison of the evaluation results of two recrutainment applications (“ausbildung.de” and “Personalamt Hamburg” = HR department of the city of Hamburg) using net diagrams based on the expert panel’s agreed mean scores.

5 4 3 2 1 2.7

2.0

1.4

2.1

2.6

1.9

1.2

3.6

0

Fig. 3. The mean results over the model’s eight criteria. The bars show the standard deviations.

5 Conclusion As pointed out in the introduction, this work aims to provide a pioneering overview of recrutainment. Based on important background aspects (gamification, assessment, gaming, and Generation Y) we first proposed a definition for recrutainment (Sect. 2). We then introduced a model with eight criteria (Sect. 3) for the evaluation of recrutainment applications. It draws from Chou’s Octalysis Framework [20] and the person-environment fit model by Kristof-Brown and Guay [21]. In our study (Sect. 4), we show that this model is well suited to describe and analyze recrutainment applications. A corpus of 42 recrutainment applications serves as the basis for the model’s application. Due to access restrictions, 25 applications could

48

O. Korn et al.

not be evaluated. We described the operative reviewing process, which is closely aligned to academic peer reviewing. Low scores should not simply be considered deficits but can reflect conscious decisions, for example to specify few job details, to refrain from implementing challenge or to provide no personalization. Nevertheless, the score profiles and their visualization in net diagrams provide a good overview of each applications’ focal points and potentially indicates areas for improvement. If we look beyond the scope of single applications at the corpus as a whole, it is obvious that recrutainment has not yet fully unfolded its potential. Only the mean scores of interactivity and visual attractiveness score above 2.5 with job-relatedness (above 3.5) being the most-developed criterion. Thus, the evaluated applications show an acceptable recruiting orientation while for many gaming criteria there is a huge potential. The “entertainment” part of recrutainment still needs reference implementations. In future work, we will investigate the acceptance of gamified applications in the target group and examine, which criteria are considered most important. We also will conduct expert interviews to substantiate our quantitative findings with qualitative data.

References 1. Dineen, B.R., Soltis, S.M.: Recruitment: a review of research and emerging directions. In: APA Handbook of Industrial and Organizational Psychology, Selecting and Developing Members for the Organization, vol. 2, pp. 43–66. American Psychological Association, Washington, DC (2011) 2. Thomas, S.L., Ray, K.: Recruiting and the web: high-tech hiring. Bus. Horiz. 43, 43–52 (2000) 3. Cross, R., Parker, A., Sasson, L.: Networks in the Knowledge Economy. Oxford University Press, Oxford (2003) 4. Sonnberger, D.: Der Einsatz von Online Games als Self-Assessment im Recruiting. Grin Publishing, Munich (2015) 5. Diercks, J., Kupka, K.: Recrutainment – Bedeutung, Einflussfaktoren und Begriffsbestimmung. In: Diercks, J., Kupka, K. (eds.) Recrutainment, pp. 1–18. Springer Fachmedien Wiesbaden, Wiesbaden (2013) 6. Deterding, S., Sicart, M., Nacke, L., O’Hara, K., Dixon, D.: Gamification: using game-design elements in non-gaming contexts. In: Proceedings of the 2011 Annual Conference Extended Abstracts on Human Factors in Computing Systems, pp. 2425–2428. ACM, New York (2011) 7. Deterding, S.: Gamification: designing for motivation. Interactions 19, 14–17 (2012) 8. Dörner, R., Göbel, S., Effelsberg, W., Wiemeyer, J. (eds.): Serious Games: Foundations, Concepts and Practice. Springer, Switzerland (2016) 9. Sinclair, J., Hingston, P., Masek, M.: Considerations for the design of exergames. In: Proceedings of the 5th International Conference on Computer Graphics and Interactive Techniques in Australia and Southeast Asia, pp. 289–295. ACM, New York (2007) 10. Bogost, I.: The rhetoric of exergaming. In: Digital Arts and Cultures (DAC) Conference, December 2005. IT University Copenhagen (2005) 11. von Ahn, L., Dabbish, L.: Designing games with a purpose. Commun. ACM 51, 58–67 (2008)

Defining Recrutainment: A Model and a Survey

49

12. Deterding, S., Dixon, D., Khaled, R., Nacke, L.: From game design elements to gamefulness: defining “gamification.” In: Proceedings of the 15th International Academic MindTrek Conference: Envisioning Future Media Environments, pp. 9–15. ACM, New York (2011) 13. Steffen, T.M., Hacker, T.A., Mollinger, L.: Age-and gender-related test performance in community-dwelling elderly people: six-minute walk test, berg balance scale, timed up and go test, and gait speeds. Phys. Ther. 82, 128–137 (2002) 14. Sternberg, R.J.: Handbook of Intelligence. Cambridge University Press, Cambridge (2000) 15. Kristof-Brown, A.L., Zimmerman, R.D., Johnson, E.C.: Consequences of individuals’ fit at work: a meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Pers. Psychol. 58, 281–342 (2005) 16. Buchanan, T.: Online assessment: desirable or dangerous? Prof. Psychol. Res. Pract. 33, 148–154 (2002) 17. Newzoo: Newzoo 2016 Global Games Market Report 18. Horovitz, B.: After Gen X, Millennials, what should the next generation be called? (2012). http://usatoday30.usatoday.com/money/advertising/story/2012-05-03/naming-the-nextgeneration/54737518/1 19. Koudal, P., Chaudhuri, A.: Managing the talent crisis in global manufacturing: strategies to attract and engage generation Y. A Deloitte Research Global Manufacturing Study, Deloitte Research (2007) 20. Chou, Y.: Actionable Gamification: Beyond Points, Badges, and Leaderboards. Octalysis Octalysis Media, Fremont (2015) 21. Kristof-Brown, A., Guay, R.P.: Person-environment fit. In: APA Handbook of Industrial and Organizational Psychology, Maintaining, Expanding, and Contracting the Organization, vol 3, pp. 3–50. American Psychological Association, Washington, DC (2011) 22. Vorderer, P., Hartmann, T., Klimmt, C.: Explaining the enjoyment of playing video games: the role of competition. In: Presented at the Proceedings of the Second International Conference on Entertainment Computing, 5 August 2003 23. Hassenzahl, M.: The effect of perceived hedonic quality on product appealingness. Int. J. Hum. Comput. Interact. 13, 481–499 (2001) 24. Birk, M.V., Atkins, C., Bowey, J.T., Mandryk, R.L.: Fostering intrinsic motivation through avatar identification in digital games. In: Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, pp. 2982–2995. ACM, New York (2016)

SciencOmat: A Gamified Research Platform for Evaluating Visual Attractiveness Christian Zagel1 ✉ , Alexander Piazza2, Yoan Petrov2, and Freimut Bodendorf2 (

)

1

2

ZukunftsDesign, Coburg University of Applied Sciences and Arts, Friedrich-Streib-Str. 2, 96450 Coburg, Germany [email protected] Chair of Information Systems, Friedrich-Alexander-University Erlangen-Nuremberg, Lange Gasse 20, 90403 Nuremberg, Germany {Alexander.Piazza,Yoan.Petrov,Freimut.Bodendorf}@fau.de

Abstract. There are many platforms on the market that support researchers and practitioners to create surveys and market studies. Nevertheless, nearly all of them focus on providing answers to textual questions. In contrast to existing systems this paper presents the concept, prototype, and evaluation of a new mobile plat‐ form for quantitative research strictly focusing on images: the SciencOmat. This platform uses pictures to evaluate products, marketing content, and other elements based on their visual attractiveness. Particular emphasis was placed on a high level of usability and user experience. The system integrates methods known from popular online dating applications (e.g., liking/disliking a product by swiping left or right) and also applies gamification elements to further drive user motivation. Next to the application and its evaluation using the User Experience Question‐ naire provided by Schrepp et al. we also present the results of two exemplary image data sets. Keywords: Gamification · Human-centered design · Systems engineering · Mobile application · Research platform · Visual attractiveness

1

Introduction

The extended usage of smartphones has created an opportunity for researchers to track users’ habits and needs [1]. Businesses today must address these changes in their marketing campaigns by making use of mobile applications to gain better consumer engagement [2]. This new way of gaining consumer engagement has also transformed market research. Market research has merged with other fields of business such as advertising, marketing, and gaming. Thus, innovative ways to get to know about customer insights such as mobile surveying have become essential. In this manner, mobile surveys are effective to collect small bits of user data that might be derived from in-situ responses, thoughts, and intentions that can give ideas to businesses to build longterm relationships with their customers and create brand loyalty among them. Mobile surveys are ideal for multiple-time users so that businesses can gain adequate informa‐ tion about their every single customer. In addition, it is possible to get more authentic, © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_5

SciencOmat: A Gamified Research Platform

51

intimate, and truthful information because mobile phones are personal devices of users [3]. Smartphone surveys are created and conducted via mobile survey applications. They are sent to smartphone users to collect information on various use cases. Usually, the application transfers the data using an internet connection. Unlike browser-based surveys, smartphone survey applications can be utilized even without a strong internet connec‐ tion, eliminating refresh and delay problems while completing the survey. Some mobile survey applications are developed to conduct surveys that include multimedia content. Also, various data like images, videos, audio and GPS can be captured to enrich the user’s survey experience. A drawback of mobile survey applications is that they might need to be developed for multiple platforms and additional programming can be required to create different kinds of surveys [4]. Mobile surveys can assist lecturers and students in higher education by providing unbiased data with low measurement error. Survey results usually tend to be influences by the so called social desirability bias. This term describes the tendency of respondents answering questions in a more favorable manner. Less social desirability bias is found with mobile surveys since respondents have the freedom to choose where to take them [5]. Mobile phones significantly shorten the length of the data collection time period which enables in-situ responses from users [6]. Also, researchers state that, independent of the end device used, a survey engine with a user-friendly inter‐ face can be very efficacious [7]. Unfortunately, users are not always motivated to participate in online surveys. Also, the type of survey platform used has a high influence on user motivation and the effec‐ tiveness of evaluations conducted [8]. This effect applies even more to business-focused studies, which are usually perceived as less attractive, not worth investing the time required [9]. The aim of this paper is to present a new type of smartphone-based research platform that focuses on continuously motivating users by increasing ease of use and integrating gamification elements as well as by leveraging the potential of micro survey tasks. The following sections discuss the theoretical background, followed by a presentation of the prototype of our gamified research platform. Also, we present the results of two studies and a user experience questionnaire to support the applications’ functionality and acceptance.

2

Platform Concept and Theoretical Background

The last decade coined a new term used in the area of motivation and efficiency research: Gamification [10]. The method refers to the use of game elements and game principles in non-game contexts and tasks to increase motivation and create improved user experiences [11]. The concept promises increased productivity while simulta‐ neously raising user satisfaction [12]. While the method already finds application in various business use cases, researchers demand for more practical examples that prove its effectiveness [13]. Previous studies show that using gamification might have a positive effect on user engagement and motivation in everyday processes [14]. A surveys’ suitability for gamification can be explained through the classification

52

C. Zagel et al.

scheme described by Zagel and Bodendorf [15]. Reasons are the high amount of process iterations and process executers as well as its perception as a typically boring and monotonous task. The idea is that participants can post their own surveys once they reach a certain level within the application. The idea of focusing on small, isolated tasks that can be solved individually and at any time can also be found in popular online dating applications. While only two decades ago finding an appropriate partner was subject to real life, the modern age has lead people to leverage the potentials of the Internet and the broad social network [16]. Online dating platforms are used by an increasing number of participants. They attract a wide range of user personalities with 30 to 50-year-old persons as the most active user group [17]. Nevertheless, the objectives of the users differ [18] and so does the focus of the dating platforms. One system widely known for strictly focusing on a counterpart’s visual attractiveness is Tinder. It resembles the system used in a “hot or not” manner as it asks users to decide if they “like” proposed partners by swiping the images to the right, or dislike them by swiping to the left. Whenever two users mutually “like” each other they get a “match” and are linked up to the system. The fact that Tinder merely focuses on profile pictures for matching users makes clear that the only aspect taken into account is the visual attractiveness of the potential partner. These modern dating applications are part or an emerging method that uses the potential of online communities to gain access to a high number of people for having issues solved quickly: Crowdsourcing. Crowdsourcing is defined as the distribution of work tasks to a large network of individuals on the Internet. Hence, it can be compared to outsourcing work to many, in most cases to unknown, people in the form of an open call [19, 20]. While the method finds wide application by companies in the consumer goods industry and the example of online dating, use cases within science and research are relatively new [21]. The main idea of the application presented in this paper is to establish a new research platform that focuses on visual elements (images), leveraging the potential of using gamification elements to further drive motivation and user engagement in the long run. While first attempts of applying gamification to science and for gamifying surveys are known [22], the focus on the combination of micro tasks (every image rating provided), crowdsourcing and usability are new.

3

Prototype

The SciencOmat research platform was realized in form of an Android-based smart‐ phone application released on the Google Play Store in January 2017. It contains the complete set of functions to register, to take part in surveys, as well as to collect points as part of the gamification concept. The backend component allows the researcher to setup new surveys. These consist of an introduction text, a start image displaying a brief description of the task to be completed, as well as a set of images organized in a database. This backend also allows the researcher to export the survey results.

SciencOmat: A Gamified Research Platform

53

3.1 SciencOmat: User Interface As stated above the frontend is implemented in Android. As soon as the application is installed and started, the initial screen allows the user to login using the Facebook API (Fig. 1 shows the login screen). Using a login via Facebook allows the system to pull relevant user information from the social platform and hence to automatically create a profile within the application. The current set of data includes user name, gender, age, email address, and current location (country and city). This information is furthermore used when survey data is exported (e.g., a comma separated value file), allowing to perform analyses in regards to demographics.

Fig. 1. Facebook login to capture important user information

After logging in to the application, the users are directed to their personal dashboard that shows the current status of the application (Fig. 2, left). This dashboard includes the user’s current level based on the number of surveys already completed as well as the distance to reach the next level. In order to further drive motivation and for being able to compete with others the application allows sharing the level on the personal Facebook profile.

54

C. Zagel et al.

Fig. 2. Gamification system and survey status overview

The second tab (Fig. 2, right) shows all active surveys including the user’s personal status (completed/not started). Each survey includes the start image as well as a short description of the task. Hence, each user has the possibility to select the surveys he/she is most interested in. After the survey is started with a simple click, a short description is displayed (Fig. 3, left). It includes basic information about the task as well as instructions for the user. The text can be instantly configured and modified for each survey. To cast a vote the user just needs to swipe left (confirm) or right (disconfirm) for each image displayed. Alternatively, it is possible to just click on either of the signs shown below the image (Fig. 3, right). It is furthermore possible to realize alternative scenarios: in case a picture shows two elements (one picture on the left, one on the right), swiping left or right can also indicate a certain preference. The intention of the respective tasks of course needs to be mentioned in the survey description. An additional user menu allows the user to log out of the application as well as to display legal information. Independent of the respective use case, the survey result always consists of a list of 0’s (no) and 1’s (yes) as ratings for every picture. These results are then aggregated with the respective user information like name, age, gender, location. The resulting export sheet covers all users and their ratings in a separate row.

SciencOmat: A Gamified Research Platform

55

Fig. 3. Example of a survey including introductory description of the task

3.2 User Experience Evaluation The User Experience Questionnaire by Laugwitz et al. [23] is used to evaluate the user experience perceived when working with the mobile application. As an established questionnaire it can easily measure perceived attractiveness, design, and quality of soft‐ ware systems by 26 scale items in six dimensions: Attractiveness, Perspicuity, Effi‐ ciency, Dependability, Stimulation, and Novelty [24]. Figure 4 and Table 1 present the results of the survey conducted amongst 27 participants.

Fig. 4. UEQ results

56

C. Zagel et al. Table 1. UEQ scale ratings. UEQ scale Attractiveness Perspicuity Efficiency Dependability Stimulation Novelty

Rating 1.012 2.741 2.019 0.731 0.361 0.315

According to Laugwitz et al. [23], the results are interpreted as follows: while values between −0.8 and 0.8 represent a neutral evaluation of the corresponding scale, values above 0.8 represent a positive, and values below −0.8 a negative assessment. This means that for the given application the scales Attractiveness, Perspicuity, and Efficiency received a very positive rating while the other dimensions were evaluated neutral. It is also possible to group the UEQ scales into the dimensions “Pragmatic Quality” (combined rating: 1.83), “Hedonic Quality” (combined rating: 1.01) and “Attractive‐ ness” (rating: 1.01). Hence, task related quality aspects were rated very positive along with the attractiveness of the system. Nevertheless, the system still lacks in hedonic quality.

4

Use Case Evaluation

Many use cases can be realized using the given application. The possibility to change the dimensions (e.g., yes/no, left/right preference) allows a constant modification and development of new tasks. Nevertheless, results need to be interpreted accordingly. In order to evaluate the application as well as its suitability to conduct surveys for measuring visual attractiveness, two exemplary use cases were developed. Both use cases are linked to the fashion industry and use a set of shoe images and can be performed as micro tasks. In total, 51 users (22 female, 29 male) aged from 18 to 63 (avg. 28.4) took part in each of the surveys. The geographic distribution of the subjects was as follows: 27 from Bulgaria, 14 from Germany, three from Austria, one from the United Kingdom, one from the Netherlands, and five unknown. 4.1 Use Case 1: Sales Prediction For companies, it is hard to predict sales numbers of their products. This way it is not possible to accordingly adjust production and especially in the consumer goods area, many products cannot be sold at a full price. They have to be sold off, while popular product designs are too quickly sold out. Particularly in the fashion industry, product design concepts are available long before the first physical product hits the market. Hence, these design sketches might be used to identify the popularity of a product and its design. Our first image set consists of ten pictures of one shoe model for men in and eight pictures of a shoe for women, each in a different color way. We focused on one particular

SciencOmat: A Gamified Research Platform

57

shoe model to avoid an additional bias due to different product features and price points and to focus only on one criterion: the color. The goal for the user is to “like” the preferred colorways. As a result, the system creates a summarized ranking including the rating of all users that took part in the survey. Afterwards, the ranking is compared with the real sales numbers of the different color ways. Our hypothesis is that we would be able to predict product sales if both the ranking of the image ratings as well as the ranking in sales data follow the same sequence (Table 2). Table 2. Use case 1: sales prediction - results (N = 51) Shoe Male 10 Male 9 Male 1 Male 7 Male 6 Male 2 Male 5 Male 8 Male 3 Male 4

Rank likes SciencOmat 1 2 3 4 5 6 7 8 9 10

Rank sales

Shoe

6 10 3 4 7 2 8 9 1 5

Female 2 Female 8 Female 1 Female 3 Female 5 Female 4 Female 6 Female 7

Rank likes SciencOmat 1 2 3 4 5 6 7 8

Rank sales 4 6 1 2 5 3 8 7

From the results of the survey no clear pattern between the preference indications and the sales performance can be identified. A reason might be the comparably small group of subjects as well as the fact that most of them live in Eastern Europe, while the sales rank only represents numbers from Germany. This might indicate different color preferences for different geographic locations. Hence, it was not possible to confirm the hypothesis. More detailed studies need to be conducted in order to better validate the feasibility of the given use case. 4.2 Use Case 2: Virtual Prototyping In the recent years, virtual product images are increasingly used for marketing purposes. Ikea for example increased the number of computer-generated images on their Web site and in printed catalogs by 50% already between 2012 and 2014 [25] while in the car industry more than 80% of the images are virtual [26]. The big advantage of using computer-generated images is cost savings: no physical prototypes have to be built in order to take pictures for marketing content which additionally increases the time to market. Nevertheless, and in comparison to other industries, using virtual prototypes in the fashion industry is comparably new [27]. Previous research [28] already was able to show that virtual product images reached a quality level where consumers are not able to distinguish virtual from real images anymore. Hence, our second image set consists of a total of 20 product images. Ten of them show a shoe model in form of computer-generated images, while the other ten represent

58

C. Zagel et al.

photographs taken from real physical products. The task for the user is to either mark it as a real picture (left swipe) or as a fake (right swipe). Our hypothesis is that no significant difference can be identified if the quality of the virtual pictures is realistic enough. The results of the study are shown in Table 3. Table 3. Use case 2: virtual prototyping - results (N = 51) Shoe Real_1 Real_2 Real_3 Real_4 Real_5 Real_6 Real_7 Real_8 Real_9 Real_10 Average

Votes “real” 21 22 24 22 30 31 26 22 29 31

Votes “fake” 30 29 27 29 21 20 25 29 22 20

Detection rate 41.18% 43.14% 47.06% 43.14% 58.85% 60.78% 50.98% 43.14% 56.86% 60.78% 50.59%

Shoe Fake_1 Fake_2 Fake_3 Fake_4 Fake_5 Fake_6 Fake_7 Fake_8 Fake_9 Fake_10 Average

Votes “real” 18 18 25 24 23 21 24 23 22 24

Votes “fake” 33 33 26 27 28 30 27 28 29 28

Detection rate 64.71% 64.71% 50.98% 52.94% 54.90% 58.82% 52.94% 54.90% 56.86% 52.94% 56.47%

The study results indicate that it is very hard for the users to distinguish real from virtual product images, even if they are aware of being confronted with both. For both cases, fake and real images, the detection rate is around 50%. This means that the hypothesis can be confirmed and that the quality of virtual product images is good enough to use them e.g., for marketing purposes.

5

Conclusions and Future Research

In this paper, we presented a novel approach to consumer research with a specific focus on images/designs as objects of study. The system is based on a smartphone application and leverages the potential of targeting the young and mobile user group with a high level of usability and ease of use. Specific functionalities known from popular online dating applications are integrated with a gamification approach to driver user motivation and engagement. The results prove the applications’ pragmatic quality and attractive‐ ness, as well as the applicability for various use cases. Nevertheless, the concept still lacks hedonic quality and a more detailed elaboration of the gamification aspects. In the future, the current gamification concept could be extended by integrating a micro-earning concept. By doing so, users could earn a small value for each survey they conduct or even for every image that is rated through the application. The integration of additional game elements will further drive user motivation and allow an engagement in competitions. This might also provide potential for future research in the area of human computation.

SciencOmat: A Gamified Research Platform

59

5.1 Annotation Researchers are invited to publish their own surveys through our app for free. If you want to do so, please get in contact with one of the authors. In return it would be great if you refer to our work within your own research.

References 1. Cook, W.: Is mobile a reliable platform for survey taking? J. Advertising Res. 54, 143–148 (2014) 2. Tarute, A., Nikou, S., Gatautis, R.: Mobile application driven consumer engagement. Telematics Inform. 34(4), 145–156 (2017) 3. Robbins, C.: Mobilizing Market Research: The state-of-the-art, future evaluations and implications of mobile data collection methods in the field of market research. Greenbook (2011) 4. Buskirk, T., Andrus, C.: Smart surveys for smart phones: exploring various approaches for conducting online mobile surveys. Surv. Pract. 5, 5–7 (2012) 5. Lynn, P., Kaminska, O.: The impact of mobile phones on survey measurement error. Public Opin. Q. 77, 586–605 (2012) 6. Wells, T.: What market researchers should know about mobile surveys. Int. J. Mark. Res. 57(4), 521–532 (2015) 7. Trucano, M.: Using mobile phones in data collection: opportunities, issues and challenges (2017). http://blogs.worldbank.org/edutech/using-mobile-phones-data-collection-opportunitiesissues-and-challenges 8. Nair, C., Adams, P.: Survey platform: a factor influencing online survey delivery and response rate. Qual. High. Educ. 15(3), 291–296 (2009) 9. Bosnjak, M., Batinic, B.: Understanding the willingness to participate in online surveys: the case of e-mail questionnaires. Online Soc. Sci. 81, 81–92 (2002) 10. Zicherman, G.: The first gamification summit rocked (2012). http://gamification.co/ 2011/01/24/the-first-gamification-summit-rocked/ 11. Deterding, S., Dixon, D., Khaled, R., Nacke, L.: From game design elements to gamefulness: defining gamification. In: Proceedings of the 15th International Academic MindTrek Conference: Envisioning Future Media Environments, pp. 9–15. ACM (2011) 12. Stieglitz, S.: Gamification – Vorgehen und Anwendung. HMD Praxis der Wirtschaftsinformatik 52(6), 816–825 (2015) 13. Gonzales-Schaller, P.: Trendthema Gamification: Was steckt hinter diesem Begriff? In: Diercks, J., Kupka, K. (eds.) Recrutainment. Spielerische Ansätze im Personalmarketing und –auswahl, pp. 33–51. Springer, Wiesbaden (2013) 14. Neeli, B.: A method to engage employees using gamification in BPO industry. In: 2012 Third International Conference on Services in Emerging Markets on Services in Emerging Markets (ICSEM), pp. 142–146. IEEE (2012) 15. Zagel, C., Bodendorf, F.: Gamification: Auswirkungen auf Usability, Datenqualität und Motivation. In: Koch, M., Butz, A., Schlichter, J. (eds.) Mensch und Computer 2014, Tagungsband, pp. 15–24, Oldenbourg Wissenschaftsverlag, München (2014) 16. Finkel, E.J., Eastwick, P.W., Karney, B.R., Reis, H.T., Sprecher, S.: Online dating: a critical analysis from the perspective of psychological science. Psychol. Sci. Public Interest 13(1), 3–66 (2012)

60

C. Zagel et al.

17. Valkenburg, P.M., Peter, J.: Who visits online dating sites? Exploring some characteristics of online daters. CyberPsychol. Behav. 10(6), 849–852 (2007) 18. Couch, D., Liamputtong, P.: Online dating and mating: the use of the internet to meet sexual partners. Qual. Health Res. 18(2), 268–279 (2008) 19. Brabham, D.C.: Crwodsourcing as a model for problem solving. J. Res. New Media Technol. 14(1), 75–90 (2008) 20. Howe, J.: The rise of crowdsourcing. Wired Mag. 14(6), 1–4 (2006) 21. Behrend, T.S., Sharek, D.J., Meade, A.W., Wiebe, E.N.: The viability of crowdsourcing for survey research. Behav. Res. Meth. 43(3), 800–813 (2011) 22. Harms, J., Biegler, S., Wimmer, C., Kappel, K., Grechenig, T.: Gamification of Online Surveys: Design Process, Case Study, and Evaluation. Human-Computer Interaction, pp. 219– 236. Springer International Publishing (2015) 23. Laugwitz, B., Held, T., Schrepp, M.: Construction and evaluation of a user experience questionnaire. In: HCI and Usability for Education and Work, pp. 63–76. Springer, Heidelberg (2008) 24. Niels, A., Lesser, T., Krüger, T.: The impact of causal attributions on the user experience of error messages. In: Advances in the Human Side of Service Engineering, pp. 173–184. Springer International Publishing (2017) 25. Parkin, K.: Building 3D with Ikea (2014). http://www.cgsociety.org/index.php/CGSFeatures/ CGSFeatureSpecial/building_3d_with_ikea 26. Southern, A.: Real or rendered? How 3D imagery is changing the way you shop (2012). http:// techonomy.com/2012/10/real-or-rendered-how-3d-imagery-is-changing-the-way-you-shop/ 27. Zagel, C., Löffler, C.: Technisches Konzept zur Service Externalisierung virtueller Prototypen in der Modebranche. In: Schumann, M., Kolbe, L.M., Breitner, M.H., Frerichs, A. (eds.) Multikonferenz Wirtschaftsinformatik 2010, pp. 1265–1276. Universitätsverlag Göttingen (2010) 28. Zagel, C., Piazza, A.: Zur Wahrnehmung virtueller Produktbilder im Online Shopping. Mensch und Computer 2016 – Tagungsband (2016)

Innovations in Quality and Safety Practices in Health Care Systems

Alarm Management: A Systems Approach to Patient Safety Vinaya Sebastian ✉ and Carolyn Hamasaki Brown (

)

Santa Clara Valley Medical Center, San Jose, CA, USA [email protected]

Abstract. The purpose of a medical device alarm is to command the attention of the clinician for notification of an immediate or potential adverse patient condition. It should be actionable and requires a response. The frequency of false alarms has increased due to increased numbers of alarming devices, lack of standardization of alarms, and alarm settings with sensitivity over specificity. In some hospitals alarm signals can reach several hundred per patient, with most requiring no clinical intervention. Clinicians can become overwhelmed and desensitized by the frequency of these alarms. During the last six years, there have been patient deaths reported that were attributed to “alarm fatigue.” Clinical alarm hazards have been cited as being among the top ten health technology hazards for several years. As a result, a national initiative is underway in the United States focusing on safe use of alarms. Appropriate use of well-designed devices and an organized work environment can reduce risks and increase safety. Alarm custom‐ ization, tailoring alarm parameters to individual patient needs, communication between clinicians, policies to empower clinicians and teamwork can help to create safer systems. Keywords: Alarm management · Alarm fatigue · Human factors · National Patient Safety Goal · Device safety · Systems analysis

1

Introduction

Technology today affects every aspect of modern society. Technological breakthroughs have changed the structure and organization of healthcare delivery. New medical devices are introduced to the healthcare industry on a regular basis. A medical device is intended for diagnosis, treatment, prevention and mitigation of diseases in human beings. It can be an instrument, apparatus, machine or any kind of related article which has a compo‐ nent or part. Medical devices range from simple tongue depressors or medical gloves to complex implantable devices. Patients in an Intensive Care Unit (ICU), Surgery or Emergency Room are connected to several devices. These devices have alarms that will alert clini‐ cians when the patient’s health condition deteriorates. Device alarms are designed to be highly sensitive. Alarms become meaningless if critical notifications are missed in the cacophony of multiple alarms and alarm artifact. The number of device alarms have increased in such a way that providers are desensitized by frequent false positive alarms © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_6

64

V. Sebastian and C.H. Brown

or clinically irrelevant alarms. This desensitization or “Alarm Fatigue” can be caused by several factors in the system [8, 9]. These factors include environment, work methods, equipment, manpower, work culture and human factors [10]. To promote safe use of alarms, one should consider all these factors and customize it to the unique patient needs to create a safe system.

2

Background

The Manufacturer and User Facility Device Experience Database (MAUDE) reports 383 deaths from January 2010 to December 2012 related to monitoring device alarm problems. Of this total, 215 of the deaths were patient monitor related and 38 were respiratory ventilator related. National attention to alarm hazards grew following reports from Federal investigators that pointed to alarm fatigue, in particular as it contributed to the death of a patient in a highly prestigious hospital in 2010. The Emergency Care Research Institute has reported clinical alarm conditions as one of the top two technology hazards since 2007. The Joint Commission, a U.S. based independent, non-profit health care accreditation organization, reported 98 alarm related sentinel events between January 2009 and June 2012 [6]. Reporting of sentinel events to this agency is voluntary therefore the volume likely under represents actual incidence. The Joint Commission has identified clinical alarm management as one of its National Patient Safety Goals (NPSG) [1, 9]. We assessed the current status of the most frequently utilized clinical alarms in all monitored units to identify contributing system factors causing alarm fatigue and to develop strategies to enhance alarm safely.

3

Device Safety Analysis and FDA Approval Process

To assess device safety, it is important to understand the device approval process. To ensure the safety and effectiveness of medical devices, The Center for Devices and Radiological Health (CDRH) develops and implements national programs and regula‐ tions [4]. CDRH is part of the Food and Drug Administration (FDA) [4]. Medical devices undergo a rigorous approval process before use is permitted. The FDA uses a risk-based classification for product approval. All of these products are subject to regulatory controls which are defined by their classification. The evidentiary threshold increases with the severity of risk for demonstrating product safety and efficacy. The performance of a medical device is often compromised by noise, stress and fatigue. FDA recommends that manufacturers incorporate principles of human factors or usability engineering processes during the development of new medical devices [4]. Manufacturers should specifically focus on user interface with a goal of eliminating harm caused by use errors. As part of design control, manufacturers conduct a risk analysis that includes risk associated with device use and measures to reduce the risk. Risk analyses and risk mitigation measures ensure that the device is safe to use in the real world with simulated stress situations [3]. Physiological monitors are considered low risk devices. For FDA approval, a low risk device needs only documentation proving

Alarm Management: A Systems Approach to Patient Safety

65

that a similar device is already in the market that is safe and effective. The devices that are in the market are subject to post-market surveillance as defined by Medical Reporting Regulations. All devices in clinical settings are subject to these regulatory rules.

4

Method

This quantitative and qualitative study was conducted in an accredited 574-bed hospital in an academic medical center. To get an insight into the types of alarms and possible measures to reduce the number of non-actionable alarms, we collected alarm data for 30 days from physiological monitors for all monitored units. A qualitative risk analysis was done for all equipment with alarms to prioritize the most important alarms to manage. We surveyed physicians and nurses to get clinicians’ opinion on various alarms and contributing factors on alarm fatigue. 4.1 Quantitative Data Analysis Alarm data was collected for a 30-day period from 10 different monitored units. There was total of 240,505 alarms during this time period (see Table 1), with an average of 7,755 alarms daily. Peripheral oxygen saturation (SpO2) was found to be among the top three most frequent alarms for each unit in 6 out of 10 units observed. The Telemetry unit alone had a total of 80,740 alarms overall with a daily average of 2,604. Of this total, only 1,000 were high priority alarms (see Fig. 1). The 24 h alarm report for the Neonatal Intensive Care Unit (NICU) revealed a daily average of 3,420 alarms. Of this total, 1,603 alarms were SpO2, which represented the highest alarm frequency in NICU (see Fig. 2). Table 1. 30 day alarm data analysis

a

Unit BURN CCU EMR MICU PEDI PICU

Total 5,612 23,129 49,666 23,087 8,498 4,984

RTC2 SICU TELE TICU Total

5,614 21,938 80,740 17,237 240,505

1 bed had 0 alarms

Beds 8 8 25 8 30 7a 8 8 84 8 187

Per bed 702 2891 1987 2886 283 712

Avg/Day 181 746 1602 745 274 161

Bed/Avg/Day 23 93 64 93 9 23

Most frequent alarm SpO2-2,151 SpO2-4,114 RR - 13,381 SpO2-3,904 RR/SpO2-2,559/2,548 RR-1,650

702 2742 961 2155 16139

181 708 2605 556 7758

23 88 31 70 517

SpO2-3,033 SpO2-3,549 HR-15,847 ABPm-3,214

66

V. Sebastian and C.H. Brown

High Priority Alarms 500 400 300 200 100 0 Asystole

VFib

Vfib/Tach

Extreme Tachy Exterme Brady

Fig. 1. 30-day high priority/critical alarms for telemetry

Non-Sustain VT Missed Beat HR Low HR End AFIB AFIB Vent Fib/Tach RR Low QTc High NBPs High Extreme Brady Desat Apnea ABP High 0

2000

4000

6000

8000 10000 12000 14000 16000 18000

Fig. 2. Detailed alarm data for telemetry

Alarm Management: A Systems Approach to Patient Safety

67

Fig. 3. 24 h alarm report for NICU

4.2 Risk Assessment Risk analysis was conducted on equipment with alarms that considered both severity and staff response to an alarm. The clinicians from all the monitored units rated the severity of injury to patients if they did not respond promptly to the equipment alarm, on a scale of 1–5. One being low likelihood of injury and five resulting in death. The clinicians also rated the probability of delayed response on a scale of 1–5 with one being the least likelihood of delayed response [11]. A risk priority index was calculated by multiplying severity rating and probability rate. For example: Any equipment with a score of 9 or above was included in the critical alarm category. The clinicians agreed that 14 out of 26 devices included in the analysis triggered critical alarms. Bedside physiological monitor, heart rate monitor, and pulse oximeter were examples such devices. 4.3 Qualitative Data Analysis A paper survey was conducted to assess clinicians’ opinions on various alarm conditions. Utilizing a 5-point Likert-scale respondents rated the significance of six alarm related issues. Two of these that were significant in the analysis were the causes of alarm fatigue and the reason for reduced attention to alarm. Of the total of 202 responses, 54% of the clinicians felt that specific alarm signals contribute to alarm fatigue and 34% agreed that frequent false alarms cause reduced attention or response to alarms when they occur (see Fig. 4).

68

V. Sebastian and C.H. Brown

Clinician Opinions of Alarms SPECIFIC ALARM SIGNALS CONTRIBUTE ALARM FATIGUE

54%

NOISE COMPETION FROM NON-CLINICAL ALARMS AND PAGES

35%

FREQUENT FALSE ALARMS LEAD TO REDUCED ATTENTION

34%

INADEQUATE STAFF TO RESPOND TO ALARMS OVER RELIANCE ON ALARMS TO CALL ATTENTION TO PATIENT PROBLEMS DIFFICULTY IN IDENTIFYING THE SOURCE OF ALARM

29% 25% 20%

Fig. 4. Clinician opinions of alarms

5

Discussion

The number of devices approved for patient or equipment monitoring and equipped with alarms continues to increase. From 1983 to 2012, there was a six-fold increase in the number of devices with alarms [12]. The number of monitoring devices attached to a single patient increases with the complexity of the patient’s condition. The patients admitted in an ICU need more intense monitoring with several devices compared to the patients in a telemetry unit. Depending on the criticality of the patient’s condition, the number of devices, noise and stress level increases. Device alarms are intended to alert the clinician or caregiver to a potential problem. Yet, when that individual encounters too many alarms, it may disrupt his or her workflow resulting in error due to alarm desensitization, distraction or omission of alarm [7]. There are few simple, easy to implement solutions to address alarm safety. For a sustainable solution and to improve patient safety, a systems approach is required. Clearly communicated written policies related to alarm safety and training on device use and management are necessary to obtain optimal outcomes [2]. Knowledge of the processes involved with monitoring, noise reduction, reviewing equipment technologies and safety features are equally important to avoid potential errors by users. Standardizing care, improving communication, and avoiding complexities can reduce errors related to human factors. A device failure or a sentinel event related to medical device use can occur for several reasons. Two primary causes reported by the FDA are: device manufacturing and design issues and device-human interaction problems. Our alarm data analysis and clinician survey demonstrated that an excess of alarms results in reduced attention (see Fig. 4). Reduced attention can result in critical alarms being missed. Application of signal

Alarm Management: A Systems Approach to Patient Safety

69

detection theory in this situation assumes that all reasoning and decision-making take place in some uncertainty. There are four possible responses to signal [5]: • Hit: The clinician identifies a critical alarm. • False alarm: A clinician identifies an alarm with no danger present. • Miss: A clinician fails to respond to a critical alarm, which may lead to a sentinel event. • Correct rejection: Nothing is wrong with the situation. Here the clinicians perform signal detection. The factors that can affect the threshold applied are their experience and ability to recognize the signal, including their psycho‐ logical and physiological state. When there is an excess of alarms, particularly those with no significance, the chance of making the right decision in recognizing and responding to a critical alarm can decrease.

6

A Systems Approach to Alarm Safety

After analyzing the data and researching best practices, we selected pilot units to imple‐ ment age specific alarm configuration. We developed an evidence-based approach to reduce false alarms. We conducted a literature review and incorporated The Joint Commission’s requirements for managing alarms. The Pediatric unit and Neonatal Intensive Care units were selected for pilot. We adapted a tested set of parameters from another university medical institution for alarm reconfiguration. The device alarms were reconfigured for the pediatric unit. Our Neonatal Intensive Care Unit is part of Vermont Oxford Network Research and their device configuration was set according to the Network’s recommendation. Decreasing SpO2 thresholds and increasing delay times were part of the pilot design strategy. Duplicate monitoring displays were eliminated to reduce noise. Nurses were educated on lead hygiene, which has been shown to reduce alarms by 50% in other facilities, proper placement of leads, and the new configuration. There was a substantial increase in alarms, one month after the reconfiguration (Fig. 5). When clinicians were surveyed on their practice, perceptions and concerns, it was evident that not all were fully compliant with the new protocols and were using the default settings instead of the age specific configuration. Additional education was indi‐ cated. Post-configuration data was collected and analyzed after four months and there was 34% reduction in alarms compared to the pre-configuration data (Fig. 6). A 30-day data analysis was conducted after nine months to ensure the effectiveness of the imple‐ mentation (Fig. 7).

70

V. Sebastian and C.H. Brown

Alarm Comparison Report

15234

8498 4984 PEDI

4645 PICU

Pre-configuraƟon Data

Post-configuraƟon Data

Fig. 5. 30 day alarm comparison report

Post-configuraƟon Alarm Data Comparison for 24hrs 275 227

161 105

PEDI

PICU Pre-configuraƟon

Post-configuraƟon

Fig. 6. Post-configuration alarm data comparison for 24 h.

Alarm Management: A Systems Approach to Patient Safety

71

Alarm Data Comparison for 30 days 15234

8498 7102 4984

PRE-CONFIGURATION

4645

POST-CONFIGURATION - JAN 2016 PEDIATRICS

3474

POST-CONFIGURATION SEPT 2016

PICU

Fig. 7. Post-configuration alarm data comparison for 30 days

To meet the unique needs of the patients, the senior leadership and alarm manage‐ ment committee developed policies that recognize individual patient needs and maintain patient safety. Clinician roles were defined. The physicians specify individual alarm parameters for a patient depending on the disease condition. The nurses set those param‐ eters on the monitor. Any change of condition is discussed with the physician and parameter adjustments are made accordingly. The clinicians were educated on the policy, lead hygiene and advised to check the patient monitor settings at the beginning of the shift, change of patient condition and change of personnel. The physicians are advised to determine the need for monitoring, and, to avoid unnecessary monitoring. Complete hand-off communication is required of all clinicians. We conducted a detailed analysis on all other equipment identified as critical. Poli‐ cies on all identified equipment with high risk were reviewed and incorporated into the systems approach.

7

Conclusion

Advancements in medical technology have helped shape health care delivery and saved lives. Increasingly, technological advances are being integrated into patient care and device alarms into the fabric of care environment. Missed alarms have caused death, injuries and jeopardized patient safety. Proper system design that demonstrates the ability to prevent and mitigate risk associated with alarm management processes improves patient safety and outcomes.

72

V. Sebastian and C.H. Brown

References 1. AAMI, FDA, TJC, ACCE, and ECRI Institute Clinical Alarms Summit (2011). http:// www.aami.org/publications/summits/2011_Alarms_Summit_publication.pdf 2. American Association of Critical-Care Nurses: Alarm management. http:// ccn.aacnjournals.org/content/33/5/83.full. Accessed Feb 2017 3. Sawyer, D.: Do it by design: an introduction to human factors in medical devices. U.S. Department of Health and Human Services. https://elsmar.com/pdf_files/FDA_files/ DOITPDF.PDF 4. FDA Guidance for Industry and Food and Drug Administration: Applying human factors and usability engineering to medical devices. https://www.fda.gov/downloads/Medical Devices/…/UCM259760.pdf. Accessed Feb 2017 5. McClelland, G.H.: Use of signal detection theory as a tool for enhancing performance and evaluating tradecraft in intelligence analysis. https://www.nap.edu/read/13062/chapter/7#98. Accessed Feb 2017 6. Joint Commission Sentinel Event Alert. http://www.jointcommission.org/assets/1/18/ SEA_50_alarms_4_5_13_FINAL1.PDF. Accessed Feb 2017 7. Cvach, M., Creighton, K.: Monitor alarm fatigue: standardizing use of physiological monitors and decreasing nuisance alarms. http://ajcc.aacnjournals.org/content/19/1/28.abstract. Accessed Feb 2017 8. Sendelbach, S., Funk, M.: Alarm fatigue: a patient safety concern. AACN Adv. Crit. Care 24(4), 378–386 (2013). http://www.aacn.org/wd/cetests/media/acc3342.pdf 9. Joint Commission on Accreditation of Healthcare Organizations: The Joint Commission announces 2014 National Patient Safety Goal. Joint Comm. Perspect. 33(7), 1 (2013) 10. Guardia-LaBar, L.M., Scruth, E.: Alarm fatigue. Advance healthcare network. http:// nursing.advanceweb.com/Continuing-Education/CE-Articles/Alarm-Fatigue.aspx. Accessed Feb 2017 11. Medical Equipment: Clinical Alarms Risk Assessment. http://www.courtemancheassocs.com/wp-content/uploads/Clinical-Alarm-Risk-Assessment-Template. Accessed Feb 2017 12. Oster, C., Braaten, J.: High Reliability Organizations: A Healthcare Handbook for Patient Safety & Quality. Sigma Theta Tau International, Indianapolis (2016)

How Is the Quality of Life of Patients with Cerebral Palsy Improved? Qualitative and Quantitative Evaluation of a Communication and Learning Assistance System Based on ICTs Jonnathan Berrezueta-Guzman ✉ , Vladimir Robles-Bykbaev, and Luis Serpa-Andrade (

)

GIIATa, Universidad Politécnica Salesiana, Calle Vieja 12-30 y Elia Liut, 010105 Cuenca, Ecuador [email protected], {vrobles,lserpa}@ups.edu.ec

Abstract. The Organization of American States (OAS) reports that Ecuador has the third highest rate of citizens with disabilities in America, approaching a level of about 12.8%. INEC (Instituto Nacional de Estadisticas y Censos) publish data that coincide with that provided by the OAS and show that in this group of disa‐ bled citizens, the most significant disability in Ecuadorian children is Cerebral Palsy (CP). ICT (Information and Communications Technology, or Technolo‐ gies) directed their development programs to create devices and systems that focus on offering a new way to support the necessities of children with CP. This paper presents the results of a study conducted over a year and a half that concerned the feasibility of implementing a didactic transmitter (that combines an electronic device with a smart app developed for Smartphones) on very young patients with spasticity in a Cerebral Palsy Institute. Additionally, we discuss results about possible improvements in the quality of life of children with cerebral palsy. Keywords: Children with cerebral palsies · Embedded system · Android OS · Learning process · Mobile applications · Neurological problems · World population · Social impact

1

Introduction

Ecuador ranks third among nations in America in the rate of citizens with disability, considering about 12.8% of their population is in a vulnerable state [1, 2]. In 2010, the national statistical institutes indicated that 2.77% of the population had some kind of disability [1, 3, 4]. Subsequently, projects of social and educational inclusion were proposed [5–7]. However, in Ecuador to date, some vulnerable groups remain unat‐ tended. There is a lack of aid projects, such as the search for alternatives for improving communication and education for children with cerebral palsy and thus improving their quality of life [8].

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_7

74

J. Berrezueta-Guzman et al.

CP is a cerebral disorder that occurs in the motor cortex causing deficiencies in communication and learning processes, in addition to significant motor deficiencies. Many theories about its causes; it is defined broadly as a disorder in motor and intel‐ lectual cerebral function [9]. This disorder is diagnosed during the first years of a child’s life, but is presumed that is occasioned in the embryonic stage by cerebral lesions or hypoxia (less level of oxygen in the brain). Many diagnostics are employed to observe the intellectual and motor behavior of the child during the first three months of life. The condition manifests in movements of extremities and mobility in three primary ways: paraplegia (lower extremities unmoved), hemiplegia (one hemisphere of the body unmoved), and quad‐ riplegia (higher and lower extremities unmoved) [10]. The treatment of children with CP is provided in specialized centers and institutes. In these centers, the children are attended by therapists to stimulate their motor abilities, and by specialized teachers to promote the development of a process learning focus about basic and elemental things. These centers or institutes are constantly looking for new methodologies to give to the patients a kind of independence in the development and conduct of their daily activities [11]. A global figure as to the number of children with CP is not known, but it is frequently stated that around the world, CP is presented among one to two births on average of each one thousand (Aranda, 2008). Based on this rate of incidence, Ecuador (a country with a poor economy) would be expected to have about fifteen thousand patients (between children, teens, and adults) with CP. However, not all of them are in a speci‐ alized center for development of their motor abilities and their learning due to their low resources [9, 12, 15]. Ecuadorian education and healthcare systems have tried to adapt their infrastructures and regulations to offer an inclusive education and free access to health service to these people, but this hasn’t been enough. Many situations exist where this vulnerable group needs personalized assistance to support their learning and communication procedures [12, 13]. This research is focused on qualitatively and quantitatively analyzing a didactic transmitter (in essence an assistance system) for children with CP at a specialized insti‐ tute. Results have been obtained by tests over one and a half years with approximately one hundred children from this institute. The next sections explain the assistance system, the tests that were conducted, the quantitative results, the qualitative interpretation, remarkable successful cases, and final conclusions [14].

2

Composition of the Assistance System

The didactic transmitter contains a very small embedded system that works to provide communication and learning process assistance to children with CP. This device uses a technology based on the processing of angular inclinations to indicate required actions on a smart device or smartphone (with an Android operating system). The user’s actions manipulate multimedia content and represent a new kind of communication [8].

How Is the Quality of Life of Patients with CP Improved?

75

The data acquisition is accomplished by an angular inclinations sensor. The data is processed by the onboard microprocessor. Connectivity with the smartphone is based on Bluetooth protocol [8]. Due to kinds of motor disorder present in patients with cerebral palsy, the device can adapt its functions to the patient’s needs. Hence, this device can be used by a hemiplegia, quadriplegia and paraplegia patients because the functionality is based on an accessory (with the embedded system) that allows the patient to use his/her body parts that can move voluntarily and generate angular inclinations. For example, head movements in quadriplegia patients, hand movements in hemiplegia and paraplegia patients, or using a head belt and a bracelet respectively as an accessory [8]. The applied methodology in this project converts this combination electroniccomputing into a system that gives the patient a way of communicating by means of slight movements that he/she makes with an area of his body that is not paralyzed [8].

3

Functionality and Tests of the Assistance System

The assistance system was tested at the institute over eighteen months with forty-seven children (20 girls and 27 boys) ages 7–12 years who have physical conditions like hemiplegia and quadriplegia (see Table 1). Table 1. Classification of the sample of children with cerebral palsy depending on the type of paralysis, gender, and ages. Age (years) 7 8 9 10 11 12 Total

Quadriplegia Boys Girls 1 1 2 4 5 2 1 2 2 3 1 0 12 12

Hemiplegia Boys Girls 2 1 4 2 2 1 5 2 1 1 1 1 15 8

Figure 1 shows the classification of the sample of children with cerebral palsy ranging from seven to twelve years. Note: In this figure, the prisms correspond to the samples of children with quadriplegia and the cubes represent a sample of children with hemiplegia. Acceptance tests consisted of making the device acceptable to the child; that is, assuring that it did not cause discomfort or irritation. When the device passed this test, its operation was evaluated. The functional evaluation consisted of evaluating the device and the child. Each one depends on the other. The device’s operation is successful only if the child understands the instructions of the training that the device gives him. Hence, the device passes the evaluation test if, after training, it properly interprets the actions that the child uses to express himself in any circumstance.

76

J. Berrezueta-Guzman et al.

Number of child's

DistribuƟon of paƟents with cerebral palsy

5 4 3 2 1 0

Boys Girls 7

8

9

10

11

12

Ages

Girls (Q)

Girls (H)

Boys (Q)

Boys(H)

Fig. 1. Distribution of the sample of children with cerebral palsy regarding their ages, gender, and states of physical paralysis, like is shown in Table 1.

4

Research Results

After analyzing the operation of the device for about a year and a half it can be concluded that the device was beneficial in more than fifty percent of patients. In addition, since

EvoluƟon of feasibility results of the device 20

19

Number of child's

18 16

14

14 12

10

10

8

8 6 4 2 0

5 1 01 to 03

6

6 4

4

07 to 09

10 to 12

5

2 04 to 06

Months

13 to 15

16 to 18

Quadriplegia Hemiplegia

Fig. 2. Evolution of the results of feasibility of the device during the eighteen months of operation and tests in children with cerebral palsy.

How Is the Quality of Life of Patients with CP Improved?

77

the test period has ended, the number of children who learn to use it to communicate and learn has gradually been increasing. The device passed a series of functionality and acceptance tests over the last eighteen months. Also, advances have been made in the number of children who have gradually acquired skill in using the device, and who have developed a new kind of communication and learning method based on manipulation of multimedia contained on a smartphone (Fig. 2). To these results are added surveys answered by therapists and teachers at the institute to determine the acceptance and appreciation of the device. Results indicate that the system as a whole represents a very innovative alternative and does not interfere with the daily activities of children in the institute. This result motivates us to propose new versions with better characteristics and smaller size (Fig. 3).

Final Results 11

Number of child`s

8 11 10 9 8 7 6 5 4 3 2 1 0

6 2

Boys

Girls

Quadriplegia Hemiplegia

Fig. 3. Graph of the results on the functional tests in the last month of testing, where the number of boys and girls and their motor disability.

The cases of failure were reduced as the training progressed with the children and with the devices until the final results (shown in Fig. 4) were reached. We intend to reduce this percentage of error cases as much as possible through continuing develop‐ ment and research. New versions of the device are being planned to eradicate previous errors.

78

J. Berrezueta-Guzman et al.

Final results with respect to the iniƟal sample 17% 43%

40%

Fig. 4. Graph of the results with respect to the initial sample where the cases of success in quadriplegia and hemiplegia are indicated with a rate of 17% and 40% respectively, while the cases of failure are shown at 43% percent.

5

Qualitative Analysis

The device not only helped to improve the process of communication and learning but also was a fundamental component of the didactic program carried out by the therapist with his patients. However, what is most striking is how the device has been providing the patient with a radical change in their life. The therapists report that the device is adapted very well to the patients, and that the contents become more extensive each time they use it, enabling boy or girl to learn more. The device covers specific areas of support both in the cerebral palsy institute (with the therapist’s or teacher’s supervision) as well as in cases where the device has been used in the homes (with the supervision of the child’s parents). Additionally, some of these benefits were not taken into account when the objectives of the system were considered. Among the beneficial results are: • Communication of the child’s needs and feelings to their therapist or a close family member. • Concurrent didactic experiences supporting learning elementary things with their meanings during therapeutic sessions. • Less supervision during the child’s recreation periods. • Less dependency by the child on his caregiver because he now has a kind of multi‐ media distraction that additionally teaches like a virtual teacher. • Adaptability of the device to children of different ages and physical conditions; its user interface is purely graphical, and it is easy to understand for the therapist as for the child.

How Is the Quality of Life of Patients with CP Improved?

79

• Very compact and accessible new version of the device; its cost is relatively low approaching US $60 (only the didactic transmitter). The following section describes some particular cases that show the effectiveness of the device even outside the institute where it was tested applied. In order to safeguard the identity of the patients (because they are children), names have been ignored in the description of these cases. Case 1. It is a girl who had problems to be in a place with little space. Her parents traveled very frequently with her, but she did not like being in the car for a long time. During a lengthy journey, she would soon become exasperated, so they would have to stop every so often so that she could take a breath and re-enter the car. The device provided a very good distraction because the system kept her entertained all the way giving instructions and interacting with her so that she no longer feels suffocated and forgets about being in a small and enclosed place while in the car. Case 2. The device has taught several children to recognize which members make up their family and school mates. Since it is open to new images and audiovisual content, the therapist or the father of the child can upload photos of the members of the household as well as the therapists and colleagues of the institute. The device has now enabled several children to know how to identify their family and friends both in games and in situations of daily activities.

6

Final Conclusions

This project had an innovative proposal and achieved excellent results. However, it required a long time to design and develop the device. Throughout this time, the tests and results showed that the device can get better each time, not only to provide an alternative communication channel for the patient but also to offer a possibility to develop an independent learning process that motivates the patient to learn to handle an Android device easily. The therapist’s surveys regarding perceptions and the transmitter interface revealed that the use is broad in support of learning. The in general believes that the device can improve the educational process, due to the multimedia content presented and the possibility of increasing the material to address individual tastes and needs. The results obtained and the analysis of several cases of success, the comments and observations provided by therapists and parents (who are in constant contact with these children) all affirm that the device has made a significant change in the communicative and didactic process. All the compiled results and the observations made during the eighteen months in the institute confirm that systems engineering has made possible a remarkable improvement of the quality of life of a child with cerebral palsy. From the technical point of view, the transmitter device meets the objectives set for the project; from a more human perspective, we can see how an innovative idea that does not require many economic resources can change the lives of these children. This research shows the impact that systems engineering, which is oriented to resolving challenging systems requirements can have when focused on a vulnerable segment of the health sector. The results encourage the making of proposals for more

80

J. Berrezueta-Guzman et al.

research and technological developments that are directed towards the purposes of health care. The purpose of this project was to show how solutions can be provided that do not represent costs that are inaccessible to those who need it. All that is required is to inno‐ vate how the elements immersed in our daily activities are used. The purpose of this project is to show how solutions can be provided that do not require extreme costs such that the only solutions that are available are those that are inaccessible to those who need it. We only need to innovate the ways in which the elements immersed in our daily activities are used. Acknowledgments. We would like to thank Dr. Hernán Tenorio, and Ing. Jessica Orellana, and all the collaborating people from Intituto de Parárilis Cerebral del Azuay (IPCA).

References 1. Organización Mundial de la Salud, & Banco Mundial: Informe mundial sobre la discapacidad (Resumen). Organización Mundial de La Salud, 27 (2011). http://www.who.int/disabilities/ world_report/2011/accessible_es.pdf?ua=1 2. Organización Mundial de la Salud: Informe mundial sobre la discapacidad. Educación (2011). http://who.int/disabilities/world_report/2011/summary_es.pdf 3. Flores, R., Yepez, D., Pramatarova, M.: Ecuador: La discapacidad en cifras. INEC, Ecuador (2005) 4. CONADIS, C.N. de la I. de D: Agenda nacional para la Igualdad en discapacidades 2013– 2017. Conadis, vol. 1. AXIS Communications, “Guía técnica de vídeo IP” (2013) 5. Educación inclusiva: DOSSIER: Educación inclusiva. Revista Trimestral de Educación Comparada, XXXVIII(0304–3053), 1–181 (2008). www.ibe.unesco.org 6. Consejo Nacional de Igualdad de Discapacidades: Ecuador (2013). http://www.conse jodiscapacidades.gob.ec/wpcontent/uploads/downloads/2015/09/estadistica_conadis.pdf 7. Instituto Nacional De Estadísticas Y Censos: Boletas Censales período 1950–2010. Quito. http://www.ecuadorencifras.gob.ec/documentos/webinec/Publicaciones/Evolucion_ variables_1950_2010_24_04_2014.pdf 8. Berrezueta-Guzmán, J., Coronel-Maldonado, F., Serpa-Andrade, L., Robles-Bykbaev, V.: A didactic transmitter to support the communication and learning process of children with cerebral palsy. In: MATEC Web of Conferences, vol. 42. EDP Sciences, January 2016 9. Rivero, A.R., García-Celay, I.M., Lorente, M.C.: El niño con parálisis cerebral: enculturación, desarrollo e intervención, vol. 79. Ministerio de Educación (1993) 10. Milani-Comparatti, A., Gidoni, E.: Pattern analysis of motor development and its disorders. Dev. Med. Child Neurol. 9, 625–630 (1967) 11. Bobath, K., Kong, E.: Zerebrale Bewqungsst8rungen Beim Kind. S. Karger-Basel, Müchen (trad. cast.: Transtornos Cerebromotores en el niiio. 1986. Buenos Aires: Panamericana) (1976) 12. Phelps, W.M.: The role of physical therapy in cerebral plasy and bracing in the cerebral palsied. En orrhopaedie appliances Arlas 1, 251–522 (1952). Edwards, Ann Arbor 13. Kleinsteuber, K., María de los Angeles, A., Varela, X.: Parálisis Cerebral. Revista Pediatría Electrónica 11(2), 54–70 (2014). http://www.revistapediatria.cl/vol11num2/pdf/ 6_PARALISIS_CEREBRAL.pdf

How Is the Quality of Life of Patients with CP Improved?

81

14. Venegas, A.O., Gutiérrez, V.R., de Paula, L.A.A.: Revisión sistemática de las intervenciones para la estimulación en niños con retraso motor de 0 a 12 meses de edad. Revista Movimiento Científico 8(1), 118–130 (2014) 15. Aranda, R.E.: Atención temprana en educación infantil. WK Educación (2008)

Prevalence and Risk Factors of Respiratory Problems Among Solid Waste Collectors in the Philippines Lizbeth A. Mariano1 ✉ and Aura C. Matias2 (

1

)

Department of Industrial Engineering, College of Engineering and Agro-industrial Technology, University of the Philippines, Los Baños, Laguna, Philippines [email protected] 2 Department of Industrial Engineering and Operations Research, College of Engineering, University of the Philippines Diliman, Quezon City, Philippines [email protected]

Abstract. In most developing countries such as the Philippines, solid waste collection is done manually, exposing the solid waste collectors to health risks. The study determined the prevalence of respiratory health problems, one of the most common health issue among these workers. The study was done at Location XYZ, and found that the prevalence of respiratory problems among solid waste collectors is at 19%. Binary logistic regression was used and found 13 significant factors contributing to the incidence results: the age of the worker, his height (m), his body mass index (BMI), number of hours of sleep, drinking frequency, number of alcoholic drinks consumed, personal protective equipment (PPE) frequency of use, decision latitude, social support, management feedback, discrimination, years in service, and whether or not he went through a job training. Based on the statistically significant factors, ergonomic interventions were recommended to help mitigate the health problems of solid waste collectors. Keywords: Respiratory · Prevalence · Solid waste collectors

1

Introduction

Humans can be at risk either directly or indirectly to the hazards of solid wastes in different stages of the waste cycle [1, 2]. According to the Environmental Management Bureau (EMB) [1], one of the groups of people at risk is the solid waste collectors’ segment. Solid waste collectors are highly susceptible to health problems, injuries, and acci‐ dents. Adverse health effects usually encountered among this work group include respi‐ ratory, musculoskeletal, dermal, gastrointestinal, irritation of the nose and eyes, fatigue, headaches, allergies, and psychological problems [3, 4]. The study conducted focused on the respiratory symptoms and their prevalence being one of the usual health issues.

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_8

Prevalence and Risk Factors of Respiratory Problems

83

1.1 Scope and Limitations Due to time restriction, the study was conducted only in Location XYZ, one of the 16 cities of the National Capital Region (NCR) in the Philippines. The location of the study was kept anonymous as requested by its management. 1.2 Literature Review Several studies in other countries have focused on possible respiratory problems of solid waste collectors. A study in Kaohsiung County, Taiwan [5] found that symptoms such as “coughing, phlegm, chronic bronchitis, low back pain, elbow and wrist pain, and injuries caused by sharp objects” are more prevalent among waste collectors than office workers (control group). Lavoie and Dunkerley [6] assessed the exposure of waste collectors to bioaerosols during summer (representing worst conditions) which are said to have adverse impacts to their respiratory system. They measured the mean concentrations of bacteria, endo‐ toxins, and molds in the air and compared with a standard limit for each. Their findings show that each bioaerosol enormously exceeded the limits set. They also found that cleaning of empty garbage trucks does not affect the exposure of the collectors from bioaerosols and that the organic waste leachate is the main source of these. Athanasiou, Makrynos, and Dounias [7] found it necessary to concentrate on the respiratory health of solid waste collectors in Keratsini, Peraeus, Greece due to the increasing evidence among this group of work-related pulmonary problems than in the general workforce. Results in this study showed that municipal solid waste workers in Keratsini, Greece have a higher prevalence of respiratory symptoms and a greater decrease in their lung function than the comparison group of local office workers. A portable electronic spirometer was used in the study to test the lung function while a slightly modified version of the Medical Research Council Respiratory Questionnaire was used to probe the respiratory problems experienced by the workers. The study results showed that there is a moderate but statistically higher prevalence of sore throat, coughing in the morning, coughing on exertion, and phlegm production among waste collectors. According to the authors, one of the limitations of the study is that no temporal relationship of factors can be determined and no specific bioaerosol or chemical expo‐ sure data was available in their study. But at the end of the study, the authors suggested that occupational exposure to bioaerosols, dust, exhaust fumes, and bad weather condi‐ tions play an important role in the development of respiratory problems.

2

Methods

Data collection for the study was done through interviews, survey, and observations of actual operations. An officer from the Solid Waste Management Office (SWMO) of the Metropolitan Manila Development Authority (MMDA) was interviewed to determine how solid waste collection is being performed in Metro Manila. Then, the representative local government unit (LGU) to study was chosen.

84

L.A. Mariano and A.C. Matias

2.1 Sample Size A total of 330 respondents were included in the study out of the 1196 total solid waste collectors in the LGU. The samples size was computed using the Slovin’s formula, Eq. (1), with 95% level of confidence. A 10% allowance was also added to the number of respondents. n=

n N e

N 1 + Ne2

(1)

is the sample population is the population is the margin of error

2.2 Variables Included The solid waste collection process was observed. The activities that the collectors perform were examined to check for possible risk factors. The literature review was also used to determine the health considerations and risk factors that should be looked into. The following are the independent variables included in the study: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

12. 13. 14.

15. 16.

Age. The age of the worker in years. Civil status. The status of the worker whether single or married. Number of children. It is defined as the number of children under the worker’s care. Height. The height of the worker in meters. BMI. The worker’s Body Mass Index. Residence. Whether the worker lives within the vicinity of the dumpsite or not. Other jobs. The presence or absence of another job for the worker aside from being a solid waste collector. Medical history. The number of critical illnesses that the worker has in their family. Sleep. The average number of sleep hours the worker gets daily. Cigarettes consumed daily. It is defined as the number of sticks of cigarettes the worker smokes daily. Drinking Freq. The frequency that the worker consumes alcoholic beverages: (0) never, (1) occasional/once a month, (2) Twice monthly, (3) once a week, (4) 2–3 times a week, (5) daily. Number of drinks. The number of alcoholic beverages the worker consumes every‐ time. PPE use frequency. How frequently the worker uses personal protective equipment (PPE): (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. Number of shifts served. There are three shift schedules available, workers need to serve for two shifts per day to receive minimum wage, more than that they will be given an overtime pay. Years on the job. The length of time in years of the worker as a solid waste collector. Job training. Whether the worker went through a training program on the job to be assigned to him beforehand or not.

Prevalence and Risk Factors of Respiratory Problems

85

17. Assignment to residential biodegradable route. Whether the worker is assigned on the said route or not. 18. Assignment to residential non-biodegradable route. Whether the worker is assigned on the said route or not. 19. Assignment to commercial route. Whether the worker is assigned on the said route or not. 20. Assignment to institutional biodegradable route. Whether the worker is assigned on the said route or not. 21. Assignment to institutional non-biodegradable route. Whether the worker is assigned on the said route or not. 22. Decision Latitude. The frequency that the workers perceive they have control over their work: (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. 23. Work Demand. The frequency the worker perceive that the worker perceives that the physical requirement of the job cannot be reached: (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. 24. Community Support. The frequency the worker perceive that community members cooperate to their job aims (waste segregation and collection schedule): (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. 25. Social Support. The frequency the worker perceive that they receive help from superiors and co-workers during work-related problems: (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. 26. Management Feedback. The frequency the worker perceive that feedback is being given by superiors and co-workers on their job performance: (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. 27. Discrimination. The frequency the worker perceive that he experienced discrimi‐ nation from the community due to the nature of their job: (4) always, (3) often, (2) sometimes, (1) seldom, (0) never. A survey questionnaire was created to determine the individual, behavioral, psycho‐ social, and work-related characteristics of the workers. Meanwhile, the last part of the questionnaire probed possible presence/absence of respiratory symptoms of the workers. The questionnaire was administered to the 330 solid waste collectors. 2.3 Data Input and Analysis Data from the survey was summarized using an MS Excel spreadsheet and data splitting was done: 70% for constructing the five models and the remaining 30% for validating the constructed models. A total of 311 data points were left from the initial of 330 after removal of those with missing values. Software used was SPSS version 23. Descriptive statistics were computed and used to characterize the subjects being studied and have an initial insight on the data charac‐ teristics. The statistical method used is binary logistic regression, Backward: LR type (Stepwise). This was used to have the full model first and avoid disregarding interactions among the variables. The seven assumptions for this method were tested before applying it in generating the models.

86

L.A. Mariano and A.C. Matias

For the validation, the SPSS scoring wizard was used to check how these models can accurately predict the presence/absence of the respiratory problem symptoms. The number of times out of the 90 validation points that the models correctly predicted the outcomes was used to determine the percentage of correctness of the models.

3

Results and Discussion

3.1 Descriptive Statistics The respondents surveyed were 100% male. Majority of them, 69.13%, are married while 27.65% are single and the rest are either separated or widowers. Half of the respondents, 50.16% have one to three children, a large portion also have no children, 34.41%, and the rest (15.44%) have four and above children. They are composed mainly (68.17%) of young to middle-aged adults, 21 to 40 years of age. In terms of residence, majority lives within the vicinity of the dump site, 77.81% and 85% depends on their job as waste collector as source of living. Some aspects of the lifestyle of the respondents were also explored. Young adults (18 to 25 years old) and adults (26 to 64 years old), need seven to nine hours of sleep daily [8]. The solid waste collectors belong to these age brackets and from the survey, it was found that only 33.76% of them are well-rested going to work. On the smoking habit, only 30.87% of the respondents were non-smokers. According to studies, there is no safe level of smoking and light smokers are still at risk of certain heart and lung diseases [9, 10]. Thus, the solid waste collectors are at risk In terms of consumption of alcoholic drinks, only 29.26% of the sample population were non-drinkers. Chronic heavy drinking leads to lower immunity and risk to bacterial and viral infections [11]. Out of the 311 solid waste collectors interviewed, 19.29% or 60 of them experience respiratory problems. 3.2 Binary Logistic Regression Result Table 1 shows the summary of the results from the binary logistic regression. The Omnibus test returned a p-value lower than alpha of 0.00 for the model and signifies that the model is statistically significant in predicting the presence/absence of respiratory problems. A relatively high value of Nagelkerke R-square is available at 59.6%. Hosmer and Lemeshow test value of 0.740 indicates that the model is not of poor fit. The overall percentage in accuracy of the model is also high at 90.3%. Table 1. Summary of results of the logistic regression for the respiratory model. Test Value Omnibus Tests of Model Coefficients (Sig.) 0.000 Nagelkerke R Square 0.596 Hosmer and Lemeshow Test (Sig.) 0.740 Percentage Correct (Model) 90.3

Prevalence and Risk Factors of Respiratory Problems

87

The statistically significant factors involved for this model are the age of worker, height, BMI, sleep hours, number of drinks, PPE use frequency, decision latitude, management feedback, discrimination, number of years as solid waste collector, and whether the worker went through job training or not (Table 2). Table 2. Independent variables found to be statistically significant. Variable

Sig.

Exp (B)

Age (years) Height (m)

0.0071 0.0161

Bmi Sleep (hours) Drinking freq Number of drinks Ppe use frequency Ppe use frequency (2) Decision latitude Decision latitude (3)

0.0192 0.0048 0.0328 0.0006 0.0048 0.0002 0.0013 7.3331 E-05 0.0353 0.0207 0.0023 0.0071 0.0002 0.0019 0.0007 0.0077

1.1186 7.8196 E-06 0.7318 0.6428

95% CI for Exp (B) Lower Upper 1.0309 1.2138 5.4190 0.1128 E-10 0.5635 0.9505 0.4730 0.8736

1.1034

1.0434

2062.4518

38.6624

87.0286

9.5699

791.4396

51.7043

1.8257

1464.2806

0.0873 0.0062

0.0148 0.0004

0.5155 0.0887

0.7737 0.0320

0.6674 0.0025

0.8970 0.4020

Social support Social support (1) Management feedback Management feedback (3) Management feedback (4) Discrimination Years on the job Job training

1.1669 110021.7200

The first significant factor, age, affects the odds of having respiratory problems of the workers such that as the years increase, the odds also increase by 1.1186. The same is true for the number of drinks consumed, as the number of alcoholic drinks consumed by a worker increases, the odds of having a respiratory problem increases by a factor of 1.1034. For the other factors such as height, BMI, sleep hours, and years as waste collector, the opposite is true. As these factors increase by one unit, the odds are reduced. The interesting result on the years of service as a solid waste collector could be due to the immunity that the workers develop as they are exposed daily to the bioaerosols coming from decaying wastes. However, the next factor of PPE use frequency interestingly indicates that a worker who “Always” uses his PPE will have a higher risk of respiratory problems by 2062.45 than those who use the PPE at “Seldom” frequency. This could be due to the nature of the material that is being used as facemask. The facemask being issued to the workers are made of cloth which needs to be re-used every day. Workers also stated during the

88

L.A. Mariano and A.C. Matias

interviews that these facemasks will be contaminated from exposure to the field due to daily usage, possibly the reason for this interesting result. Meanwhile for the psychosocial factors, decision latitude, social support, manage‐ ment feedback, and discrimination, the results are different. For decision latitude, as the workers perceive more freedom in their job, the more will be their odds of having the disease. This is because if a worker gets too much freedom, some safety precautions such as use of PPE may be overlooked so the higher is the risk for them. Social support variable has the same result, as the workers perceive more support from their co-workers (Seldom to Always), the odds increases, possibly in some instances other workers would want to have more independence than being supported too much by their co-workers (there is a limit for them). For the management feedback variable, as the feedback from management increases in frequency from “Sometimes” or “Often” to “Always”, the odds of having respiratory disorder decreases. This is due to the fact that workers need some form of feedback, may it be in the form of appreciation or correction, in order for them to feel the purpose of the responsibilities that they are doing and the results coincide with this theory. The model was validated using SPSS scoring Wizard and a percentage correct of 70% was attained.

4

Conclusion

The study found that respiratory problems among solid waste collectors in the Philip‐ pines indeed exist and a sector that should somehow be given attention. The risk factors found would help in determining areas for improvement in this endeavor. Acknowledgements. The authors would like to thank the workers who participated in the study and the local government unit members who assisted, the panel members who helped improve the study, and the ERDT Program and UP System for the financial assistance.

References 1. Environmental Management Bureau (EMB): National state of the brown environment report (2005 – 2007) (2009). http://emb.gov.ph/eeid/publications.htm 2. Cointreau, S.: Occupational and environmental health issues of solid waste management special emphasis on middle- and lower-income countries (2006). http:// documents.worldbank.org/curated/en/2006/07/7826263 3. Bogale, D., Kumie, A., Tefera, W.: Assessment of occupational injuries among Addis Ababa city municipal solid waste collectors: a cross-sectional study. BioMed Cent. 14, 169 (2014) 4. An, H., Englehardt, J., Fleming, L., Bean, J.: Occupational health and safety amongst municipal solid waste workers in Florida. Waste Manage. Res. 17, 5 (1999) 5. Yang, C.Y., Chang, W.T., Chuang, H.Y., Tsai, S.S., Wu, T.N., Sung, F.C.: Adverse health effects among household waste collectors in Taiwan. Environ. Res. 85(3), 195–199 (2001) 6. Lavoie, J., Dunkerley, C.J.: Assessing waste collectors’ exposure to bioaerosols. Aerobiologia 18(3), 277–285 (2002)

Prevalence and Risk Factors of Respiratory Problems

89

7. Athanasiou, M., Makrynos, G., Dounias, G.: Respiratory health of municipal solid waste workers. Occup. Health 60(8), 618–623 (2010) 8. Hirshkowitz, M., Whiton, K., Albert, S.M., Alessi, C., Bruni, O., DonCarlos, L., Hazen, N., Herman, J., Adams Hillard, P.J., Katz, E.S., Kheirandish-Gozal, L., Neubauer, D.N., O’Donnell, A.E., Ohayon, M., Peever, J., Rawding, R., Sachdeva, R.C., Setters, B., Vitiello, M.V., Ware, J.C.: National sleep foundation’s updated sleep duration recommendations: final report. J. Nat. Sleep Found. 1(4), 233–243 (2015) 9. Bjartveit, K., Tverda, A.: Health consequences of smoking 1–4 cigarettes per day. Tobacco Control. 14, 315–320 (2005). doi:10.1136/tc.2005.011932 10. Schane, R.E., Ling, P.M., Glantz, S.A.: Health effects of light and intermittent smoking: a review. NIH Public Access. 121(13), 1518–1522 (2010). doi:10.1161/CIRCULATIONAHA. 109.904235 11. Barr, T., Helms, C., Grant, K., Messaoudi, I.: Opposing effects of alcohol on the immune system. Prog. Neuropsychopharmacol. Biol. Psychiatry 65, 242–251 (2016). doi:10.1016/ j.pnpbp.2015.09.001

Prevalence, Severity, and Risk Factors of Work-Related Musculoskeletal Disorders Among Stevedores in a Philippine Break-Bulk Port Terminal Keneth Sedilla(&) and Aura Matias Department of Industrial Engineering and Operations Research, University of the Philippines Diliman, Quezon City, Philippines [email protected], {kbsedilla,acmatias}@up.edu.ph

Abstract. While it plays a vital role in a country’s economic development, port operations host a wide spectrum of health problems and work-related diseases. This study aimed at determining the prevalence, severity, and significant risk factors that contribute to the occurrence of work-related musculoskeletal disorders among stevedores in a Philippine break-bulk port terminal. The stevedores were asked to identify the work-related musculoskeletal symptoms they had experienced over the last three months. In this cross-sectional study, the results showed that self-reported prevalence and severity of work-related musculoskeletal symptoms were significantly higher in the upper and lower back, shoulder, and forearm regions. Some of the identified risk factors associated with the severity of musculoskeletal symptoms are age, work experience, number of working days per week, vessel work assignment, type of steel commodity, level of job satisfaction, and perceived job security among others, suggesting that the etiology of this condition is multifaceted. Keywords: Human factors
Work-related musculoskeletal disorders Break-bulk port terminal
Stevedoring operations
Stevedores




1 Introduction Musculoskeletal disorder (MSD) is also known as repetitive strain injury, repetitive motion injury, cumulative trauma disorder, occupational overuse syndrome, musculoskeletal injury, and sprain and strain [1–3]. These terms are based on the idea that injuries in the muscles, tendons, ligaments or any musculoskeletal regions are caused or made more severe by intense repeated, or sustained use, or by insufficient recovery from exertions, motions of the body, vibration, or cold [1]. Moreover, there are some clarifications to be aware of regarding the use of the terms “disorder” and “injury”. An injury occurs when there is a mechanical disruption of tissues while a disorder occurs even without disruption of tissues. The onset of a disorder may be gradual while injury is sudden [1].Also, we note that musculoskeletal disorders are not always necessarily work-related, because the disorder may be due to other aspects of daily life, © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_9

Prevalence, Severity, and Risk Factors

91

including, for example, sports activities and frequently repeated household activities such as gardening. MSDs are work-related when they are significantly contributed to or exacerbated by work-related factors such as work postures, work shifts and so on [4]. Musculoskeletal symptoms are often intermittent and episodic, especially in the early stages [5]. They may develop over weeks, months or years [1, 5–7]. Musculoskeletal disorders arise from several simultaneously contributing factors and hence are multifactorial in origin [1, 5, 8–11]. Musculoskeletal disorders affect all workplaces [8]. With its substantial health and economic implications [12], musculoskeletal disorders have received increasing attention across different industries [8]. Any major port terminal is an important working environment for the local economy and the world market [13]. Such facilities are situated on a coastline where vessels can dock and transfer passengers to or from land. One of its primary functions is to efficiently manage the loading and/or unloading of goods or commodities to or from cargo vessels [14]. Lift-on/lift-off operations are the most common type of cargo vessel operations, where a vessel crane or mobile crane is used to transfer the cargoes from the ship’s lower decks through deck hatches to the port’s apron and/or vice versa. Break-bulk port terminals accommodate wide varieties of break-bulk cargoes. Such cargoes are loaded into ships as individual pieces or are unitized on pallets, in bundles, and not containerized. While playing a vital role in a country’s economic development, port operations also contribute to a wide spectrum of worker health problems and work-related diseases. Globally, there are 2.34 million work-related deaths per year; around 86% are due to work-related diseases, while others are accident-related [13]. Workers involved in port operations such as stevedores or dockworkers carry out a set of tasks, which, if not properly designed and managed, may result to a wide variety of musculoskeletal disorders [15]. There had been a few studies looking into understanding the possible causes of MSD among port workers. Of the 6 related studies, 3 are focusing on port equipment/machinery operators [16–18]. In addition, there is some literature focusing on the prevalence of occupational risks among dockworkers in container port terminals [19–21].

2 Methods 2.1

Sampling Frame

This cross-sectional study was conducted in the largest break-bulk cargo terminal within the Port District Office of Manila/Northern Luzon [22]. There are 561 stevedoresunder the port’s accredited stevedoring agencies. Generally, the stevedores’ work assignment varies depending on the number and type of vessels at berth. Thus, to increase the objectivity of the results of the association of risk factors to the symptoms of work-related musculoskeletal disorders, purposive sampling was used. In order to qualify as a respondent in this study, the stevedore must be previously assigned to steel vessels more frequently than other types of commodities such as logs, lumber, bagged cargo, and general cargo. Based on the port’s stevedoring database, steel has the highest annual cargo throughput. In addition, the stevedores who had any history of major musculoskeletal trauma regardless of whether it was work-related or not were excluded in the study. Based onthe selection criteria, fifty-two percent of the stevedores (290) were qualified to participate in the study.

92

2.2

K. Sedilla and A. Matias

Data Collection

Preliminary interviews and worksite observations were conducted prior to the deployment of the employee survey questionnaires. Questionnaires with self-reported musculoskeletal symptoms were used in conducting surveillance of work-related musculoskeletal disorders. The questionnaires include stevedores’ individual attributes such as age, sex, marital status, highest educational attainment, body mass index, etc. Further, there are sets of questions pertaining to work-related factors such as work characteristics and psychosocial factors. In addition, the stevedores were asked to identify the work-related musculoskeletal symptoms (pain, tingling sensation, etc.) they had experienced over the last three months using the body discomfort chart. Subsequently, the frequency and intensity of each musculoskeletal symptom per body region (neck, shoulder, upper back, lower back, forearm, etc.) were also identified.

2.3

Statistical Analysis

In determining the significant predictors of the symptoms of work-related musculoskeletal disorders, ordinal logistic regression models for the body regions with the highest prevalence rates were generated using IBM SPSS statistics software (version 24). All the individual and work-related factors were included in the regression models so that the effect of their interactions to the severity of work-related musculoskeletal symptoms per body region is considered. In this study, there are eleven (11) individual factors and fifteen (15) work-related factors linked to the severity of symptoms of work-related musculoskeletal disorders.

3 Results 3.1

Sample Demographics

Two hundred ninety stevedores participated in the study. All the respondents were male, aged 19 to 56 years old. The stevedores’ average height and weight are 161.59 cm and 63.50 kg respectively. A majority of the respondents were married (81%). Moreover, eighty percent (80%) of the stevedores are high school graduates. Approximately three-fourths (72%) of the stevedores do not exercise regularly and/or perform very light physical activities intermittently. The average weekly consumption of alcoholic drinks is 3.08 alcohol units. Moreover, seventy seven percent of the respondents are smokers. On the average, a stevedore consumes 3.61 cigarettes per day. The stevedores work an average of 11.56 (SD 0.26) hours per shift. The respondents’ average working days per week is 3.43. Forty-six percent of the stevedores were assigned to steel billets more frequently than wire rods and coils. A little more than half (53%) was frequently assigned to work ondock/quayside. In regards to the type of shift in which stevedores were assigned to, the results showed that the frequency for day shift and night shift is nearly equal. In relation to psychosocial factors, almost half (44.48%) of the stevedores are slightly dissatisfied with their job while 39.66% gave a

Prevalence, Severity, and Risk Factors

93

satisfactory rating for their job satisfaction. Most of the stevedores (53.45%) seldom experienced conflicts at work. Moreover, 53.79% of the respondents receive little support from their co-employees. Of the 290 stevedores, 126 (43.35%) stated that they are receiving little support from the management. More than half (53.79%) of the stevedores have experienced low level of decisional latitude at work while 182 of the total respondents stated that they had low of perceived job security.

3.2

Prevalence of Work-Related Musculoskeletal Disorders Among Stevedores

Table 1 presents the prevalence of work-related musculoskeletal symptoms on various body regions among the stevedores. The body regions with high prevalence of self-reported work-related musculoskeletal symptoms are the upper (60%) and lower back (58%), shoulder (64%), and forearm (62%). Table 1. Prevalence of symptoms of work-related musculoskeletal disorders per body region. Body region Neck Shoulder Upper back Lower back Forearm Hand/Wrist Hip/Buttocks Thigh Lower leg Ankle/Foot

3.3

No. of stevedores (Percent) 69 (24%) 187 (64%) 175 (60%) 169 (58%) 180 (62%) 88 (30%) 65 (22%) 56 (19%) 41 (14%) 27 (9%)

Severity of Work-Related Musculoskeletal Disorders Among Stevedores

Table 2 presents the summary of the severity of the symptoms of work-related musculoskeletal disorders per body region. Based on the results, the upper and lower back, shoulder, and forearm region has the higher percentage of stevedores experiencing medium to high severity of work-related musculoskeletal symptoms.

3.4

Risk Factors of Work-Related Musculoskeletal Disorders (Shoulder)

Work experience in a break-bulk port terminal (expressed in yrs.) has a strong positive association with the severity of the work-related musculoskeletal symptoms. The odds ratio (OR) indicates that stevedores with more years of work experience are 2.53 times more likely to experience high severity than those with no/less experience working in break-bulk port terminal. Further, the regression model shows that a unit increase in

94

K. Sedilla and A. Matias

Table 2. Severity of all the symptoms of work-related musculoskeletal disorders per body region (average rating of all musculoskeletal symptoms per body region) Body region No. of stevedores (Percent) Low Medium Neck 230 (79.31%) 52 (19.93%) Shoulder 161 (55.52%) 99 (34.14%) Upper back 150 (51.72%) 118 (40.69%) Lower back 132 (45.52%) 133 (45.58%) Forearm 186 (64.14%) 76 (26.21%) Hand/Wrist 208 (71.72%) 73 (25.17%) Hip/Buttocks 236 (81.38%) 43 (14.83%) Thigh 259 (89.31%) 26 (8.97%) Lower leg 252 (86.90%) 31 (10.69%) Ankle/Foot 277 (95.52%) 13 (4.48%)

High 8 (2.76%) 30 (10.34%) 22 (7.59%) 25 (8.62%) 28 (9.66%) 9 (3.10%) 11 (3.79%) 5 (1.72%) 7 (2.41%) 0 (0%)

stevedore’s age (expressed in yrs.) is associated with a twenty-percent increase in the odds of having high severity. The results also shows that weekly consumption of alcoholic drinks has a positive association with work-related musculoskeletal symptoms. With regard to psychosocial factors, the ordinal logistic regression model indicates that stevedores who were completely dissatisfied with their job (as compared to those who have a very high job satisfaction) had 185.50 times higher odds of reporting high severity of musculoskeletal. Likewise, stevedores who felt that they have no assurance of keeping their job had 105.69 higher odds of having high severity level of shoulder work-related musculoskeletal symptoms. The identification of stevedores’ vessel work assignment as a shoulder MSD risk factor was an interesting finding of the study, and one that incurred a high odds ratio (OR: 66.05). Moreover, the type of steel commodity specifically the steel billets is also a significant work-related risk factor with odds ratio of 18.97. With regard to stevedores’ average working days per week, there is 3.22 higher odds of experiencing high severity of musculoskeletal symptoms for every one-day increase in a stevedore’s average working days per week.

3.5

Risk Factors of Work-Related Musculoskeletal Disorders (Forearm)

The ordinal logistic regression model implies that there is an 8% increase in the probability of experiencing high severity of symptoms for every additional year in a stevedore’s age. The stevedores’ weekly consumption of alcoholic drinks has a positive association with the severity level of musculoskeletal disorders. This indicates that a 20% increase in the probability of having high severity of symptoms of MSD for a unit increase in stevedore’s alcoholic drinks consumption per week. The results show that stevedores who are working inside the vessel hatch (as compared to those who are working on-dock) have 12.79 times higher odds of experiencing high severity of MSD symptoms. Moreover, the regression model specifies that for every one day increase in stevedores’ average working days per week, there is

Prevalence, Severity, and Risk Factors

95

3.23 higher odds of having high severity of work-related musculoskeletal symptoms. The psychosocial factors associated with the severity level of work-related musculoskeletal disorders are job satisfaction and perceived job security. Apparently, the results are nearly similar with the shoulder region in which the values of odds ratio are also high. The regression model states that stevedores who were completely dissatisfied in their job (as compared to those who have a very high job satisfaction) had 10.69 times higher odds of showing high severity of MSD. With regard to perceived job security, the findings show that those who felt that they have no assurance of keeping their job in the port terminal for a period of time had 4.11 higher odds of having high severity level of musculoskeletal symptoms (Tables 3). Table 3. Ordinal logistic regression model for risk factors associated with the severity of musculoskeletal symptoms in the shoulder region (only significant factors are shown below) Factors Age Yrs. in break-bulk port Alcoholic drinks consumption Average days/week Vessel work assignment Quayside (Ref.) Vessel hatch Steel commodity Coils (Ref.) Wire rods Steel billets Job satisfaction Very satisfied (Ref.) Satisfied Slightly dissatisfied Dissatisfied Job security High Medium Low None

3.6

Coefficient 0.184 0.928 0.381 1.169

P value 0.000* 0.004* 0.001* 0.002*

Odds ratio (95% CI) 1.20 (1.10–1.31) 2.53 (1.34–4.77) 1.46 (1.17–1.83) 3.22 (1.52–6.82)

– 4.190

– 0.000*

1.00 66.05 (15.73–277.35)

– 0.830 2.943

– 0.343 0.000*

1.00 2.29 (0.41–12.78) 18.97 (4.20–85.64)

−0.856 2.802 5.223

– 0.411 0.008* 0.001*

1.00 0.43 (0.06–3.27) 16.49 (2.09–129.72) 185.50 (9.43–3649.37)

−0.042 2.054 4.660

– 0.965 0.041* 0.001*

1.00 0.96 (0.15–6.25) 7.80 (1.08–56.03) 105.69 (7.14–1564.30)

Risk Factors of Work-Related Musculoskeletal Disorders (Upper Back)

Table 5 shows the results of the ordinal logistic regression model for the respondents’ upper back region. Among the individual attributes, age is a significant risk factor for upper back MSD. The model specifies that there is a 10% increase in the probability of experiencing high severity of work-related musculoskeletal disorders for an additional

96

K. Sedilla and A. Matias

Table 4. Ordinal logistic regression model for risk factors associated with the severity of musculoskeletal symptoms in the forearm region (only significant factors are shown below) Factors Age Alcoholic drinks consumption Average days/week Vessel work assignment Quayside (Ref.) Vessel hatch Job satisfaction Very satisfied (Ref.) Satisfied Slightly dissatisfied Dissatisfied Job security High Medium Low None

Coefficient 0.072 0.183 1.174

P value 0.015* 0.032* 0.000*

Odds ratio (95% CI) 1.08 (1.01–1.14) 1.20 (1.02–1.42) 3.23 (1.79–5.85)

– 2.549

– 0.000*

1.00 12.79 (4.51–36.25)

−1.379 0.471 2.370

0.096 0.546 0.038*

1.00 0.25 (0.05–1.28) 1.60 (0.35–7.40) 10.69 (1.14–100.27)

−0.109 0.659 1.413

– 0.881 0.373 0.016*

1.00 0.90 (0.22–3.74) 1.93 (0.45–8.24) 4.11 (0.66–25.46)

year in stevedores’ age. In addition, weekly consumption of alcoholic drinks and cigarette smoking are also significant predictors for the severity of the symptoms of work-related musculoskeletal disorders. Table 5. Ordinal logistic regression model for risk factors associated with the severity of musculoskeletal symptoms in the upper back region (only significant factors are shown below) Factors Age Cigarette consumption Alcoholic drinks consumption Vessel work assignment Quayside (Ref.) Vessel hatch

Coefficient 0.099 0.055 0.069

P value 0.000* 0.014* 0.028*

Odds ratio (95% CI) 1.10 (1.06–1.15) 9.47 (8.80–10.20) 1.07 (0.94–1.21)

– 1.202

– 0.001*

1.00 3.33 (1.61–6.87)

Vessel work assignment is still a useful predictor for predicting high severity of symptoms of upper back MSD. Stevedores who are working inside the vessel hatch (as compared to those who are working on-dock) have 3.33 times higher odds of experiencing high severity of work-related musculoskeletal symptoms. Based on the results, there are no psychosocial factors identified as significant risk factors for the symptoms of upper back MSD.

Prevalence, Severity, and Risk Factors

3.7

97

Risk Factors of Work-Related Musculoskeletal Disorders (Lower Back)

The results of the ordinal logistic regression model for the respondents’ lower back region is quite similar with the upper back region. Vessel work assignment is still associated with high severity of symptoms of upper back MSD. Stevedores who are working inside the vessel hatch (as compared to those who are working on-dock) have 5.15 times higher odds of experiencing high severity of WMSD symptoms. Age is a significant risk factor for lower back MSD. The model specifies that there is a 12% increase in the probability of experiencing high severity for an additional year in a stevedore’s age. Moreover, weekly consumption of alcoholic drinks and cigarette smoking are also significant predictors for work-related musculoskeletal symptoms (Table 6). Table 6. Ordinal logistic regression model for risk factors associated with the severity of musculoskeletal symptoms in the lower back region (only significant factors are shown below) Factors Age Cigarette consumption Alcoholic drinks consumption Vessel work assignment Quayside (Ref.) Vessel hatch

Coefficient 0.116 0.077 0.040

P value 0.000* 0.046* 0.003*

Odds ratio (95% CI) 1.12 (1.07–1.18) 9.26 (8.60–10.00) 1.04 (0.92–1.18)

– 1.640

– 0.000*

1.00 5.15 (2.41–11.02)

4 Discussion Table 7 shows the summary of significant risk factors of work-related musculoskeletal disorders per body region. The occurrences of work-related musculoskeletal symptoms in the upper and lower back, shoulder, and forearm region can be associated with the nature of manual handling activities performed by the stevedores during vessel loading and unloading operations. These activities include securing and manual adjustment of slings to lifting gears, reaching and/or hoisting for proper positioning of cargoes upon landing, and releasing of cargo (hooking/unhooking) from the lifting gears. Table 7. Summary of significant risk factors of work-related musculoskeletal disorders Significant factors Age Yrs. in break-bulk port Cigarette smoking Alcoholic drinks consumption Average days/week Vessel work assignment Steel commodity Job satisfaction Job security

Shoulder Forearm Upper back Lower back U U U U U U U U U U U U U U U U U U U U U U

98

K. Sedilla and A. Matias

Age is one of the major risk factors for prevalent regional pain [23]. It is a significant predictor of work-related musculoskeletal symptoms for the body regions with high prevalence rates. Stevedores who reported drinking alcoholic drinks were more likely to experience symptoms of work-related musculoskeletal disorders. Possible explanation for this relationship is difficult to surmise as few, if any; other studies have looked at alcohol as predictor of musculoskeletal disorders. However, the relationship with cigarette smoking was more specific, with it being a strong predictor for upper and lower back region. Tobacco is as a risk factor in several studies, suggesting a probable biological relationship with musculoskeletal disorders [24]. Work experience (expressed in yrs.) contributes to the occurrence of work-related musculoskeletal disorders for the shoulder region. The results showed that an increase in work experience of stevedores, leads to an increase in probability of experiencing high severity of shoulder pain as compared to those with less experience working in break-bulk port. This finding can be associated to the nature of tasks performed by the stevedores on a regular basis, which require overhead reaching and hoisting of heavy lifting gears. The identification of vessel work assignment as a significant risk factor for all the body regions is an interesting finding of the study. The stevedores who are working inside the vessel hatch (as compared to quayside/on-dock) had higher odds of showing high severity of musculoskeletal symptoms. This result may be associated with the stevedores’ exposure to postural load, forceful exertions and handlings, non-neutral or awkward postures (e.g. bending, twisting) while working inside the vessel hatch. In addition, mobility inside the vessel hatch is more restricted than quayside because of the various cargoes stowed inside. In addition, the stevedores on-board are also in-charge of assessing the conditions of all steel cargoes on board (stowed inside the vessel hatch) before the start of discharging/loading operations. Moreover, the type of steel commodity specifically, steel billets appeared to be a predictor of musculoskeletal symptoms in the shoulder region. Handling steel billets is more physically demanding compared to other steel commodities because the required lifting gears for steel billets are chain slings with hook, wire sling, and or belt slings, which are heavier than other lifting gears. In addition, hooking and unhooking of these gears entails forceful exertion that may contribute to the development of work-related musculoskeletal disorders. The number of working days per week is a significant risk factor for work-related musculoskeletal symptoms in the shoulder and forearm region. The results showed that for a unit increase (expressed in days) in the average number of working days per week, there is a 3.22 and 3.23 higher odds of experiencing high severity of shoulder and forearm pain respectively. Psychosocial factors appear to play a role in the occurrences of musculoskeletal symptoms in the shoulder and forearm region. This is consistent with the emerging evidence from various literatures. In this study, the most effective predictors in most of the body parts are job dissatisfaction and perceived job security. The results showed that stevedores who were dissatisfied in their job had higher odds of showing high severity of work-related musculoskeletal symptoms in the shoulder region. Likewise, those who felt that they have no assurance of keeping their job had higher odds of having high severity level of symptoms of MSD. This finding can be associated with the respondents’ type of employment in the port terminal. All of them are under the company’s stevedoring agencies and employed on a contractual basis. Majority of the

Prevalence, Severity, and Risk Factors

99

stevedores mentioned that they are always worried about the future of their job employment in the port terminal. In addition, most of them are anxious in losing their job because they are the breadwinners of their family. Low job satisfaction can lead to poor performance at work, eventually leading to occurrence of musculoskeletal disorders [25]. In several studies, psychosocial factors, such as low job satisfaction, high job strain, and low level of perceived job security, have been reported to increase the occurrence of work-related MSDs.

5 Conclusion In conclusion, the results of this study showed that the prevalence of work-related musculoskeletal disorders among stevedores is significantly higher in the upper extremities. It was determined that age, work experience in break-bulk port terminal, cigarette smoking, alcoholic drinks consumption, vessel work assignment, type of steel commodity assigned to, average number of days of work per week, level of job satisfaction, and perceived job security are the significant predictors amongst the risk factors considered in the study. The variety of these factors suggests that the etiology of this condition is both complex and multifaceted. The magnitude of the reported period prevalence and severity level of work-related musculoskeletal disorders justifies the application of further research to determine other possiblerisk factors associated with the occurrences of MSDs among port workers. This study contributes to the existing body of literatures that in one way or another provide bases in the formulation of policies and intervention programs in port industry - particularly on stevedoring operations in a break-bulk port terminal. Acknowledgement. The authors would like to acknowledge the funding support from the Engineering Research and Development for Technology (ERDT), Department of Science and Technology (DOST), Philippines. Special thanks to Jairus Gogorza, Bryan Gayagoy, Kenneth Soldevilla, James Lomeda, Bannie Alcantara, and Michelle Estrada for their significant contribution in data collection.

References 1. Grozdanović, M.: Human activity and musculoskeletal injuries and disorders. Facta Univ. Ser. Med. Biol. 9(2), 150–156 (2002) 2. Occupational Health and Safety Council of Ontario: MSD prevention toolbox: getting started. Musculoskelet. Disord. Prevent. Ser. 3, 1–58 (2008) 3. Silverstein, B., Stetson, D.S., Keyserling, W.M., Fine, L.J.: Work-related musculoskeletal disorders: comparison of data sources for surveillance. Am. J. Ind. Med. 31(5), 600–608 (1997) 4. Buckle, P.W., Devereux, J.J.: The nature of work-related neck and upper limb musculoskeletal disorders. Appl. Ergon. 33(3), 207–217 (2002) 5. Punnett, L., Wegman, D.H.: Work-related musculoskeletal disorders: the epidemiologic evidence and the debate. J. Electromyogr. Kinesiol. 14(1), 1–23 (2004)

100

K. Sedilla and A. Matias

6. Chapman, L., Meyers, J.: Ergonomics and musculoskeletal injuries in agriculture: recognizing and preventing the industry’s most widespread health and safety problem. In: Agricultural Safety and Health Conference Proceedings (2001) 7. Singh, S., Arora, R.: Ergonomic intervention for preventing musculoskeletal disorders among farm women. J. Agri. Sci. 1(2), 61–71 (2010) 8. Buckle, P.: Ergonomics and musculoskeletal disorders: overview. Occup. Med. 55(3), 164– 167 (2005) 9. Cabeças, J.M.: Occupational musculoskeletal disorders in europe: impact, risk factors and preventive regulations. Enterp. Work Innov. Stud. 2, 95–104 (2006) 10. David, G.C.: Ergonomic methods for assessing exposure to risk factors for work-related musculoskeletal disorders. Occup. Med. 55(3), 190–199 (2005) 11. Marras, W., Cutlip, R., Burt, S., Waters, T.: National occupational research agenda (NORA) future directions in occupational musculoskeletal disorder health research. Appl. Ergon. 40, 15–22 (2009) 12. Waters, T.R.: National efforts to identify research issues related to prevention of work-related musculoskeletal disorders. J. Electromyogr. Kinesiol. 14(1), 7–12 (2004) 13. Cezar-Vaz, M.R., Capa Verde de Almeida, M., Bonow, C.A., Rocha, L.P., Borges, A.M., Piexak, D.R.: Casual dock work: profile of diseases, injuries, and perception of influence on health. Int. J. Environ. Res. Public Health 11(2), 2077–2091 (2014) 14. Roa, I., Peña, Y., Amante, B., Goretti, M.: Ports: definition and study of types. Sizes Bus. Models 6(4), 1055–1064 (2013) 15. Health and Safety Executive: A Quick Guide to Health and Safety in Ports (2011) 16. Bovenzi, M., Pinto, I., Stacchini, N.: Low back pain in port machinery operators. J. Sound Vib. 253, 3–20 (2002) 17. Edwin, M., Ploychronopoulos, L.: Musculoskeletal discomfort in crane and forklift operators in a New Zealand port (2010) 18. Hassan, M.A., Mburu, C.: Prevalence of occupational low back pain among drivers of terminal tractors in the port of. Mombasa, Kenya. Int. J. Cur. Trends Res. 2, 358–364 (2013) 19. Cavalcante, F., Gomes, A., Nogueira, R., De Farias, J., Pinheiro, J., De Albuquerque, E., Farias, A., Cabral, G., Magalhaes, F., Gomide, M.: Occupational risks among dock workers in the port of Mucuripe, Fortaleza, Brazil. Ciencia Saude Coletiva. 10, 101–110 (2005) 20. Waskiewicz, J.: The effect of heavy work on the musculoskeletal system of dockers. Bull. Inst. Marit. Trop. Med. Gydnia 47, 25–32 (1996) 21. De Almeida, M., Cezar-Vaz, M., Soares, J., Da Silva, M.: The Prevalence of Work-Related Musculoskeletal Disorders among Casual Dockworkers (2012) 22. Philippine Ports Authority (PPA): PPA Annual Report (2015) 23. Palmer, K.T., Goodson, N.: Ageing, musculoskeletal health and work. Best Pract. Res. Clin. Rheumatol. 29(3), 391–404 (2015) 24. Smith, D.R., Mihashi, M., Adachi, Y., Koga, H., Ishitake, T.: A detailed analysis of musculoskeletal disorder risk factors among Japanese nurses. J. Saf. Res. 37, 195–200 (2006) 25. Nevala-Puranen, N.: Effects of occupationally-oriented rehabilitation on farmers’ work techniques, musculoskeletal symptoms, and work ability. J. Occup. Rehabil. 6(3), 191–200 (1996)

Smart Service World

Applying Value Proposition Design for Developing Smart Service Business Models in Manufacturing Firms Jens Neuhüttler ✉ , Inka C. Woyke, and Walter Ganz (

)

Competence Team Service Management, Fraunhofer Institute for Industrial Engineering IAO, Nobelstr. 12, 70569 Stuttgart, Germany {Jens.Neuhuettler,Inka.Woyke,Walter.Ganz}@iao.fraunhofer.de

Abstract. The digital transformation of economies is one of the megatrends of today’s world. The advancing equipment of machines and products with sensors and connectivity as well as their intelligent interconnection among each other leads to a mergence between physical and virtual worlds. This development is not limited to smart production systems in terms of an “Industry 4.0” only, but will also affect the increasingly important service business of manufacturing companies. Regarding “Smart Services” (data-based bundles of services, digital applications and technologies), the development of appropriate business models is seen as a key challenge for companies. Our paper illustrates how the approach “Value Proposition Design” can support manufacturing firms to develop Smart Service Business Models. Thus, we present insights from an applied research project, in which two different Business Model alternatives have been derived from applying Value Proposition Design. Keywords: Smart services · Business model design · Service 4.0 · Value proposition design · Condition monitoring

1

Introduction: Smart Services in Manufacturing Firms and the Need for New Business Models

During the last decades, an increasing number of manufacturing firms started shifting their focus from mainly producing and selling capital goods towards providing services and solutions, i.e. [1]. Due to the manifold potentials of the service business, this devel‐ opment - often referred to as “Servitization” - can be observed for companies across many industries and countries worldwide [2]. The intangible and thus “hard-to-imitate” character of services, for example, provides a sustainable source for differentiation in times of growing commoditization and shrinking margins in the core business of capital goods [3]. Offering a wide range of value-added services also enables manufacturing firms to provide customer-centric, individualized solutions, leading to an increasing customer satisfaction and consequently to a higher customer retention [4]. Moreover, providing product-related services creates a continuous revenue stream around the life‐ cycle phases of capital goods, which becomes even more important in rather precarious economic times when investments in new machinery drop. As the service business becomes more and more important, manufacturing companies strive for exploiting © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_10

104

J. Neuhüttler et al.

opportunities for increasing service quality and productivity accordingly. Regarding these aspects, the ongoing digital transformation of manufacturing and the associated prevalence of technology-driven concepts, such as the “Internet of Things” or an “Industry 4.0”, hold substantial potentials in form of so-called “Smart Services” [5]. The term “smart services” describes data-based, individually configurable bundles consisting of physical and electronic services as well as physical elements and technol‐ ogies, which are usually performed and organized on integrated service platforms [6, 7]. The equipment of products and machinery, with information and communication technology, such as sensors, actuators and web-enabling modules, leads to a growing amount of continuously transferable data. By capturing, storing, analysing, combining and visualizing machine data, manufacturing firms can transform their services into customer-oriented solutions that lead to added value for both, providers and customers [8]. One central component of Smart Services are integrated service platforms, on which unstructured data is turned into structured data and physically delivered services, digital services and physical elements are combined to solutions. Based on contextualized customer information, the combination is conducted rather customer- than provideroriented. In advanced Smart Service Systems, solution elements and resources are combined even across company boarders. Therefore, significant changes of operational processes, business models and even whole value creation systems are required. However, the disruptive potential of smart services highly depends on the digital maturity level of branches as well as characteristics of targeted solutions. Figure 1 illustrates the different development levels of Smart Services in manufacturing firms.

Fig. 1. Development levels of smart services and the need for developing new business models

The first level “Web-based services” represents rather rudimentary versions of Smart Services that do not necessarily fit all elements of above stated definition. However, it shows a first step towards digitalized service offers and represents the current develop‐ ment stage for many manufacturing firms. Web-based service comprise digital offers,

Applying Value Proposition Design

105

such as providing an online shop for ordering spare parts or a web portal to file electronic documents (e.g. service reports, contracts or manuals). The second Smart Service development level can be described as “Stand-alone Smart Services”. Companies start to offer first services that are based on machine data collected via different sensors and transmitted to a back-end system. A widely known example is Condition Monitoring. Based on the continuous collection and analysis of machine data for conspicuous patterns, manufacturers can identify potential failures in advance and thus prevent breakdowns of their plants and machines. On the development level of “Smart Service Platforms”, manufacturing firms offer a wide range of Smart Services via their company platform. The first stand-alone services (e.g. condition monitoring) are enhanced with a portfolio of other data-based, valueadded applications and services, such as an integrated performance cockpit for the whole machine park, predictive maintenance solutions or a fleet management system). Besides offers for customers, the platform can be used to provide internal services for employees, such as service technicians, too. One example for an internal smart service is seen in the combination of machine data of customers, with routes information and the equipment of service vans in order to optimize the coordination of service operation fleets. One of the main challenges of smart services providers is to finding appropriate business models, which monetarize the value provided by analyzing, visualizing and transferring data into new service concepts [9]. Thus, currently applied service business models need to be adapted to the new value proposition, value creation processes and resources needed for providing Smart Services. The complexity and need for re-development grows with the development level stated in Fig. 1. However, the concept of Smart Services comprises more than a digital improvement of existing services. The collection and combination of data from different sources (e.g. devices, sensors or machines) holds vast potentials for developing completely new service ecosystems and business models. The fourth development level therefore describes the opening of the formerly company-specific Smart Service Platform to offers of other providers and ecosystem stakeholders. These “Smart Service-Industry Plat‐ forms” take the role of fully automated market places, where products, services, produc‐ tion capacities or data is traded between suppliers and consumers across company borders [6]. Occupying customer interfaces of other providers and orchestrating their solution elements outlines activities far away from the core business of manufacturing firms and thus requires the development of a separate, disruptive business model (e.g. two-sided market models [10]). The following sections present a relatively new approach for developing new business models and thus support manufacturing in coping with these challenges.

2

Developing Business Models Using Value Proposition Design

Simplified, a business model describes the value companies provide to their customers and partners (‘Value proposition’), how this value is generated by combining different resources and activities (‘value creation’) and how provided value can be monetarized by the company in return (‘value capture’) [10, 11]. For developing sustainably

106

J. Neuhüttler et al.

successful business models, it is important to start with creating a value proposition to customers that describes how the products and services of a company help customers to get their jobs done better than other alternatives [12]. Despite the obvious logic of this first step, the realization can be rather challenging for manufacturing companies. Although, for example, the advantages of condition monitoring are beyond question (e.g. increased machine availability due to preventing unplanned breakdowns, avoid‐ ance of material and personal damages as well as a reduction of maintenance costs and time), a lot of manufacturing companies struggle with designing customer-specific solu‐ tions and sustainably successful business models [9]. An approach that supports compa‐ nies in developing value propositions systematically and has become increasingly important during the last years is Value Proposition Design by [13]. In a simplified version, the method consists of three consecutive steps, which are illustrated in Fig. 2: (1) Mapping the Customer Profile: In a first step, customer jobs are analyzed. The underlying theory, the “Jobs-to-be-done” concept by [12], assumes that customers are hiring products and services in order to fulfill different jobs during their work or leisure time. Besides their functional dimension, customers jobs also possess social and emotional dimensions, which explains why customers prefer some solu‐ tions more or less (e.g. because it makes them look good or addresses basic needs, such as security, better). Moreover, it is important to consider the context (e.g. personal or professional situation), in which a job is performed. In the second step, central pains (describing problems, obstacles and negative outcomes) as well as gains (describing desired outcomes, needs and preferred aspects) of managing jobs are analyzed and derived. (2) Mapping the Value: Firstly, all technological, physically delivered services and digital applications, which a company plans to offer to their customers in order to support them in fulfilling their functional, emotional or social jobs, have to be listed. Secondly, pain relievers and gain creators are stated, describing how the listed solution elements address the pains and gains of the customers (e.g. by taking away responsibility or a particular unpleasant activity from the customer job [14]). (3) Checking the fit between Customer Profile and Value Map: A good value propo‐ sition is characterized by addressing all essential pains and gains of customers better than alternatives, without containing elements that do not generate value. Therefore, it is important to prioritize jobs, pains and gains and add respectively eliminate redundant elements from the value map during the value proposition design. Developing a value proposition based on the functional, emotional and social Jobs of customers allows companies to gain insights about the decision-making logic of customers and thus helps to focus on product and service elements that really matter to customers. Moreover, other business model elements have to be designed accordingly and contribute to providing the value proposition, starting from communication channels or chosen partners and ending in the profit formula and paying options. In this matter, the approach proved to be superior to classic market segmentation by socio-demographic characteristics, since customer jobs could vary heavily among these groups.

Applying Value Proposition Design

107

Fig. 2. Generic structure of value proposition design [13]

In the next section, we illustrate how stated approach for designing value proposi‐ tions can be applied to develop business models for Smart Services using the example of Condition Monitoring.

3

Use Case: Applying Value Proposition Design for Developing a Business Model for Condition Monitoring Services

3.1 Introduction to the Use Case The insights presented in this section were gained during a joint project between Fraunhofer IAO and a worldwide leading manufacturing company, which is producing customized suction and extraction machines for production plants of their industrial customers. Their machines are specialized in capturing and separating a variety of pollutants (e.g. industrial dusts, aerosols or oils) and reducing emissions in industrial applications. Thus, the main value proposition is to keep air in manufacturing environ‐ ments clean and to support customers meeting the needs of environmental and safety regulations. Customers often perceive suction and extraction machines as supporting machine, since they are not producing parts themselves. Consequently, they are often neglected and expected to run smoothly and without a lot of maintenance activities. However, in case of a failure or breakdown of supporting machines, the primarily focused production machines must be switched off immediately in order to prevent consequential damages of the machines or shop floor employees. Based on their high importance for the production system and the expectation towards a low-maintenance machine, the development of a condition monitoring system seemed to be highly attrac‐ tive for supporting suction and extraction machines. However, questions arose regarding the specification, communication and monetarization of the expected value. Moreover, customer requirements about technical, personal and usability aspects of the smart service had not been understood sufficiently. Thus, a joint project was set up to answer

108

J. Neuhüttler et al.

these questions by applying Value Proposition Design and to develop an appropriate business model around it. In the following paragraphs, we present exemplary insights and results from the project. 3.2 Analyzing Customer Jobs, Pains and Gains In order to build a basis for the Value Proposition Design approach, multiple qualitative in-depth interviews were conducted among key business-to-business customers of the manufacturing firm. As to the interviewees, we selected persons responsible for oper‐ ating and maintaining machinery in their companies. The main objective of these inter‐ views was to analyse central jobs, pains and gains of customers as well as to identify key requirements to condition monitoring services. Moreover, the interviews aimed for generating ideas for future data-based service offers, such as predictive maintenance. The interviews followed the approach of a “Means-End-Analysis” in order to gain deeper insights about why customers are doing jobs in certain ways and what their underlying motives are. Based on the results of the interviews, customer profiles were developed according to the structure of Fig. 2. Being aware of the high importance of employee satisfaction on the external service value and thus customer satisfaction (c.f. the Service-Profit Chain by [15]), similar interviews have been conducted with service staff from management (e.g. service coordinators) as well as operational level (e.g. service technicians). Due to limited space, our paper only exemplifies results from customer interviews. One of the most interesting results regarding the customer profiling was that there are two types of customers, which have to be distinguished for developing value prop‐ osition and business models for condition monitoring or other smart services: “Serviceoriented” and “Data-oriented” customer types. These two types do not represent distinct companies observed during the project and should be understood as idealized forms. However, for most surveyed companies, it was rather easy to allocate them to either one of the two types. The main difference between the two customer types is that, although they pursue the same main job (assuring machine availability), they follow different approaches in doing so and have to fulfil their goals in diverging job contexts. To meet the specific needs, pains and gains of each customer type, two distinct value propositions and business model alternatives had to be developed during the project. In the following sections, we present our insights divided into the two types, to illustrate the different approaches of customers. The Service-oriented customer type is characterized by a rather low experience level of collecting and analysing machine data. There is no company-wide strategy for making use of data from different production plants and machines or for improving the production system systematically based on extracted information. Thus, using data provided by the condition monitoring system of suction and extraction plants does not provide a distinct value to these customers. The main job of their maintenance depart‐ ment is to guarantee a high degree of production machine availability and to prevent failures of supporting machines accordingly. Usually service-oriented customers work with a small maintenance team possessing high expertise regarding the production machines but fewer skills in respect to maintaining neglected supporting machines.

Applying Value Proposition Design

109

Despite the rather small size of the maintenance team, reliably operating machines are a key requirement and are of high importance to these companies. Thus, they involve external service providers or the service technicians of the machine manufacturers on a regular basis to reach the internal goal of high machine availability. Consequently, data will be deployed for receiving a higher service quality of external providers or for opti‐ mizing internal service processes. Among the main pain points of service-oriented customers that were stated by the interviewees are: (1) a higher complexity for employees due to the amount of data; (2) non-transparent usage of data; (3) bad usability of the system and: (4) institutional barriers for introducing data-based service solutions. In this context, two companies described that they would face problems to convince the works council as well as the top management, if there wasn’t any support for issuing data security concepts offered by the condition monitoring provider. Two identified key gains of a condition monitoring system were: (1) a faster reaction of service providers in case of suspicious data patterns or pronouncing problems as well as (2) a simplified process for locating potential errors and failures by internal service staff. Moreover, documentation of machine condition, automated reaction of service provider in case of problems, as well as safety in the sense of a “4-eyes principle” were mentioned by the interviewees. In contrast, the data-oriented customer type already possesses rich experiences with condition monitoring systems for other machine types. Moreover, the company already forged short- or mid-term plans for interconnecting production machines, collecting data and using it for managing and controlling the production plant in the sense of a smart production system (c.f. industry 4.0). Furthermore, data-oriented customers consider machine data as a strategic resource for their business and therefore have a high sensibility for topics, such as data security or system control access by external partners. Usually, the internal maintenance team has sufficient size in terms of employees, resources and skills for carrying out their main job: guaranteeing a high degree of machine and plant availability without hiring external service providers daily. Only in special or emergency situations, such as an unexplainable machine breakdown or unidentifiable failure, would they fall back on third-party service providers or the OEM service technicians. As main pain points, (1) a misuse of data, (2) loss of data sovereignty and (3) security flaws in a condition monitoring system, which could lead to cyberattacks addressing the entire smart production system, were mentioned during our study. Moreover, a flow of machine data across the company’s system boundaries was regarded as a massive risk and was thus rejected. A further potential risk regarding the condition monitoring service was seen in providing slow system support or in missing a helpline request to react in the case of a failure of the condition monitoring system itself. Data-oriented customers expect gain regarding the usage of a condition monitoring system for machines of different manufacturers and the manifold potentials of collected data for their own operations. As the most important potentials, the use of data for an integrated production system management, for analysing interdependencies between different production and supporting machines or for planning internal service operations, were named by the interviewees.

110

J. Neuhüttler et al.

3.3 Designing Value Propositions and Business Models for Customer Profiles Based on the idealized customer profiles, two different value propositions and business model variants were derived accordingly. For a structured illustration, we used the Business Model Canvas framework of [16]. The value proposition and Business Model elements for service-oriented Customers were characterized as follows: (1) Value proposition: Service-oriented customers are offered a holistic single-source condition monitoring solution wherein the service provider will carry out most relevant tasks and activities. This includes installation, pre-configuration and main‐ tenance of the sensor-box for collecting data during the implementation. A simple, pre-configured sensor box provides additional value, since customers don’t intend to collect additional data or combine and analyse them by themselves. The technical solution should be resilient rather than complex and also customizable. Technical infrastructure, such as connectivity, data hosting and storage, needs to be part of the solution too. By recording and analysing data constantly, the service provider can address the strong need for security and reduce risks for machine operator and internal maintenance employees. The internal maintenance team can focus their core tasks, such as monitoring and controlling production machines. Since serviceoriented customers do not possess a centralized control station that integrates data from different machines, data must be visualized in a simple and decentralized way, e.g. on displays attached to the suction and extraction machines. In addition to displays, a browser-based application is provided for insights about data. Moreover, a guide for undertaking emergency measures is illustrated on the display, which supports employees to react and localize problem sources quickly and efficiently. The main value proposition for service-oriented customers is that external service providers are enabled to react quicker in cases of suspicious data patterns and to plan service operations more efficiently and customer-oriented. The main job – increasing machine availability and ensuring the smooth support of production machines – is supported by condition monitoring. When limit values are exceeded, the service provider will contact the customer actively and discuss next steps, such as providing basic remote services or guiding customers to carry out emergency measures themselves. In order to do so, the provider needs full access to real-time data and machine system control, which requires a joint definition of data security and establishing permission concepts with customers. Moreover, the provider needs to support the service-oriented customer with added-value services, such as nego‐ tiating with work councils or union representatives or providing a line of argumen‐ tation and a solid concept against data misuse. In periods when no limit values are exceeded, a short report with service recommendations will be sent at the period, since employees won’t analyse data proactively. (2) Customer relationships and channels: By establishing a condition monitoring system, the relationship between the manufacturing companies and their serviceoriented customers is intensified substantially. On the one hand, there is a new digital interface between the provider, the customer and their suction and extraction machines, due to the consistent data stream. Thus, the service unit of the

Applying Value Proposition Design

111

manufacturing company gains substantial competitive advantages towards third party service providers. On the other hand, the relationship turns into a pro-active and customizable service provision based on machine data. Although the number of routine service operations (e.g. machine audits) might be reduced, long-term customer relationships can be established and intensified based on the exclusive data access and value estimation of the machine. Existing customer channels, such as service hotline, will not be dropped but complemented by digital channels, which help to increase productivity of the service provider too. (3) Key activities, resources and partners: One of the core activities is the collection and interpretation of data via intelligent algorithms. Due to the pool of data from different customers’ machines, an extensive knowledge base will be set up. Knowl‐ edge pooling and editing data for different applications (e.g. for the construction department or for planning service operations efficiently) is one of the key tasks. This also applies for defining service modules as reaction to occurring data patterns, which will be recommended and priced immediately. The second key activity is to assure data security. Although service-oriented customers do not have a problem with sharing data with external partners, they are still sensitive when it comes to a misuse of production-related data. This implies that all software systems need to be updated consistently to prevent cyberattacks. Key resources are data but also employees that are essential for introducing and providing smart services, such as data analysts for developing algorithms or service employees that can explain the solution in detail and convince customers to participate. Moreover, service techni‐ cians need to be trained to work with the condition monitoring system in their daily work routine. Key partners are needed for providing basic ICT infrastructure, such as connectivity, data hosting and cloud services. Moreover, a sensor-box provider is needed for supporting the standard pre-configuration of sensor-box and back end. In order to develop a sufficient standard data security concept, external support of a law firm is required. (4) Cost structure and revenue streams: Increasing costs can be linked to implemen‐ tation of additional hardware into the suction and extraction machines as well as to qualification measures of service technicians and back office employees. Moreover, the operation and support of the condition monitoring software system will lead to additional costs. Conversely, service operation costs are expected to sink, due a more efficient coordination and preparation of service technicians (e.g. by bringing specifically required spare parts or tools). Additional service fees of the condition monitoring service system generate main revenue streams. Market entry barriers have to be kept to a minimum level, especially during market introduction. There‐ fore, sales prices for the technology basis (e.g. sensor-box or displays for visualizing data) should be kept low. One possibility to reduce perceived investment risks on the customer side is the application of a “prepaid”-business model pattern [10]. Customers can buy service credits instead of paying periodic service fees and thus cap perceived risks of unexpected costs caused by condition monitoring services. Credits are only booked, if a service operation or a remote service took place. In case none of the digital or data-based service offers was utilized, customers do not have to do anything. From a service provider’s view, this seems to be a negative

112

J. Neuhüttler et al.

consequence, but the budgeted credits will be used for the company’s service even‐ tually. Moreover, the collected information can be used to evaluate the specific value of a machine and service operation for the customers. Consequently, the willingness to pay of customers can be skimmed individually, leading to higher incomes on the provider and increasing satisfaction on the customer side. Beyond that, future business models, such as BOT (build-operate-transfer) models, can be evaluated based on machine lifecycle data. Again, in contrast, the value proposition and business model elements for dataoriented Customers were characterized during the project as follows: (1) Value proposition: The main value proposition of condition monitoring for dataoriented customers is the continuous and reliable provision of machine data for the customer’s smart production system. Since requirements of data usage and existing infrastructures can vary heavily between customers, used technology needs to be customizable and interfaces (e.g. to a central control stand) need to be adaptable. Configuring and adapting the standard solution to the requirements of customers is part of the service offering. Guaranteeing a high availability of the condition moni‐ toring system as well as a quick support reactions in case of failures or breakdowns are expected by customers in order to prevent an interruption of a highly networked production system. Since data security is a highly sensitive aspect for data-oriented customers, data needs to be stored locally on customers’ servers. Therefore, aspects such as data hosting or connectivity are not included in the offered service bundle, while providing safety-relevant software updates and a contact person for defining data security is a key requirement of data-oriented customers. Although the machine data won’t be transferred, service operations of the external provider are still expected to make use of data. In order to enable service technicians to provide databased services, alternative concepts for data access have to be defined. Alternatives could range from receiving printouts from the customer’s smart production control system on site to transferring the aggregated and history-based data to a mobile device of the service technician. This ad hoc confrontation with new data leads to high requirements regarding the “smart skills” of service technicians, who need to be able to analyze and interpret data in different centralized control systems. However, the main value proposition refers to service improvements of the internal maintenance team and a holistic optimization of the production system. (2) Customer relationships and channels: The customer relationship stays rather reac‐ tive, since the service unit of the manufacturer will be contacted only in distinct situations, e.g. mandatory safety checks or machine failures. The existing relation‐ ship and the high experience of maintaining the suction and extraction machines are expected to remain the main driver for customer retention. Regarding customer channels, only marginal changes will take place. Main touchpoints between manu‐ facturer and customer are personal contact during service operations and via the hotline. Digital touchpoints are restricted to the exchange of historic and static data in case of emergencies. (3) Key activities, resources and partners: Among the key activities are all tasks that are directed at guaranteeing a continuous and safe data stream for the customer.

Applying Value Proposition Design

113

This includes customer support regarding the sensor-box, the local back end and interfaces to the customers’ central control stand. During the implementation phase, this could also comprise trainings and adaption activities. The interpretation of data during service operations is a new activity for the provider’s service technicians. Thus, they need to be able to react to data patterns visualized in different ways flexibly. Reference data from service-oriented customers could support them and thus represent a key resource for providing service to data-oriented customers too. It is essential that service technicians have access to these reference data cases and patterns during their service operations. Thus, mobile devices and especially skilled service technicians are further key resources. (4) Cost structure and revenue streams: Expenditures for the technological basis and its integration into the smart production system (e.g. adaption of interfaces to the central control centre) are the main cost drivers of condition monitoring. Moreover, qualifying service technicians and the maintaining the condition monitoring system generates additional costs. Since additional services offered won’t be demanded, revenues need to be generated by selling the sensor equipment and back end systems to customers. Moreover, the manufacturer could try to establish a fee for generated data. The risk of this option lies in the inducement for customers to manipulate the data flow into the central system themselves. In order to prevent discovery, they might choose a third-party service provider instead of the OEM service unit.

4

Conclusion and Implications

Our paper presented Value Proposition Design as an appropriate approach for devel‐ oping Smart Service Business Models in Manufacturing firms using the example of Condition Monitoring. The approach is based on the “Jobs-to-be-done” concept and thus focuses not only functional, social and emotional jobs, but also the job context. Thus, new insights about customer requirements for business models can be derived and help to develop sustainably successful business models. Our Case study has shown that customers with the same functional job (machine availability), but varying contexts need distinct business model alternatives: For service-oriented customers, all business model elements must be focused on providing a higher service quality and productivity for the customer. For data-oriented customers, an undisturbed data flow and security measures are the key value proposition and thus need to be centered by the business model. The distinction of these two customer types is not only relevant for condition monitoring services but also for more advanced solutions (e.g. predictive maintenance or the devel‐ opment of Smart Service Platforms).

References 1. Oliva, R., Kallenberg, R.: Managing the transition from products to services. Int. J. Serv. Ind. Manag. 14(2), 1–10 (2003) 2. Neely, A.: Servitization in Germany: An International Comparison. Working Paper of the Cambridge Service Alliance, Cambridge (2013)

114

J. Neuhüttler et al.

3. Baines, T.S., Lightfoot, H.: Servitization of the manufacturing firm: exploring the operations, practices and technologies that deliver advanced services. Int. J. Oper. Prod. Manag. 34(1), 2–35 (2013) 4. Reinartz, W., Ulaga, W.: How to sell service more profitably. Harvard Bus. Rev. 86(5), 90– 98 (2008) 5. Herterich, M.M., Uebernickel, F., Brenner, W.: Industrielle Dienstleistungen 4.0. Springer, Wiesbaden (2016) 6. Smart Service Welt Working Group, acatech (eds.): Smart Service Welt – Recommendations for the Strategic Initiative Web-based Services for Businesses, Berlin (2015) 7. Bullinger, H.-J., Ganz, W., Neuhüttler, J.: Smart Services – Chancen und Herausforderungen digitalisierter Dienstleistungssysteme für Unternehmen. In: Bruhn, M., Hadwich, K. (eds.) Forum Dienstleistungsmanagement: Dienstleistungen 4.0 (2017, forthcoming) 8. Neuhüttler, J., Ganz, W., Liu, J.: An integrated approach for measuring and managing quality of smart senior care services. In: Ahram, T.Z., Karkowski, W. (eds.) Advances in the Human Side of Service Engineering, pp. 309–318 (2016) 9. Wünderlich, N., et al.: “Futurizing” smart service - implications for service researchers and managers. J. Serv. Mark. 29(6), 442–447 (2015) 10. Gassmann, O., Frankenberger, K., Csik, M.: The Business Model Navigator. Pearson Education Limited, Harlow (2014) 11. Johnson, M.W., Christensen, C.M., Kagermann, H.: Reinventing your business model. Harvard Bus. Rev. 86(12), 51–59 (2008) 12. Christensen, C.M., Cook, S., Hall, T.: Marketing malpractice – the cause and the cure. Harvard Bus. Rev. 83(12), 74–86 (2005) 13. Osterwalder, A., Pigneur, A., Bernada, G., Smith, A.: Value Proposition Design – How to Create Products and Services Customer Want. Wiley, Upper Saddle River (2014) 14. Bettencourt, L.A., Ulwick, A.W.: The customer-centered innovation map. Harvard Bus. Rev. 86(5), 109–114 (2008) 15. Heskett, J.L., Sasser, E.W., Schlesinger, L.A.: The Service Profit Chain. The Free Press, New York (1997) 16. Osterwalder, A., Pigneur, Y.: Business Model Generation: A Handbook for Visionaries, Game Changers, and Challengers. Wiley, Hoboken (2010)

Wisdom Service Systems: Harmonious Interactions Between People and Machine Md. Abul Kalam Siddike1 ✉ , Kazuo Iwano2, Kazuyoshi Hidaka3, Youji Kohda1, and Jim Spohrer4 (

1

)

Graduate School of Knowledge Science, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi, Ishikawa 923-1292, Japan [email protected], [email protected] 2 Japan Science and Technology Agency, Science Plaza, 5-3, Yonbancho, Chiyoda-ku, Tokyo 102-8666, Japan [email protected] 3 School of Environment and Society, Tokyo Institute of Technology, 3-3-6, Minato-ku, Tokyo, 108-0023, Japan [email protected] 4 IBM Almaden Research Center, San José, CA 95120, USA [email protected]

Abstract. Wisdom is the highest-level decision-making performance, acquired by gaining experience that can be applied in complex, conflict situations as well as interacting with other wise people. We conceptualize wisdom service systems as socio-technical systems in which people use cognitive mediators (smart machines that know people in some ways better than they know themselves and provide a type of weak immortality for people) to augment interaction and deci‐ sion-making capabilities in a kind of Reality 2.0. Someday cognitive mediators may provide scaffolding and offer options (appropriate recommendations) asso‐ ciated with a situation to help people make wiser decisions to manage complex situations more wisely, overcoming some of the problems of bounded-rationality and becoming more T-shaped in their skillsets and mindsets to communicate better with people from other disciplines, systems, or cultural depth areas. To introduce this concept, we develop a framework for wisdom service systems showing how appropriately designed cognitive mediators may eventually harmo‐ niously interact with people and machines, adapt to evolving situations and contexts, as well as provide prioritized recommendations. The digital cognitive systems framework provides a way to understand the progression: tool, assistant, collaborator, coach, mediator, and sheds light on one interpretation DIKUW: data, information, knowledge, understanding, and wisdom. Finally, this paper concludes with future research directions to help realize this vision. Keywords: Cognitive mediators · Cognitive service systems · Service systems · Smart service systems · Wisdom computing · Wisdom service systems

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_11

116

1

Md.A.K. Siddike et al.

Introduction

Technologies with models of the world are emerging and re-shaping how people interact with each other and make tough decisions in conflict situations [1, 2]. People have limited life spans and limited cognitive capabilities, suffering from “bounded rationality” in deci‐ sion making—the condition noted by a Nobel Prize Laureate Herbert Simon [3]. As knowledge accumulates in society, Jones [4] identified the condition known as the “knowledge burden” as well as Arbesman [5] notes that there is a “half-life of knowl‐ edge” in any innovation-oriented society. To address these problems, great universities have appeared as “knowledge facto‐ ries” known for teaching (learning), discovery (research), and application of knowledge (entrepreneurship and policy making) [6]. Nevertheless, people need more resilient and sustainable approaches to address the rapidly evolving, and increasingly knowledgerich nature of socio-technical systems [4, 7]. In addition, people flooded with data, information, knowledge, and intelligence, need “wisdom” [8–13]. Traditional scholars of wisdom focus on acquiring this special type of knowledge from experience, and some even mention the importance of a sense of humor to cope with life’s challenges, including “limited life span”, “bounded ration‐ ality” and “knowledge burden.” Nowadays, new technologies can augment the cognitive and social capabilities of people, just as machines from an earlier age augmented the physical strength and capa‐ bilities of people. Today for example, cognitive computing [2] and sensor technologies [14] have begun to emerge to augment and scale the capabilities of people in specific ways. Eye tracking as well as motion tracking have the power to monitor people’s movements, so these technologies can build better models of their users, the task at hand, and relevant aspects of context and the world. The American engineer and inventor, and an early computer and Internet pioneer, Douglas Engelbart [15] urged society to work quickly to augment human intellect to address complex, and urgent problems. As technological capabilities have advanced, so have proposals for augmenting people, including such capabilities as weak telepathy, weak immortality and weak cloning [16]. Naturally, the suggestion of these types of advanced capabilities give rise to ethical concerns, and the need for a vision for harmonious interactions between people and machines [16]. Therefore, this paper introduces one way of thinking about the new reality, one which has been termed by some researchers as Reality 2.0 [41–43]. The focus of this paper is to explore the following questions: What is wisdom? Are rapidly evolving socio-technical systems becoming wiser in a measureable way (achieving mode of the common good)? What are wisdom service systems? How can interactions between people and machines becoming more harmonious? What are cognitive mediators, and how might they help create wiser systems and more harmo‐ nious interactions in a Reality 2.0? The rest of the paper is organized in the following order: Sect. 2 explores the topic of wisdom and wiser systems; Sect. 3 introduces the concept of wisdom service systems; Sect. 4 introduces the concept of cognitive mediators and wisdom decision processes; Sect. 5 concludes the paper with discussion and future research directions.

Wisdom Service Systems: Harmonious Interactions

2

117

Wisdom (the Common Good)

2.1 Practical Wisdom (Phronesis) Historically, wisdom has been seen as an intellectual virtue [8, 11]. The concept of phronesis originates with Aristotle and means a kind of practical wisdom, which is generally understood as the ability to determine and undertake the best action in a specific situation to serve the common good, not just the good of a specific individual. Phronesis is the ability to make the best judgment for the common good in a particular context, usually a novel, complex, conflict situation of some sort [17]. Nonaka and Toyoma [18] claimed that the effective strategic management requires distributed wisdom. 2.2 Wisdom as Expert Knowledge More recently, wisdom research has focused on wisdom as a special kind of expert knowledge [19]. Carr (pp. 216–217) wrote: Paul Baltes (1939–2006) and his colleagues at the Max Planck Institute for Human Development in Germany were world leaders in wisdom research… Baltes’ team developed five criteria by which the quality of a wise judgment or behavior may be evaluated. First, wisdom entails a rich store of declarative knowledge. Second, wisdom involves a rich repertoire of procedural knowledge about how to perform certain skills and routines such as complex decision making about interpersonal problems or conflict resolution. Third, wisdom involves lifespan contex‐ tualism. This is an appreciation of the many themes and contexts of life such as self, family, peer group, school, workplace, community, society and culture, and the varia‐ tions and interrelationships among these across the lifespan. Fourth, wisdom entails an appreciation of the relativism of values and life priorities with a tolerance for differences in values and priorities help by individuals and society in the service of the common good. The wise person is respectful of the unique set of values that other people hold, since the common good can be achieved by many routes. Finally, wisdom entails a recognition and management of uncertainty and a tolerance for ambiguity. It involves an appreciation that when solving any problem, each of us has access to incomplete information about the past and present; uncertainty about the future; and limited infor‐ mation-processing capacity. So wisdom… is the use of practical intelligence in a way that balances one’s own interests and those of others involved in the problem and the wider community to achieve a common good for all. 2.3 DIKIW Hierarchy and Wiser Systems Another possible starting point is the work of Ackoff [12]. Within the systems sciences and information systems communities, the DIKW (Data-Information-KnowledgeWisdom) hierarchy is frequently introduced to explain what is meant by data, informa‐ tion, knowledge, and wisdom, though the use of this framework is also criticized as lacking rigor [20, 21]. Nevertheless, advocates and critics agree, that Ackoff was the source of the DIKW hierarchy’s introduction and popularity, and that his original

118

Md.A.K. Siddike et al.

formulation included five levels, not four [10]: Data, Information, Knowledge, Intelli‐ gence (Understanding), and Wisdom (DIKIW). Ackoff associated (1) data with agents making measurement and creating symbols (examples included industrial sensors and scientific instruments), (2) information with agents performing communication and computation processes that can also answer basic-questions, (3) knowledge with agents decision-making and purposeful actions that can also answer how-to-questions, (4) intelligence with agents abilities to efficiently acquire more knowledge and answer whyquestions, and (5) wisdom with agents effectively using knowledge, understanding, and judgment to achieve a balance between individual and collective human values [10]. Ackoff wrote: Intelligence is the ability to increase efficiency; wisdom is the ability to increase effectiveness. The difference between efficiency and effectiveness—that which differentiates wisdom from understanding, knowledge, information, and data—is reflected in the difference between development and growth. Growth does not require an increase in value; development does. Therefore, development requires an increase in wisdom as well as understanding, knowledge, and information. Wisdom deals with values. It involves the exercise of judgment. Evaluations of efficiency are all based on a logic that, in principle, can be programmed into a computer and automated. These evaluative principles are impersonal. We can speak of the efficiency of an act inde‐ pendently of the actor. Not so for effectiveness. A judgment of the value of an act is never independent of the judge, and seldom is the same for two judges. People are the agents performing processes at all five levels of Ackoff’s original DIKIW hierarchy with data (at the bottom) being associated what people choose to measure and wisdom (at the top) being associated with what people choose to (collec‐ tively) value. In related work, Kline [22] shows that the link between measureable improvements in capabilities (more objective) and tasks that humans want done (more subjective) is a “basic pattern” of sociotechnical systems [22]: We have already defined sociotechnical systems as systems that link people with human-made hardware to perform tasks that humans want done… Many of the increases in human powers created by the use of sociotechnical systems have now reached beyond a billion times [what is] obtainable by a single human unaided by sociotechnical systems.

3

Wisdom Service Systems

Service systems are a type of socio-technical system where the focus is on value cocreation interactions [24]. The main function of a service system is to share access to resources with other entities to become more viable, robust and usefully complex [25]. In this age of information explosion or big data, service systems are becoming more complex, and people are often drowning in information, but only able to use a fraction of it to make decisions (bounded rationality). People understand, learn, reason, and interact, while smart machines model, adapt, infer, and interact by building, modifying, and sharing knowledge representations of people, each other, the world [2]. Tanca [26] stated that “not only do we want to make sense of the data, whether they are big or not, but we can, and should extract from them a worth that makes us wiser”. By definition, smart decisions help an individual, while wise decisions advance the common good as

Wisdom Service Systems: Harmonious Interactions

119

well. Bounded rationality limits how smart or wise the decision making of people can be. The main reason why we need wisdom service systems is that achieving the common good is becoming more difficult, and more important. The information may be available but cannot be processed efficiently or effectively enough to make a smarter decision (individuals co-create more value) or a wiser decision (the common good is served, as well as individuals co-creating value). Our conventional wisdom (human wisdom) hardly grasps a perspective or situation and understands it fully. Thus, people make a choice from their guts with limited information. Thus, the complexity, quantity, diversity of data, information, and expert knowledge are in some ways making our decision making harder. 3.1 Complexity of Growth As the population of the world continues to grow, the evolving ecology of nested, networked service system entities is becoming more complex, in terms of number value propositions connecting entities. The service systems are surrounded by a vast amount of data and information. Day by day, it becomes more complex. Therefore, we need to introduce some way to use connections and the information to improve the value prop‐ ositions and better serve the common good. 3.2 Quantity of Data Firms today must maintain both systems of record (back stage) and systems of engage‐ ment (front stage). Organizations collect data from a variety of sources, including busi‐ ness transactions, social media and information from sensor or machine-to-machine data. In addition, data streams in at an unprecedented speed and must be dealt with in a timely manner. Furthermore, RFID tags, sensors and smart metering are driving the need to deal with torrents of data in near-real time. In the vision of Reality 2.0, cognitive computing has the power to process the huge amount of information that exists in the real world and cyberspace. In Reality 2.0, people make wiser decisions because of their ability to use cognitive computing. 3.3 Diversity of Data Data comes in all types of formats – from structured, numeric data in traditional data‐ bases to unstructured text documents, email, video, audio, stock ticker data and financial transactions. An additional dimension, in addition to the increasing velocities and vari‐ eties of data, are highly inconsistent periodic peaks. Is something trending in social media today, sparked by a usual event? Daily, seasonal and event-triggered peak data loads can be challenging to manage. Even more so with unstructured data.

120

Md.A.K. Siddike et al.

3.4 Diversity of Using Information The use of information is diverse. People from all over the world can use the same information at the same time. In terms of value, information has the global to local space specification. In addition, the value of information is changing continuously. For example, the information which is valuable today, might not be valuable tomorrow. Furthermore, context of using information changes from person to person. Some infor‐ mation might be valuable in an American context, but the same information may not be valuable in a Japanese context, and vice versa. In such complex situations, making wiser decisions is very difficult for people. 3.5 Wisdom Computing The Japan Science and Technology Agency (JST) and several researchers conceived area of “wisdom computing” [27–31]. Wisdom computing is concerned with the design, management, use, and implications of information technologies for discovering, creating, sharing, and supporting wisdom [27]. Wisdom computing [28–30] discusses discovery, aggregation, search, and application of wisdom as a foundational technology platform in Reality 2.0 [41–43], that supports policy-based decision making with an emphasis on ELSI (Ethics, Legal and Societal Issues) and SSH (Social Sciences and Humanities) issues associated with business and society. In [41–43], the term “Reality 2.0” is coined and it implies that an inseparable fusion of cyber and physical things will lead to new concepts of identities for individual, community, institute, and nation. Moreover, [41] discusses a new service platform providing various societal functions based on the Reality 2.0 concept. Wisdom computing is required to make Reality 2.0 a healthier and wealthier world for people, with improved quality of life based on better decision making in complex and ever changing situations. It makes human life better in quality as people gain abilities and benefit from machines with ever-increasing capa‐ bilities that might threaten human jobs if not used to augment cognitive and social inter‐ actions in Reality 2.0 [28]. Wisdom computing focuses on interactions between people and machines for accumulation, propagation, exploration, and actuation of wisdom. In this era of overwhelming information explosion, people can access widespread “infinite” information in real time, but it is indeed unclear that people have become wiser than ever with the abundant knowledge and information [29]. On the other hand, machines are attaining enormous capabilities in accessing and analyzing information and control‐ ling objects such as airplanes and automobiles [30]. Research activities in Japan related to wisdom computing focus on understanding and developing wisdom by sublimating distributed and heterogeneous data and information. This research asks the question: “How can greater levels of individual and societal wisdom be achieved through collaboration between people and machines, in order to better balances living in both the physical and cyber realms of Reality 2.0” [29, 41–43]. It also focuses on ethical, legal, and social issues related to social responsibilities for actuation of wisdom [30]. In the area of cognitive computing, almost all the people of our society including doctors, physicians, patients, bankers, policymakers, tourists, customers, as well as community people can be greatly augmented by the cognitive mediators [32].

Wisdom Service Systems: Harmonious Interactions

121

However, beyond cognitive computing, wisdom computing deals with issues relating to legal, ethical, and public policy – all to ensure the common good. Cognitive systems allow cognitive assistants for all occupations in smart service systems, boosting individual people’s creativity and productivity. Wisdom systems improve sustainability and justice [31, 33]. In this paper, we define wisdom service system as system in which people have cognitive mediators that offer appropriate options by actuating the context and situation that help people make wiser decisions (common good) to solve complex problems more efficiently and effectively, overcoming some of the problems of bounded-rationality [23]. A wisdom service system helps our next generation to build and re-build from the scratch. In wise service systems, cognitive mediators have the mighty power to generate insights from trillions of unstructured and structured data, process the data and provide precise recommendations that are termed as “artificial wisdom”. On the other hand, people have their own wisdom termed a “human wisdom” that is acquired by sensing the real world environment, using knowledge and experiences. People’s recommenda‐ tions are based on their experiences, knowledge and skills to solve complex problems. In this way, people and machine collaborate harmoniously and generate win-win value co-creation for individuals and the common good. In Reality 2.0, win-win value cocreation goes beyond individuals and includes policies for the common good as well (See Fig. 1).

Fig. 1. Wisdom service system structure

4

Cognitive Mediators

Figure 2 represents our proposed wisdom service systems process in which people have cognitive mediators, equipped with sociometers to help them build better models of the user, themselves, the world, and the task at hand. Blue dots indicate entities in the real world, yellow dots are metrics (sociometrics) observed by sociometers and green dots indicate appropriate options (recommendations) generated by mediators.

122

Md.A.K. Siddike et al.

Fig. 2. Wisdom service system process

4.1 Observing Situations, Context and Culture Traditionally, people interact with people and objects; understand people behavior, situation, context and culture. In addition, people actuate the situation, context and culture by physically visiting the real world. Furthermore, intuition, knowledge and experience of human are important to actuate the situation, context and culture. But in our proposed wisdom service systems process, sociometers (sensors/agents) able to contextualize human communication behaviors; situation, context and culture through the trusted communication signals [14]. Using the sociometer, people signaling situa‐ tions can accurately be predicted. On the other hand, cognitive mediators are also able to observe trillions of unstructured data related to human/machine. 4.2 Understanding Situations, Context and Culture Sociometers are able to predict and measure the social context of people. Mediators have the ability to process the unstructured and structured data to understand situations, context and culture based on their models and knowledge representation of their user, themselves, the task at hand, the context and the world. 4.3 Grasping Situations, Context and Culture After understanding the complex situations, sociometers and mediators can grasp the evolving patterns of human situations, context and culture. This involved modeling the world, adapting those models, extended the models, and sharing the models with the cognitive mediators of other people.

Wisdom Service Systems: Harmonious Interactions

123

4.4 Providing Options After observing, understanding and grasping people situations, context and culture, mediators generate and provide recommendations for their users. In this way, the deci‐ sion making capabilities of people are enormously augmented by the cognitive media‐ tors. Thus, people will be better able to make wise decisions for solving complex prob‐ lems of business and society the benefit the common good. As a result, every service system in which people have cognitive mediators will become a wisdom service system, and overall business and society will become wiser using the artificial wisdom needed in Reality 2.0. 4.5 Wisdom Decision Process In this section, we propose a step by step iterative wisdom decision process as illustrated in Fig. 3.

Fig. 3. Wisdom decision process

Step 1: First of all, we have to figure out the common good and how we can measure common good of society, individuals, communities and machines? In order to under‐ stand social good or common good, we have to understand the mutual goal of the society by setting up the mutual understanding. In some cases, gross national happiness (GNP) shows one kind of measure of social good or common good. In this case, how agents and humans can understand the common good by articulating the situation, context and culture of humans and machines. Then, if situation changed, common goods will also be changed. There might be some common goods for human might not be changed for thousand or million or billion years. Then, there might be cultural contexts namely Japanese, Americans or Indians. Then, individuals’ culture might come up with some common goals. For example, Japanese culture might be harmonic society that might not be changed in the next 100 years. As a result, we have to survive as species.

124

Md.A.K. Siddike et al.

Step 2: In the second step, how we can measure common good by sociometrics? Shall we peruse values which can be measured? Common good might not be a measurable thing. In this case, we have to compromise sub-measures (a, b, c). Or set of sub-measures (a, b, c…) which is helpful for mediators guessing the status of society, individuals and communities. Step 3: In the third step, mediators can understand and interpret the status of individ‐ uals, society and communities and come up with options estimating effects. In this case, artificial wisdom with compassion to people might harmoniously interact with human’s wisdom. In addition, sociometers (agents) and artificial wisdom will provide recom‐ mendations that might be considered by human side or objects. This artificial wisdom is real intellectual wisdom based on individuals, society and community. Values, we want to pursue it might be changed after certain period of time. Then, this kind of socio‐ meters and artificial wisdom might be changed based on the changed in the context. In this case, we have to repeat the step 1. Step 4: In this step, how individuals, society and community reach wiser choice of options. Note that there is a gap between the ideal state and the state realized by this wise choices. To fill this gap, we need wisdom. Then how our wisdom computing helps to fill the gap through providing satisfaction to individuals, society and community. This process iterates and the status of the individuals, society and community will be changed providing situated service.

5

Discussions and Future Research Directions

This paper builds on earlier attempts to understand wisdom service systems (cognitive mediators; common good), and how they are different from smart service systems (cogni‐ tive assistants, individual efficiency and effectiveness) [33, 34]. Intelligence involves language, learning, and levels of confidence in cognitive systems [34]. On the other hand, wisdom connects with human values and stands the test of time and perspectives of multiple generations; wisdom connects to courage and not cowardice; wisdom does not control the narrative cosmetically or politically, but does suggest a higher purpose for our actions than individual selfishness or complete altruism [33]. Bostrom [35] in his book “Superintelligence” writes about ethical artificial intelligences, and highlights the work of American artificial intelligence researcher Eliezer Yudkowsky, who writes about coherent extrapolated volition (CEV), timeless decision theory, and the complex value systems that may well be required to realize valuable futures [34–36]. For sociotechnical systems in general, understanding if there are smart and wise “speed limits” or “no speed limits” for the growth of sociotechnical system capabilities is an open question [6, 37, 38]. These are issues for the human-side of service engineering to study in the context of increasing customer capabilities through improved value co-creation interactions [39]. This paper explores a new way of actuating wisdom in service systems through an elaboration of the processes for the harmonious collaboration between people and machines in Reality 2.0. We predict that all people will have cognitive mediators that help them overcome some of the problems of bounded rationality in a world where data,

Wisdom Service Systems: Harmonious Interactions

125

information, and knowledge are growing at exponential rates. In this Reality 2.0, cogni‐ tive mediators will have models of their user, themselves (what they know and do not know), the task at hand, and also aspects of the world in order to help people interact and make decisions. Cognitive mediators are a type of software social organisms [40]. People working in all occupations in business and society will potentially benefit from cognitive assistants and cognitive mediators [34]. If cognitive mediators can shift people’s decision-making to better serve the common good, then Reality 2.0 will become an evolving ecology of nested, networked wisdom service systems with harmonious interactions between people and machines. To achieve this Reality 2.0 vision, a new curriculum is under development to help people build, understand, and work with cognitive mediators in wisdom service systems.

References 1. Gelernter, D.: Mirror worlds: or the day software puts the universe in a shoebox… How it will happen and what it will mean. Oxford Paperbacks (1992) 2. Kelly III, J.E., Hamm, S.: Smart Machines: IBM’s Watson and the Era of Cognitive Computing. Columbia University Press, New York (2013) 3. Simon, H.A.: Administrative Behavior: A Study of Decision-Making Processes in Administrative Organizations. The Free Press, New York (1997) 4. Jones, B.F.: The burden of knowledge and the ‘death of the renaissance man’: is innovation getting harder? NBER Working Paper, no. 11360, May 2005 5. Arbesman, S.: The Half-Life of Facts: Why Everything We Know has an Expiration Date. Penguin, New York (2013) 6. Spohrer, J., Giuiusa, A., Demirkan, H., Ing, D.: Service science: reframing progress with universities. Syst. Res. Behav. Sci. 30(5), 561–569 (2013) 7. Spohrer, J.: Service science. In: Bainbridge, W.S. (ed.) Leadership in Science and Technology: A Reference Handbook. Sage Publications, New York (2011) 8. Nonaka, I., Takeuchi, H.: The wise leader. Harv. Bus. Rev. 89, 58–69 (2011) 9. Baltes, P.B., Staudinger, U.M.: Wisdom: a metaheuristic (pragmatic) to orchestrate mind and virtue toward excellence. Am. Psychol. 55(1), 122–136 (2000) 10. Sternberg, R.J.: Wisdom, Intelligence and Creativity Sysnthesized. Cambridge University Press, Cambridge (2003) 11. Schwartz, B., Sharpe, K.: Practical Wisdom: the Right Way to do the Right Thing. Riverhead Books, New York (2010) 12. Ackoff, R.L.: From data to wisdom. J. Appl. Syst. Anal. 16, 3–9 (1989) 13. Ardelt, M.: Wisdom as expert knowledge system: a critical review of a contemporary operationalization of an ancient concept. Hum. Dev. 47, 257–285 (2004) 14. Pentland, A.: Honest Signals: How They Shape Our World. MIT Press, Cambridge (2008) 15. Engelbart, D.C.: Augmenting human intellect: a conceptual framework. Summary Report, Stanford Research Institute, on Contract AF 49(638)-1024, pp. 1–134, October 1962 16. Lenat, D.B.: WWTS (what would Turing say?). AI Mag. 37(1), 97–101 (2016) 17. Nonaka, I.: Wise leader: building communities of sustainable innovation, October 2011. http:// www.theaimblog.com/wp-content/uploads/2011/11/Lecture-by-Prof.-Ikujiro-Nonaka.pdf 18. Nonaka, I., Toyama, R.: Strategic management as distributed practical wisdom (phronesis). Ind. Corp. Change 16(3), 371–394 (2007)

126

Md.A.K. Siddike et al.

19. Carr, A.: Positive Psychology: The Science of Happiness and Human Strengths. Routledge, New York (2011) 20. Rowley, J.E.: The wisdom hierarchy: representations of the DIKW hierarchy. J. Inf. Sci. 33(2), 163–180 (2007) 21. Frické, M.: The knowledge pyramid: a critique of the DIKW hierarchy. J. Inf. Sci. 35(2), 131– 142 (2009) 22. Kline, S.J.: Conceptual Foundations for Multidisciplinary Thinking. Stanford University Press, Palo Alto (1995) 23. Spohrer, J., Siddike, M.A.K., Kohda, Y.: Rebuilding evolution: a service science perspective. In: Proceedings of the 50th Hawaii International Conference on System Sciences, Hilton Waikoloa Village, Hawaii, pp. 1663–1672, 4–7 January 2017 24. Maglio, P.P., Vargo, S.L., Caswell, N., Spohrer, J.: The service system is the basic abstraction of service science. IseB 7(4), 395–406 (2009) 25. Spohrer, J., Vargo, S.L., Casewell, N., Maglio, P.P.: The service system is the basic abstraction of service science. In: The Proceedings of 41st Hawaii International Conference on System Sciences, pp. 1–10. IEEE Press, New York (2008) 26. Tanca, L.: Wisdom: a “double V” for big data. ACM SIGMOD Blog, February 2016. http:// wp.sigmod.org/?p=1948 27. Dalal, N.: Wisdom computing: toward a framework for wisdom search in information systems. In: The Proceedings of the Eighteenth Americas Conference on Information Systems, Seattle, Washington, 9–12 August 2012. http://aisel.aisnet.org/amcis2012/proceedings/PerspectivesIS/14 28. JST: Wisdom computing – research and development for creative collaboration between humans and machines (2014). http://www.jst.go.jp/crds/en/publications/CRDS-FY2013SP-07.html 29. Iwano, K., Motegi, T.: Wisdom computing: toward creative collaboration between humans and machines. J. Inf. Process. Manag. 58(7), 515–524 (2015). [Translation] 30. Iwano, K.: Wise computing: collaboration between people and machines. Panel Discussion at AAAS Annual Meeting, San Jose, CA USA (2015). https://aaas.confex.com/aaas/2015/ webprogram/Session9386.html 31. Spohrer, J.: From cognitive computing to wise (or wisdom) computing: a service science perspective. AAAS Annual Meeting, San Jose, CA, USA, 15 February 2015. http:// www.slideshare.net/spohrer/wise-computing-20150215-v3 32. Spohrer, J.: Innovation for jobs with cognitive assistants: a service science perspective. In: Nordfors, D., Cerf, V., Senges, M. (eds.) Disrupting Unemployment, pp. 157–174. Ewing Marion Kauffman Foundation, Missouri (2016) 33. Spohrer, J., Bassano, C., Piciocchi, P., Siddike, M.A.K.: What makes a system smart? Wise? In: Ahram, T., Karwowski, W. (eds.) Advances in the Human Side of Service Engineering, pp. 23–34. Springer, Cham (2017) 34. Spohrer, J., Banavar, G.: Cognition as a service: an industry perspective. AI Mag. 36(4), 71– 86 (2015) 35. Bostrom, N.: Superintelligence: Paths, Dangers, Strategies. Oxford University Press, Oxford (2014) 36. Yudkowsky, E.: Intelligence explosion microeconomics. Technical report 2013-1, Machine Intelligence Research Institute, Berkeley, CA, USA (2013) 37. Mitroff, I.I., Linstone, H.A.: The Unbounded Mind: Breaking the Chains of Traditional Business Thinking. Oxford University Press, New York (1993) 38. Wright, R.: Non-zero: The Logic of Human Destiny. Vintage/Random House, New York (2000)

Wisdom Service Systems: Harmonious Interactions

127

39. Freund, L.E., Spohrer, J.C.: The human side of service engineering. Hum. Fact. Ergon. Manuf. Serv. Ind. 23(1), 2–10 (2013) 40. Forbus, K.D.: Software social organisms: implications for measuring AI progress. AI Mag. 37(1), 85–90 (2016) 41. Iwano, K., et al.: Expectation of societal innovation by ICT – REALITY 2.0. Center for Research and Development Strategy, Japan Science and Technology Agency, JST, CRDSFY2015-XR-05, November 2015. (in Japanese). http://www.jst.go.jp/crds/pdf/2015/XR/ CRDS-FY2015-XR-05.pdf 42. Iwano, K., Anzai, Y. and Kuma, K: Cyber society and human being – Reality 2.0 and our coming future. Science Agora Keynote Session, 13 November 2015, http://www.jst.go.jp/csc/ scienceagora/program/session/fb_506/ 43. JST: Reality 2.0. Summit Report. CRDS, JST, 27–28 September 2015. http://www.jst.go.jp/ crds/pdf/2015/WR/CRDS-FY2015-WR-03.pdf

Do You Like According to Your Lifestyle? A Quantitative Analysis of the Relation Between Individual Facebook Likes and the Users’ Lifestyle Alexander Piazza1(&), Christian Zagel2, Julia Haeske1, and Freimut Bodendorf1 1

2

Friedrich-Alexander-University Erlangen-Nuremberg, Lange Gasse 20, 90403 Nuremberg, Germany {Alexander.Piazza,Julia.Haeske, Freimut.Bodendorf}@fau.de Coburg University of Applied Sciences and Arts, Friedrich-Streib-Straße 2, 96450 Coburg, Germany [email protected]

Abstract. The performance of companies depends on the ability to leverage data to create insights and to target consumers with personalized messages Like marketing content or product offerings. One key element for personalized targeting are expressive user profiles, which are the basis for predictive models to estimate individual consumers’ preferences. Traditionally user profiles are mainly based on demographic attributes like age, gender, or occupation. Due to changes in society, consumers’ behaviors are less stable, and therefore these demographic factors are less effective. Alternatively, the consumers’ lifestyle has a significant impact on their purchase and consumption behavior. This paper investigates the relationship between Facebook Likes and the lifestyle of individuals based on the activity, interests, and opinion (AIO) model. Therefore, 14482 user-Like combinations from 214 participants were collected together with lifestyle information and a correlation analysis is conducted. The results indicate weak monotonic correlations between the AIO and the Like information. Keywords: Social network segmentation
Facebook data

data
Consumer
AIO-model

profiling




Lifestyle

1 Introduction In industry and academia, the awareness rises that the success of companies strongly depends on their ability to leverage data to generate insights. One of the first studies revealing the relationship between data-driven decision making (DDD) and company performance were Brynjolfsson et al. (2011) in their article “Strength in Numbers: How Does Data-Driven Decisionmaking Affect Firm Performance?” [1]. In that research, 179 large publicly traded companies were analyzed, resulting in the insights that the productivity of these adopting DDD is in average 5–6% higher than expected based © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_12

Do You Like According to Your Lifestyle?

129

their other investments. This positive impact was confirmed in [2] where 3000 participants were interviewed, coming from various management levels, 30 different industries, and 108 countries. This study revealed that companies belonging to the top-performing organizations in their industry are twice as likely to apply analytics to their daily business as the mid- to low-performing organizations in the same industry. Companies selling via e-commerce especially benefit from DDD. As [3, p. 1] states: “Big Data enables merchants to track each user’s behavior and connect the dots to determine the most effective ways to convert one-time customers into repeat buyers.” By analyzing the available data, e-commerce companies can personalize product offerings and the communication towards individual consumers what leads to higher sales and consumer engagement. For instance [3] reported that the online store Bikeberry.com used their consumer data to target consumers individually with optimized product offers that lead to increased sales as well as on-site engagement. In general, it is reported that personalization can increase sales by 10% or more [4]. The effectiveness of personalization depends on the accuracy of the predictive models for estimating consumers’ preferences and behavior. In general, the predictive accuracy of the models increases when more characteristic information about the consumers are added e.g. by applying data fusion [5, 6]. For example in [7] the accuracy of the predictive model for targeting individual consumers in the automotive industry was increased by merging transaction data of consumers with market research data. Especially nowadays, it is important to leverage additional consumer information characterizing their purchase behavior, as in general consumers are showing less predictable consumption patterns [8]. As a consequence, traditional criteria for grouping customers into meaningful subsegments, like demographics, are getting less effective [9]. Alternatively, lifestyle attributes reflect the consumers’ purchase behavior more adequately. In parallel, the amount of data organizations and individuals generate is increasing. Particularly in using the Internet, people generate data describing their behavior, an inadvertent outcome often referred as “leaving digital traces”. Emerging technologies and analytics methods enable organizations to analyze these data using what is frequently subsumed under the term Big Data analytics [10]. In 2016 Abbasi, Sarker, and Chiang presented a research agenda for Big Data in Information Systems, collecting and systemizing relevant questions regarding the impact of Big Data on organizations [11]. In this paper, they emphasize the need for further research to understand better the predictive power of the available Big Data to predict events on from individuals or micro-level. They raise the question: “How can one leverage the volume and variety of big data to develop novel artifacts for predicting/describing macro versus individual/micro-level phenomena or events?” [11, p. xvii]. One recent research investigating this kind of question demonstrated that Facebook Likes have a high predictive rate of success for estimating users’ characteristics like personality traits. In [12] the authors showed that there is a significant relationship between personality traits operationalized as Big Five values and the Facebook profile information. In a following study [13], the authors are able to predict various private traits and attributes like sexual orientation, religious and political views, intelligence, happiness, age, and gender based on the people’s Facebook Like data. The objective of this paper is to investigate whether it is possible to infer a person’s lifestyle characteristic by analyzing their Facebook Likes. According to marketing literature, an individual’s lifestyle can be represented as the collection of their activities,

130

A. Piazza et al.

interests, and opinions (AIO) (see Sect. 2.1). From the authors’ perspective, this is similar to fan pages people usually like on Facebook. Therefore, the assumption in this research is that by grouping the Likes data according to the AIO lifestyle dimensions, a correlation to the users’ lifestyle values should be revealed. Therefore, lifestyle information are collected within a study by using validated item batteries as well as Facebook Likes. These Likes are then grouped according to the AIO lifestyle dimensions by two different approaches, and finally, a correlation analysis is conducted to assess the relationship. The contribution of this research is to provide insights regarding the relationship between Facebook Likes and lifestyle dimensions as well as insights about the collection of Facebook Likes from individuals.

2 Fundamentals 2.1

The Lifestyle Concept

The lifestyle of people has a significant influence on how people purchase and consume goods [14]. Lazer gave the first definition of lifestyle in the marketing context in 1963. He defined lifestyle as “… a systems concept. It refers to the distinctive or characteristic mode of living, in its aggregative and broadest sense, of a whole society or segment thereof” [15, p. 130]. According to this definition, the lifestyle of individuals, as well as groups, can be clearly distinguished. Another definition of lifestyle is given by Wind [16, p. 302]: “Life style reflects the overall manner in which people live and spend time and money.”According to this definition, lifestyle can be operationalized in two ways: (I) by the products a person consumes or (II) by his/her activities, interests, and opinions (AIO). In this paper, the latter approach is used, referred as the AIO-model. In 1976, Plummer concretized the lifestyle concept by focusing it to individuals only and gave examples for the AIO-dimensions (see Table 1) [17]. In marketing research, the lifestyle of individuals is typically collected by conTable 1. Lifestyle dimensions according to [17] Activities Work Hobbies Social events Vacation Entertainment Club memberships Community Shopping Sports

Interests Family Home Job Community Recreation Fashion Food Media Achievements

Opinions Themselves Social issues Politics Business Economics Education Products Future Culture

Demographics Age Education Income Occupation Family size Dwelling Geography City size Stage in life cycle

ducting surveys based on developed and evaluated item batteries [19]. The length from the surveys ranges from 18 items in [18], 56 items in [19], and 300 in [20]. A general problem with surveys is that people have low motivation in participating, especially if they are time-consuming. Furthermore, from an industry perspective, these data have to

Do You Like According to Your Lifestyle?

131

be frequently recollected, as the lifestyle of people changes over time. Therefore, alternative measures are promising, determining the lifestyle of people in an implicit way, without asking people to give explicit statements.

2.2

Facebook

Facebook is one of the dominant social networks, having been founded in 2004 and on average currently having 1.23 billion daily active users [21]. Within this network, people can establish relationships, share content like text, images, or videos and express a favourable opinion towards internal objects like content, groups, or fan pages by assigning a so-called Like to it. According to [22] the main reason for members to use Facebook are staying in connection with friends, and the value of the network as an information source. Based on a sample dataset of 13,619 users, [23] showed that the median of Likes of all users is 152, but also that the distribution of the number of Likes per number of users is approximating a power law distribution. Therefore, many users are liking only a few pages, and few users are liking many pages.

3 Experimental Setup The data collection process is illustrated in Fig. 1. During data collection, the participants were also asked to state their opinion regarding specific examples of clothing in step (1), and to fill out a personality survey to determine the Big Five personality factors in step (3). Since for the objective of this research these attributes are not relevant, this part was omitted. The study was conducted using the survey platform Unipark1 in combination with a

(1) EvaluaƟon of Clothings

(2) Lifestyle Items

(3) Personality Items

(4) Demographic quesƟons

(5) Gaterhing of Facebook Likes via Facebook Login

Fig. 1. Steps within the survey

self-developed Facebook login based on the Graph API for extracting the Facebook Likes. To be able to connect the responses from the Unipark survey with the Facebook data, the participants were asked to give their e-mail address during the survey, which they also used to register to Facebook.

1

http://www.unipark.com/.

132

A. Piazza et al.

3.1

Survey Structure

To determine the lifestyle attributes of the participants, a selection from the item battery introduced in [19] was used. Originally, this item battery consisted of 56 items for determining ten lifestyle dimensions. To shorten the survey 20 randomized items were chosen in a way, so two items per dimension were available. The selected attributes are listed in Table 2. Table 2. Selected 20 items for the lifestyle dimensions based on the 56 items introduced in [19] Dimension Family orientation

Self-consciousness Fashion consciousness Explorer Community oriented Practical Homebody Health orientation Cost conscientious

Conservatism

Selected items (1) When a family member of mine is ill in bed, I drop most everything else in order to see to their comfort (2) My family is the most important thing in my life (3) I think I have more self-confidence than most people (4) I am more independent than most people (5) I have one ore more outfits that are the latest style (6) An important part of my life and activities is dressing smartly (7) I often try new stores before my friends and neighbours do (8) I would like to take a trip around the world (9) I do volunteer work on a fairly regular basis (10) I would like to work on community projects (11) I like to sew and frequently do (12) I often make my own clothes (13) I would rather spend a quiet evening at home than go out to a party (14) I like parties where there is lots music and talk (15) I buy more low calorie foods than the average person (16) I have used diet foods at least one meal a day (17) I shop for specials (18) I find myself checking the prices in the grocery store even for small items (19) I often seek out the advice of my friends regarding which brand to buy (20) I spend a lot of time talking with my friends about products and brands

The language of the survey is English and German. Therefore, the items where translated into German. Per item, the participants could choose on a five-level Likert scale.

3.2

Facebook like Data Collection

For extracting Like information from the participants, a Facebook app was developed in the programming language PHP and the data was stored via SQL in a MySQL database. By designing the login as a Facebook app, the login was not limited to a

Do You Like According to Your Lifestyle?

133

specific platform but could be used on every computer, tablet, or mobile. After developing this app, it had to be reviewed by Facebook to get their approval. Therefore, the app together with a description had to be uploaded to the Facebook development platform. After five days, Facebook authorized the developed app for the usage within the survey (Fig. 2).

Fig. 2. Facebook login screen during the survey

To ensure the privacy of the data, the Facebook extraction was developed and run on a dedicated server with only the authors having access to the raw data. Before analyzing this data, it was anonymized and the raw data was deleted. As server, a Banana PI single board computer running on a Bananian Linux was used. Participants were informed about all data which was extracted after the login to ensure transparency to the participants as well as to increase their acceptance and willingness to participate. Only data, which was necessary for this study, like the e-mail address for matching with the survey data, or the Like of fan pages, were extracted. No further information for example Likes of images or comments, comments, or images were extracted.

4 Data Collection and Analysis The survey was open for participation from February 2nd till March 1st 2016. The main channel for recruiting participants were Facebook groups, as this ensured to target people having a Facebook account is necessary for the participation. As illustrated in Fig. 3, 2596 participants started the survey. In the step of the survey where the people were asked to login into Facebook, only approx. 40% of the people proceeded. It seems that the majority of the participants had significant concerns about sharing data from their Facebook account. In total 498 participants finished the complete study. Unfortunately, because of technical issues from 194 participants no Facebook Like information were stored and therefore had to be excluded from the analysis. Furthermore, in 54 cases duplicated entries were found, which were excluded as well. For 14 participants, the Facebook data could not be assigned to the survey data because the participants used different logins in Facebook and the survey. The data from 226 participants could be considered in the final data analysis.

134

A. Piazza et al.

Fig. 3. Number of participants per step within the survey

4.1

Descriptive Analysis

The group participants consist of 55 male (=26%) and 159 female (74%) subjects. The higher number of women might be explained by the focus on fashion products in this survey. As already stated in Sect. 3, the fashion related questions of the survey is not part of this research and are therefore omitted. The average age is 25.5 years and having 92% of participants being younger than 30. The age distribution is illustrated in Fig. 4.

Fig. 4. Age distribution of participants

In total 14,482 user-Like combinations were collected. To each Like the related Facebook Like category was stored as well. In Fig. 5 the distribution of Facebook Likes is illustrated. As mentioned above, only Likes of fan pages were extracted. For the authors, it was unexpected that a large number of participants had less than 100 Likes and 92 participants had exactly 100 Likes. As mentioned in subsection 2.2 another study reported that the average user in their dataset had 152 Likes. Possible reason for this could be that the users in our study had stricter privacy rules allowing

Do You Like According to Your Lifestyle?

135

only the extraction of a fraction of their Likes, or technical reasons, such as the processing power of the single board computer used was too low for handling peaks during the data collection phase.

Fig. 5. Distribution of number of Likes

4.2

Facebook Likes Preprocessing

To correlate the data with the results of the survey, the 14,482 Facebook Likes were manually classified within the ten lifestyle dimensions defined in [19] based on two approaches. In the first approach, the Likes were categorized in the ten dimensions, whereas in the second approach the Likes were categorized based on their Facebook internal Like categorization to the lifestyle dimensions. An example of the first approach, as well as the overall fraction of Likes per dimension, are illustrated in Table 3.

Table 3. Manual assignment of the Likes to ten lifestyle dimensions AIO lifestyle dimension Family orientation Self-consciousness Fashion consciousness Explorer Community oriented Practical Homebody Health orientation Cost conscientious Conservatism

Fraction of Likes Example Likes 11% Babyfotos 24; Thermomix 14% I am capable; I’m a muslim & I’m proud 9% ZARA; Shopping Queen 15% Rock im Park; ISt Sprachreisen 11% Fairtrade; UNICEF 1% Do-it-Yourself; do your own fashion 19% Lieferando.de; YouTube 13% Alnatura; Tennis 5% myDealZ; Flixbus 1% Auswärtiges Amt; Deutsche Bank

136

A. Piazza et al.

The Like data is converted to a ranking representation having five levels. Therefore, the proportion of each Like per dimension for each participant is calculated. Afterwards the scale values are determined based on the following rules: scale value 1 for 0% of Likes, 2 for 1–5%, 3 for 6–20%, 4 for 21–49%, and 5  50%. An example conversion for a user having 100 Likes is illustrated in Table 4. Table 4. Example for the conversion of the number of Facebook Like to scale values for one participant having 100 Likes AIOdimension

Family orientation

Selfconsciousness

Fashion consciousness

Explorer

Community oriented

Practical

Homebody

Health orientation

Cost conscientious

Conservatism

Number Likes

21

7

7

7

13

1

2

24

12

6

Percentage

21%

7%

7%

7%

13%

1%

2%

24%

12%

6%

Scale value

4

3

3

3

3

2

2

4

3

3

4.3

Correlation Analysis

The correlation analysis was conducted using the SPSS statistics software2. As both the data determined by the itemsets as well as the preprocessed Like data are on a ordinal scale, the Spearman’s rank correlation coefficient was applied. The first correlation analysis focused on the first approach of grouping the individual Likes manually to the lifestyle dimensions with the lifestyle dimension gathered by the AIO-items from the survey. In Table 5 the resulting correlation coefficients are illustrated. The gray marked fields indicate results having significant correlations. In the diagonal fields, the correlation coefficient between the same lifestyle dimension coming from the two data sources. The results indicate a relation between the family orientation, fashion consciousness, and community orientation dimension. The correlation between the first to dimensions are even significant at a level of 0.01 which indicates a strong significance. Between the remaining dimensions, there are no significant correlations. Furthermore, there are correlations between different dimensions, like the fashion consciousness and the community orientation. However, the correlation analysis resulted in only small coefficients below 0.3 what means that even some of the correlations are significant, the correlation is only weak. The next analysis is based on the second approach of grouping the Likes based on their Facebook categories and the lifestyle dimensions based on the AIO-items. This analysis only lead to 10 significant correlations in comparison to the 15 in the first approach. This result suggests that the Facebook categories seem to be less appropriate compared to the first approach for grouping Likes into the lifestyle dimension in a meaningful manner.

2

https://www-01.ibm.com/software/de/analytics/spss/.

Do You Like According to Your Lifestyle?

137

Table 5. Correlation matrix of the AIO-dimensions from collected by the items, and the AIO-dimensions of the Likes assigned to the ten AIO-dimensions

,085

,087

,191 **

-,001

,081

-,001

-,028

Explorer

,017

,065

Community oriented

,149 *

-,155 *

Practical Homebody

,009 ,092

,029 -,118

-,084

Conservatism

,026

Cost conservatism

,079

Health orientation

,067

Homebody

,151 *

Practical

-,023

Community oriented

Fashion consciousness

,185 **

Explorer

Self-consciousness

Family orientation Selfconsciousness Fashion consciousness

Family orientation

Determined by Items

Determined by Likes

-,047

-,081

-,019

,056

,085

-,067

-,018

,023

-,040

-,093

,061

-,064

-,018

-,107

,079

-,074

,006

,008

,090

,054

,000

-,028

-,051

,007

-,112

,181 **

,126

,004

-,002

-,161 *

-,045

,027

-,067

-,007

-,017

,043

,095

,016 ,030

-,107

,132

-,081

,035

-,161 *

-,214 ** -,165 * ,291 ** ,186 ** ,040

Health -,175 ,033 ,154* ,090 -,021 * orientation Cost ,060 ,007 -,034 ,087 -,045 conservatism ,042 ,094 -,071 -,115 Conservatism -,062 * = The correlation is significant on the 0.05% level ** = The correlation is significant on the 0.01% level

-,004

,118

,078

-,138 *

-,053

-,061

,053

,124

-,049

-,005

,020

-,012

-,110

,108

5 Conclusion The objective of this research was to investigate the relationship between Lifestyle dimensions gathered by developed and evaluated item sets and Facebook Likes. Therefore Facebook Likes were aggregated manually based on the name of the Like, and secondly based on the related Facebook category. The correlation analysis indicates significant but weak correlations between the Like data and the lifestyle information. The grouping of Likes based on their name lead to 15 significant correlations, whereas the grouping based on the Facebook categories lead only to 10 significant correlations. One indication of this result is, that the first approach seems to better represent the lifestyle than the second approach. The overall result gives first indications that a meaningful relationship between Facebook Likes and the Lifestyle attributes based on the AIO-model exists. Limitations are especially the relatively low number subjects which could be used in the analysis. The reason for this is the high dropout rate, the incomplete Like data extraction per user as well as not missing Likes in the database because of technical issues. One reason for

138

A. Piazza et al.

the high dropout rate seems to be privacy concerns of the participants. Therefore the authors suggest to make the Like data extraction and data processing more transparent to the participants, to increase trust and therefore the willingness to participate. From the technical perspective, special focus on testing the reliability of the Like data extraction mechanism should be set. Furthermore, the processing power of the used single board computer seems not to be sufficient for gathering the data, especially when multiple users are simultaneously accessing the survey. Therefore it is recommended for future studies to evaluate the extraction system not only from a functional perspective, but also to conduct performance tests.

References 1. Brynjolfsson, E., Hitt, L.M., Kim, H.H.: Strength in numbers: how does data-driven decisionmaking affect firm performance? SSRN eLibrary (2011). doi:10.2139/ssrn.1819486 2. LaValle, S., Lesser, E., Shockley, R., Hopkins, M.S., Kruschwitz, N.: Big data, analytics and the path from insights to value. MIT Sloan Manag. Rev. 52(2), 21–32 (2011) 3. Jao, J.: Why big data is a must in ecommerce (2013). http://www.bigdatalandscape.com/ news/why-big-data-is-a-must-in-ecommerce. Accessed 2 Mar 2017 4. Liebowitz, J.: Big Data and Business Analytics. CRC Press, Boca Raton (2013) 5. Van Hattum, P., Hoijtink, H.: The proof of the pudding is in the eating. Data fusion: an application in marketing. J. Database Mark. Cust. Strat. Manag. 15(4), 267–284 (2008) 6. Van Hattum, P., Hoijtink, H.: Improving your sales with data fusion. J. Database Mark. Cust. Strat. Mana. 16(1), 7–14 (2009) 7. Wortmann, C., Piazza, A., Brauner, D., Jäck, H.: Evaluating data integration in the CRM context at the example of the automobile industry. In: 8th International Workshop on Data Mining in Marketing, pp. 21–32. ibai-publishing, Fockendorf (2015) 8. Miguéis, V.L., Camanho, A.S., Falcão e Cunha, J.: Customer data mining for lifestyle segmentation. Expert Syst. Appl. 39(10), 9359–9366 (2012) 9. González, A.M., Bello, L.: The construct “lifestyle” in market segmentation. Eur. J. Mark. 36(1/2), 51–85 (2002) 10. Chen, H., Chiang, R.H.L., Storey, V.C.: Business intelligence and analytics: from big data to big impact. MIS Q. 36(4), 1165–1188 (2012) 11. Abbasi, A., Sarker, S., Chiang, R.H.: Big data research in information systems: toward an inclusive research agenda. J. Assoc. Inf. Syst. 17(2), 3 (2016) 12. Bachrach, Y., Kosinski, M., Graepel, T., Kohli, P., Stillwell, D.: Personality and patterns of Facebook usage. In: Proceedings of the 4th Annual ACM Web Science Conference, pp. 24– 32 (2012) 13. Kosinski, M., Stillwell, D., Graepel, T.: Private traits and attributes are predictable from digital records of human behavior. Proc. Natl. Acad. Sci. 110(15), 5802–5805 (2013) 14. Hawkings, D.I., Best, R.J., Coney, K.A.: Consumer Behavior: Building Marketing Strategy. McGraw-Hill, Boston (2010) 15. Lazer, W.: Life style concepts and marketing. In: Greyser, S.A. (ed.) Towards Scientific Marketing, pp. 130–139. American Marketing Association, Chicago (1963) 16. Wind, J.: Life style analysis: a new approach. In: Allvine, F.C. (ed.) AMA 1971 Combined Proceedings, pp. 302–305. American Marketing Association, Chicago (1971) 17. Plummer, J.T.: The concept and application of life style segmentation. J. Mark. 38, 33–37 (1974)

Do You Like According to Your Lifestyle?

139

18. Lee, H.J., Lim, H., Jolly, L.D., Lee, J.: Consumer lifestyles and adoption of high-technology products: a case of South Korea. J. Int. Consum. Mark. 21, 153–167 (2009) 19. Kucukemiroglu, O., Harcar, T., Spillan, J.E.: Market segmentation by exploring buyer lifestyle dimensions and ethnocentrism among Vietnamese consumers: an empirical study. J. Asia Pac. Bus. 7(4), 55–76 (2007) 20. Wells, W.D., Tigert, D.J.: Activities, interests and opinions. J. Advert. Res. 11(4), 27–35 (1971) 21. Facebook Inc. http://newsroom.fb.com/company-info/ 22. Foster, M., Francescucci, A., West, B.: Why users participate in online social networks. Int. J. E-Business Manag. 4(1), 3–19 (2010) 23. Zhang, Y., Pennacchiotti, M.: Predicting purchase behaviors from social media. In: Proceedings of the 22nd International Conference on World Wide Web, pp. 1521–1532 (2013)

An Online Tool to Promote Citizenship Roles in Urban Community Organizations Adrian Haurat and Li Liu ✉ (

)

Department of Computer Science, California State University Northridge, Northridge, CA 91330, USA [email protected], [email protected]

Abstract. A community does not only provide a circle where people make friends for life, it also promotes helping behaviors among members. Community members are capable and willing to contribute to community co-development because of the similar ideas and beliefs that they share. This paper introduces a community engagement tool which promote self-help activities in an organization by associating people who need help to responsible members who can offer support in an urban community setting. The online tool increases the degree of helping behavior by maximizing the effect coverage of potential contributors and using an automated online matching mechanism. The system eliminates perturbing community members who are not able to participate by using a targeted messaging method. In this paper, we will discuss how human factors are consid‐ ered during the design and implementation process when building such an inclu‐ sive tool for a diverse population of users. Keywords: Social presence · Community development · Volunteers · Involvement

1

Introduction

Many applications exist to facilitate communication between members of an organiza‐ tion or community. They try to address various issues within a group by using different combinations of technical solutions. However, an elegant solution to one frequently occurring question from a community has not be answered yet: How can a person easily and quickly find someone to help them in their daily life? When an individual realizes, they need help with some tasks – anything from moving a heavy piece of furniture to replacing a broken garbage disposal – many questions must be answered, and compli‐ cated steps must be taken to address the need. First, the person who needs help must determine whom to ask. While this may seem trivial, it is often the most difficult step due to the number of variables involved and barriers that must be overcome. One must determine whom they know that has the necessary skill to help with the task, lives in proximity, and might be available and willing to help. Then they must determine the best way to contact those individuals. If the person in need does not know of anyone who can help with the task, the problem compounds. One might resort to asking friends or using mass communication methods © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_13

An Online Tool to Promote Citizenship Roles

141

(social media post, mass email to coworkers, etc.) that have a broad reach but often do not effectively target people who are likely to help. All of this requires time and effort for the individual who needs help, while also taking time from a large number of people who must read the message to determine if it applies to them. Furthermore, people often begin to ignore mass email, and social networking messages since their high-frequency distribution causes distraction and fatigue [1]. Many workplaces discourage or forbid these kinds of mass emails since they often are not directly work-related. Posting a need on Facebook or other social networks may be ineffective depending on one’s connections and many factors out of the individual’s control such as who happens to see the message in their ever-changing stream of posts. All of these challenges beg for a solution that can efficiently connect people with needs to others who are most likely to help. Prior research suggests that individuals within an existing community are more likely to help fellow community members for altruistic reasons than they are to help total strangers. Leider et al. found that the degree to which individuals help each other has a high correlation to their “social distance” (i.e. individuals tend to help family, friends, friends of friends, and strangers in that order with decreasing frequency). The effec‐ tiveness increases when there is a possibility of future reciprocation [2]. Given this reality and the fact that the poorest members of communities tend to have the greatest needs, an ideal solution to the abovementioned problems would leverage an individual’s existing network of relationships so they can avoid having to hire people for help with tasks that could be done for free by their connections. Furthermore, to do so in a context where users will have opportunities to reciprocate will provide the highest probability of finding help when needed and produce the most substantial benefits to the overall community [3, 4]. To improve the effectiveness of a “time banking” model where users earn “currency” by spending time assisting other users, a community engagement tool should [5], 1. Minimize time and effort on the part of the requester. 2. Be easy to access, be simple to use for a wide variety of users with varying levels of technological experience. 3. Be versatile to handle various kinds of requests. Gilbert et al. suggests that people living in urban is likely to interact with someone else in a community even over a weak tie [6] in their CHI publication. However, our survey with some community leaders in the Greater Los Angeles area shows that community members are feeling too busy for one another because of professional lives, traffic and other stresses living in a city [7, 8]. They also expressed the need to increase interactions between community members. Both members and the community will benefit from self-help activities within the group [9–11]. What we found is missing is a tool which facilitate these self-help activities. An optimal solution would not only accomplish objectives in [5] but also inform individuals who are likely able and willing to help without bothering those who are not. In this paper, a web-based community engagement system named Helpfull is introduced that meets these needs through an automated targeted messaging system. Empirical testing shows that the tool promotes helping behavior among members of a community by connecting who seek help to others who can provide it without using monetized rewards.

142

2

A. Haurat and L. Liu

Related Work

This section reviews existing applications which attempted to provide assistance to the members of a community promptly by comparing their strengths and weaknesses regarding types of services (paid vs. free), volunteer recruitment, and community engagement. There are four popular online/mobile applications in this domain: TaskRabbit, GiveGab, The City, and NextDoor. TaskRabbit1 helps users find “Taskers” (paid workers) whom they can hire to complete various categories of tasks. TaskRabbit provides access to its services through their web application as well as through native apps for iOS and Android devices. This app matches users with paid service providers for certain categories of tasks such as cleaning, repairs, moving, event planning, or running errands. For users with disposable income who are willing to pay for these services, TaskRabbit provides a solution. If help is needed right away, users can allow the app to message Taskers to see who is available and thus allow the user to get help in a relatively short amount of time. If the task does not have to be completed right away, the user interface provides information to help users select a Tasker such as how many tasks each provider has completed, how often they received a positive review, and what they charge per hour of work. This allows users to select a Tasker based on their chosen priorities (experience, cost, and/or positive reviews). Tasker’s hourly rates vary for different task categories, but an approximate range is $25 to $60 per hour for services (with running errands being the least expensive and home repair being the most expensive) as of 2017. TaskRabbit’s availability is currently limited to certain metropolitan areas because it relies on having a large pool of nearby service providers. Consequently, persons in less populated areas currently cannot utilize the service. TaskRabbit has a defined list of request categories that are included in their service and thus is not able to fulfill requests that fall outside of those categories. GiveGab2 is a web application that exists to meet needs in a different way: by lever‐ aging existing nonprofit organizations and giving them a communication channel with potential volunteers. Since many organizations exist that seek to meet the needs of communities through philanthropic activities, GiveGab has created a platform that allows those organizations to reach out for help with funding and volunteering from a pool of willing individual app users who specify the kinds of causes for which they are willing to donate and/or volunteer. This model provides a way for organizations to communicate needs with users and provides those organizations with the ability to track individuals’ giving (monetary gifts and volunteer hours). It also enables various means of communication within the app as well as social media integration to spread the word about various needs. While GiveGab and other similar volunteering-focused apps provide a way for organizations to let motivated individuals know when needs arise so they can decide whether or not to help out, they do not provide a way for app users to communicate their needs. In other words, they focus on providing tools that are primarily

1 2

www.taskrabbit.com. www.givegab.com.

An Online Tool to Promote Citizenship Roles

143

useful to organizations seeking help and users who are looking for opportunities to help, as opposed to being a tool for the individuals who are seeking help. Unlike TaskRabbit’s paid provider model or GiveGab’s organizational outreach model, The City3 and NextDoor4 use variations of a community-based model. The City provides a social networking platform (similar to Facebook) targeted for individual religious communities via a web application and native apps for Android and iOS devices. Through the platform, users have a private online connection to members of their community and can send out messages to the whole community or sub-commun‐ ities. This provides a way for users to share needs and requests, although the commu‐ nication cannot be specifically targeted toward members of the community based on factors such as having a particular skill or being available at the requested time. Essen‐ tially, it provides a way to communicate en masse with the entire community or a chosen sub-community. Users of The City do have the option to specify skills in their user profile, but viewing and/or messaging members of the community with a particular skill is a feature only available to administrators. In other words, an administrator could view and send a message to any or all community members who have a particular skill (e.g. photography) if a need arises for a photographer. This feature facilitates finding members to meet needs of the religious organization itself but does not provide the majority of users with a method of targeting messaging to members who might be able to help them with particular tasks. To connect with an individual who might have a particular skill, the administrator must actively engage in the communication and manually determine who is the best match for a given need through directly communicating with a potentially large group of individuals, which might require significant time and coordination. Overall, The City provides general functionality for communication, community engagement, and event planning within a religious organization, but it is not tailored for use cases where an individual is reaching out for help. NextDoor creates an online community for persons who live near to one another and gives neighbors a means to communicate with each other through their web application and iOS/Android apps. Members are required to verify their residence in a particular locale to be admitted into the online community, so users can have a relatively high degree of confidence that the other members of their community within the app do, in fact, live in their same neighborhood. NextDoor allows users to post messages about a wide variety of topics ranging from yard sale announcements to school fundraisers, wildlife sightings, lost and found items and crime alerts. The format is somewhat like a forum or bulletin board in the sense that anything posted to the community page can be seen by everyone within the community, with the additional ability for users to add comments. The fact that NextDoor connects people who live within proximity of one another (usually within a 2-mile radius or so) makes it useful for certain kinds of needs and requests. For someone making a request, the platform gives the individual a broader reach than they might have with other tools because it allows them to communicate with strangers outside of their usual social circles. Also, users who ask for help from other

3 4

www.acstechnologies.com. www.nextdoor.com.

144

A. Haurat and L. Liu

members could conceivably get help promptly given the proximity of the community members. However, because community members may not know each other or may not have strong relational ties, there may be little motivation for users to help with tasks that require significant effort. While a neighbor may be willing to give someone from their neighborhood a cup of sugar or lend them a hammer, they may not be ready to help them install a garbage disposal for example (at least not for free). Morever, NextDoor’s connection between persons of loose ties but proximity is both the app’s primary feature and its limitation. While users can post messages to the whole community, or to hand‐ picked individuals, there is no automated way to target all neighbors who are, for example, photographers or gardening experts. The exception to this limitation is that users can create community subgroups based on particular areas of interest, but these are opt-in, and a user has to seek actively and join such a group. So if for example, a community happens to have a “photography” community group and a user needs a photographer, they could temporarily join that group and send a message to its members. Although this feature cannot reach all photographers in the larger community (only the ones who found and joined the subgroup), it does provide a platform for communication that relates to specific activities and areas of interest. In summary, existing applications provide mechanisms for communication of needs and requests or getting help with particular tasks. Some are focused on finding suitable service providers who can be hired to complete a task. Others provide ways for organ‐ izations to reach out for help from willing volunteers. Many apps generally focus on the issue of connecting members of a community and providing various means of commu‐ nication that can be used for a wide variety of purposes. Lastly, some existing apps give neighbors a way to communicate about community-specific issues and foster ties within geographical communities. Each of these models has strengths that make them suitable for particular use cases. However, a solution that focuses on enabling a person to quickly and simply connect with members of their community who are highly likely to be willing and available does not yet exist. A solution that provides these connections without users having to pay for services would be useful to a broad range of users.

3

System Architecture

Helpful enters a space where some other similar tools exist, but it has many key differ‐ entiators which make it optimal for connecting members of communities with others in their community who can provide the help they need. A Model-View-Controller (MVC) design pattern is selected to support the development of the system. MVC allows the view/presentation layer modules to decouple from the controllers (application logic) and the models (application data structures, relationships, and database interactions) to maximize reusability and extensibility of modules of the system. MVC also enables Helpfull to be accessible across numerous devices with various operating systems suggesting a mobile-friendly web application users could interact with any web browser. A database provides the necessary mechanism for persisting data related to user profiles, community settings, and help requests. Server-side logic and processing user requests

An Online Tool to Promote Citizenship Roles

145

is handled by the web application module to route requests appropriately, persist data, send messages, and provide useful views to the user. A time-based job scheduler handles the periodic checks for new providers and sending messages for pending requests. The system is designed to be deployed on a cloud-based server via Amazon Web Services. Figure 1 shows the system architecture.

Fig. 1. System architecture

3.1 Community Model The system is architected around the concepts of a community-based model. A community-based model begins with the idea that users register for an account that is tied to an existing organization or “closed community”. In this context, a closed community specifically refers to an organization with clearly defined members in who have some relationship to each other as part of an organization (such as a school, work‐ place, club, or faith community). The system connects members of a closed community and provides means to share and fulfill requests for help of various kinds. Each community has one or more administrators who are able to add or approve new members within the app. These administrators are also responsible for establishing communitywide settings that tailor the app’s behavior to the specific needs of their community. For example, administrators can create custom request types that are relevant to their organ‐ ization (e.g. Tutoring requests within a school community) and/or disable certain request types that may not be suitable for their organization (e.g. ride-sharing, or childcare requests might be disabled for organizations concerned about liability).

146

A. Haurat and L. Liu

3.2 Request Process When users register for the application, they enter their profile information including contact information, request types for which the user wishes to receive notifications, preferred locations for fulfilling help requests, and limits to the number of incoming requests. In this way, the system knows whether to contact a given individual via email and/or text message, which locations and request types are relevant for that user, and whether the user is available to receive requests at a given time. Because each user’s profile contains this information, the App can handle request messaging in an automated way that does not require any work for users making a request beyond submitting it to the system. Below is a diagram of how a request flows through the App when a user submits it (Fig. 2).

Fig. 2. User request flow

To request help, users select a category of task/request and submit the request to the app without manually selecting a particular “service provider”. A service provider (or, for brevity, “provider”) is a user who responds to - and potentially fulfills - a request for help based on skills in their user profile, thus, • Helpfull is “user to user” in that it connects users to each other directly rather than to paid service providers employed by a company (as TaskRabbit does), or to a nonprofit (as GiveGab does), or to a community administrator.

An Online Tool to Promote Citizenship Roles

147

• Requesters and service providers know each other outside the context of the App. In this way, Helpfull is meant to facilitate the enhancement of existing communities and relationships. In the case of The City, users know each other but do not have a way to connect with skill-based and/or interest-based service opportunities without involving administrators in request fulfillment and messaging. Users of The City also have the option of sending a request out to members of their entire group at once, but this is essentially the same as a mass email, and it lacks the automation and targeted messaging that Helpfull provides. • Targeted messaging. Users submit and receive requests that are relevant to them and do not have to browse through potential service providers or lists of volunteer oppor‐ tunities to help or be helped. Minimizing friction for users during the process of making or fulfilling a request for help is a central focus of Helpfull’s design to promote the maximum use and adoption by community members. • Users engage with the app both as requesters and providers. The functionality is as much about helping others as it is about having one’s own needs met. By making users aware of opportunities they might otherwise not have known about, the app enables and challenges users to help members of their community in a novel and engaging way.

4

System Evaluation and Analysis

The system is deployed on a Ubuntu 14.04 server hosted through Amazon Web Services. A Google Analytics script tag was added to all application screens to track page views and data regarding users’ web browsers and operating systems. Additionally, a custom GA timing event was added on the request form to detect how long users spent making requests. All other analysis would be performed on information gathered directly from a request and message data collected in the database of Helpfull. A total of 32 volunteers living in the Greater Los Angeles area signed up for the trial use of the system. They were split between two virtual communities (named Gryffindor and Slytherin), each with two members designated to be community administrators who would create custom request types. In the case of Gryffindor, which had the user management setting set to “manual”, administrators were also responsible for approving users’ requests to join the community. Quantitative usage data of the test run is collected by a tracking form placed at the bottom of each app screen. A question survey is admin‐ istered after the test run to gather qualitative feedback. The test was conducted over the course of 38 h. Users were asked to register for the app within the first 10 h and submit fictional requests for help throughout the test run. They were also asked to respond to requests for help from other users – sometimes accepting, sometimes declining, and at other times ignoring the requests. This behavior was intended to simulate a level of activity which might normally occur over a period of several days, weeks, or even months in a typical community. Users were instructed to choose a subset of request types from home repair, cooking, cleaning, tech help, childcare, elder car, event setup, trans‐ portation, hardwork tutoring and other. Then, users are asked to provide a radius of service area for which they were willing to be notified if there is a request. We measured

148

A. Haurat and L. Liu

request submission time and request response time to evaluate the interaction between requesters and providers. Figure 3 shows 192 requests submitted in total (124 in Gryffindor and 68 in Slytherin). 84.53% of requests were reviewed and accepted in less than 120 s. 56.88% were reviewed and accepted in less than 60 s. The Average time spent submitting the request form was 89.52 s. A total of 104 requests were eventually accepted by other users (54% fulfillment rate). Of the 175 requests where a matching provider was found and contacted, 110 requests received a response of either “yes” or “no” (63% response rate). 88 requests were not fulfilled and 17 of these requests had no matching providers when initially submitted. 6 requests were made where all matching providers declined. One third, or 65 requests, never received a response during the trial period.

Fig. 3. Request submission time breakdown (provided by Google Analytics)

Figure 4 shows the response times for the 104 requests that were accepted during the test. Response time was measured from the time a request was submitted to the time it was accepted by a provider after reviewing the request. A service provider uses this time to gather his or her resource before making the commitment of fulfilling the request. 61 of the requests (59%) were accepted in under 1 h. 73 of the requests (70%) were accepted in under 2 h. As the response rate leveled out, 81 of the requests (78%) were accepted in under 3 h. For accepted requests, the average time from submission to

An Online Tool to Promote Citizenship Roles

149

acceptance was 145 min (2.4 h) with a standard deviation of 284.23 min. This high standard deviation was primarily due to a few outliers visible throughout the testing period.

Fig. 4. Request response time

This data reveals that the vast majority of app users were able to ask for help in under 5 min, and on average receive a response from a provider within 2–3 h, which is a promising result. Average interactions per user is 11.3 (5.8 requests submitted and 5.5 responses given). The request submission duration and response time could even be significantly shorter with more users when messages are automatically sent to new providers every 20 min, and the pool of potential providers is larger. Figure 5 is the result of usability survey conducted after the testing period.

Fig. 5. Subjective rating of usability

150

5

A. Haurat and L. Liu

Conclusions

Community members are capable and willing to contribute to virtual co-development of their living environment [9, 10]. Such contributions relieve an individual’s weight of sole responsibility in a community and make members stake in the community as [8]. In this paper, we report an online tool which enables members to play their citizenship role, which is the common thread that connects all people in urban communities. Helpfull addresses the concerns to meet day to day needs of individuals and encourages community members to help each other to foster a sense of community rather than to receive financial compensation. It is architected in a fashion that it could easily be extended to other problem spaces. The system is designed to address the common needs of such a tool including accessibility, simplicity, minimal effort for request submission, the maximum possibility of request fulfillment, user privacy, and reusability. It is capable of producing results in real-time based on the pilot study. The system could be utilized in various contexts, where users are tagged with various characteristics and are notified automatically. The current iteration of the system dramatically increases the extent to which members of communities assist each other and increase community engagement in a metropolitan area.

References 1. Romero, D.M., Galuba, W., Asur, S., Huberman, B.A.: Influence and passivity in social media. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp. 18–33 (2011) 2. Leider, S., Markus, M.M., Rosenblat, T., Do, Q.: Directed altruism and enforced reciprocity in social networks. J. Econ. 124(4), 1815–1851 (2009) 3. Hampton, K.N.: Persistent and pervasive community. Am. Behav. Sci. 60(1), 101–124 (2016) 4. Ren, Y., Kraut, R., Kiesler, S.: Applying common identity and bond theory to design of online communities. Organ. Stud. 28(3), 377–408 (2007) 5. Bellotti, V., Carroll, J.M., Han, K.: Random acts of kindness: the intelligent and context-aware future of reciprocal altruism and community collaboration. In: 2013 International Conference on Collaboration Technologies and Systems (CTS), pp. 1–12. IEEE Press, New York (2013) 6. Gilbert, E., Karahalios, K.: Predicting tie strength with social media. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 211–220. ACM, 4 April 2009 7. Partecke, J., Schwabl, I., Gwinner, E.: Stress and the city: urbanization and its effects on the stress physiology in European blackbirds. Ecology 87(8), 1945–1952 (2006) 8. Rose, N.: Community, citizenship, and the third way. Am. Behav. Sci. 43(9), 1395–1411 (2000) 9. Foster-Fishman, P.G., Berkowitz, S.L., Lounsbury, D.W., Jacobson, S., Allen, N.A.: Building collaborative capacity in community coalitions: a review and integrative framework. Am. J. Commun. Psychol. 29(2), 241–261 (2001) 10. Maton, K.I., Salem, D.A.: Organizational characteristics of empowering community settings: a multiple case study approach. Am. J. Commun. Psychol. 23(5), 631–656 (1995) 11. Seyfang, G.: Time banks: rewarding community self-help in the inner city? Commun. Dev. J. 39(1), 62–71 (2004)

Service Science Education and T-Shaped Professionals

Factors Influencing Academic Success for Design Students: A Study of Curricular Expectations and Ethical Issues Debra Satterfield ✉ (

)

California State University Long Beach, Long Beach, CA, USA [email protected]

Abstract. Federal legislation, America’s College Promise, urges community colleges and 4-year universities to comply with guidelines for college graduation rates and improved academic success outcomes as indicated in this bill. This study examines these student success factors. A pilot study of 22 students in a design degree program at a state university in a major metropolitan area was conducted to identify non-cognitive factors that impact academic success and graduation rates. A literature review was also conducted to examine the impact of financial aid, social situations, transfers between two-year and 4-year institutions and perceived satisfaction with course content and expectations of the role of the professor in the educational process. This research will be used to inform the second phase of research examining course content delivery methods and valueadded strategies for improving education for students in design majors and in urban university settings based on reported satisfaction and demographic survey data. Keywords: Design education · Ethics · Academic success · User experience design · Service design · America’s college promise · California SB-412

1

Introduction

University students are faced with many factors that determine their academic success. Specifically, society changes such as urban environments, financial situations, and other non-cognitive factors contribute greatly to the academic success of students. For students in design majors, additional factors such as creative thinking processes, the cost of materials, long hours in studio, lack of computer labs or high tech fabrication equipment, and high-stress studio reviews also contribute to the non-cognitive burdens faced by students in these majors. Therefore, a phase one pilot study, “Career Goals and Educa‐ tional Preferences,” was given to senior design majors in a 4-year, state university in a major metropolitan area to identify curricular preferences, as well as non-cognitive barriers to academic success. The first phase of this research was used to identify student perceptions about careers, skills, curricular and learning preferences and demographic data relating to age, gender, and hours worked per week. Student responses to this pilot study and the implicit ethical issues are identified and future research discussed with regard to improving graduation rates and academic success as mandated by the federal legislation, America’s College Promise. © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_14

154

2

D. Satterfield

Non-cognitive Factors that Influence Academic Success

According to the US Department of Education report, “Fact Sheet: Focusing Higher Education on Student Success,” nearly half of students who enter college do not complete a degree within a six-year time frame. In addition, the financial burden of huge student loans that do not translate into a college degree cause these students to default at three times the rate of students who do graduate. Therefore, focusing on less time in school, higher graduation rates, and lower education-related debt are imperative to success in higher education [1]. A student’s ability to perform adequately in progressing towards the completion of a college degree can be affected by a wide variety of factors including both cognitive and non-cognitive variables. Non-cognitive factors are those things that are not related to ability or intelligence, yet can inhibit student success in academia. These factors may be related to resilience, hope, positive attitude, finances, long commute times, heavy workloads, family responsibilities, and mental or physical health issues. In Flow: The Psychology of Optimal Experience, two main strategies are presented that improve quality of life. The first is to make external conditions match our goals. The second is to change how experience external conditions are perceived and make them fit existing goals better [2]. Csikszentmihalyi’s theory of flow explains the concept behind the academic success strategy referred to as “hope.” Hope is a non-cognitive, problem-focused, coping factor that is measured by a student’s ability to generate multiples strategies to achieve a goal and to view barriers as challenges rather than obstacles. Hope is the ability to see more than one strategy or pathway to success. The factor of hope also gives students a sense of control. Optimism, in contrast to hope, is the belief that good things will happen, but is not attached to belief in pathways, goals, or strategic problem solving, all of which are critical to academic success. Two types of students were also identified, those with learning goals and those with performance goals. Students with learning goals were more actively engaged in their education. Students with performance goals were less committed to changing strategies when faced with barriers and were more inclined to take the easier pathways that required less effort to achieve results [3].

3

Transfer Students

While the barriers differed for freshman, sophomore and upper division students, first year transfer student success can also be effectively predicted with non-cognitive factors. Transfer students may have few transfer credits or complete two-year degrees, they may also be married or supporting families, working part or full time, and may be older than traditional students. In addition, they transfer for reasons such as financial, grades, personal growth, degree completion, or fulfilling career aspirations. A spring 2002 university study measuring the academic performance and persistence of 369 transfer students indicated that at the end of the first year 26% were failing to perform up to success measures as indicated by GPA and 21% failed to enroll for the following

Factors Influencing Academic Success for Design Students

155

semester. In this study, for all levels of students, non-cognitive factors were the best predictors of academic success [4]. Low-income and first-generation transfer students have additional barriers to success. A 2008 Pell Institute report states that after six years in school only 11% of low income first generation students had earned a college degree compared to 55% of their peers. It also stated that first generation students that started their college education at a 4-year institution were 7 times more likely to complete their degree than those who started at two-year institutions. However, only 25% of first generation college students started at 4-year institutions and the remainder who didn’t often didn’t transfer to complete their degrees. In addition, when entering two-year colleges, 63% of lowincome first generation students indicated their intention to ultimately graduate with a 4-year degree. However, 6 years later only 5% of these students had in fact completed their 4-year degrees [5].

4

Financial Conditions

Financial aid comes in the form of grants, work-study, and loans. Grants do not have to be repaid and students must qualify for a set amount of work study money that is then paid for jobs on campus and applied toward a student’s college tuition. College loans vary in their interest rates and do have to be repaid upon leaving school. In order to determine eligibility for each of these types of financial assistance, a student must compete the “Free Application for Federal Student Aid (FAFSA)” at fafsa.gov [6]. According to a 2008 Pell report, a student financial aid Pell grant in 1980 covered 77% of costs of attending a 4-year institution. By 2008, this same grant had been reduced to cover only 38% of these soaring university cost. In addition, only 37% of grant aid went to low income, first generation students with 61% going to students without any demonstrated financial need. This was due in part to the fact that low income, first generation students tend to enroll in low cost institutions, 2-year institutions, or forprofit universities all of which have less financial aid available. This left many low income, first generation students with financial need in a position of having less financial aid available, increasing costs over previous years, and higher debt loads. These factors all contribute to lower rates of transfer or degree completion for these students. In 2008, for students who left after only one year of college had on average over $6,500 of debt. Students who left without a degree after 4 years of college had an average loan debt of over $16,500. They faced the increased financial difficulty of paying back their loans without the extra earning advantage of better paying jobs based on a degree [7]. Even after major investments in the Pell Grant program by the federal government in 2015, current Pell grants still only cover about 30% of the costs of a 4-year degree. Addition‐ ally, in the 2015 report more than 40% of students enrolled in 4-year college programs do not finish their degrees after 6 years. For these students who do not complete their degrees, the default rates on loan repayment are three times higher than for those who do complete their degrees [8].

156

5

D. Satterfield

America’s College Promise

In 2015, President Obama announced the America’s College Promise (ACP) Grants. This $100 million dollar grant competition program offered free tuition for those students attending one of the participating two-year community colleges. The program specifically detailed its intention to serve all students including those in low income and first generation categories. Also since its inception in 2009, federal government programs have invested $66 billion dollars to fund 19 million Pell grants for students in community college programs [9]. The ACP shifts federal funding emphasis onto two-year community college programs. Specifically, its main targets for funding were those community college programs with low costs and an emphasis on outcomes in the job market [8]. California Senate Bill-412, a state level companion to the federal ACP bill, reads as follows: “This bill would establish the California Promise, which would require specified minimum numbers of campuses of the California State University to establish a California Promise program by which the campus would enter into a pledge with a student who satis‐ fies specified criteria to support the student in earning a baccalaureate degree within 4 academic years, or if the student is a community college transfer student who earned an associate degree for transfer, within 2 academic years, of the academic year of the student’s first year of enrollment, as specified. The bill would require the trustees to submit, by July 1, 2021, a report to the appropriate policy and fiscal committees of the Legislature that includes specified information about students who participate in the program and a summary description of significant differences in implementation of the program by campuses. The bill would require the trustees to submit recommendations, by March 15, 2017, to the appro‐ priate policy and fiscal committees of the Legislature regarding potential financial incen‐ tives that could benefit students who participate in the program. The bill’s provisions would be repealed as of January 1, 2026” [10].

6

Factors Impacting Academic Success in Design Education

Creative people have unique traits and personalities. According to Csikszentmihalyi, crea‐ tive people experience the world in novel and original ways. They appear to be unusually bright and naturally express themselves in creative domains. The concept of creativity is defined as any act, idea, or product that fundamentally changes or transforms an existing domain into a new one [11].

7

Design Thinking and Creativity

Creative people are by definition agents of change. Experts in creative domains have exten‐ sive knowledge about how they organize, represent and interpret their environment. The key elements of this unique learning style include: 1. Having the ability to notice meaningful patterns in information; 2. Possessing a large amount of content knowledge and deep learning;

Factors Influencing Academic Success for Design Students

3. 4. 5. 6.

157

Having the ability to apply their knowledge; Able to retrieve important information with little additional effort; May not be able to transfer their deep knowledge to others; and Varying in their ability to be flexible in novel situations [12].

Creative people are often times both emotional and analytical in how they think and solve problems. This unique method of problem solving is often referred to as “design thinking.” With regard to complex problem solving, research in neuroscience says that cognition actually includes thinking, reasoning, intelligence and emotion. Therefore, cogni‐ tion and emotion can both be considered unconscious processes and are equal in an allencompassing view of the mind. One definition of cognition expands on this by saying that emotions can be considered reasoned thoughts about a situation. In terms of neural processing speed, the emotional meaning of a stimulus can begin to be applied by the brain before the cognitive systems have finished processing [13]. Therefore, creativity as it is part of design thinking can be interpreted as a cognitive style with unique methods of problem solving in the areas of attentiveness, memory, representation, cognitive images, metaphors, and diagrams [14].

8

Methods

A thorough literature review of non-cognitive factors influencing academic success was conducted. In addition, an IRB approved pilot study (PROJECT TITLE: [860934-2] Career Goals and Educational Preferences Survey, REFERENCE #: 16-232) was conducted in a senior level design course during Fall 2016. All respondents (n = 22) were majors in the Bachelor of Arts (BA) in Design degree program in a 4-year state university. The study was conducted in an on-campus section of a senior level design studio class. The students repre‐ sented included 7 males and 15 females and with ages ranging from 21–29 years old. The outcomes of this pilot study were used to inform and create a larger study on non-cognitive factors influencing academic success and timely graduation rates in design degree programs. This study will be conducted in 2017.

9

Discussion

As a newer degree program in a design department in a college of the arts, the career goals and curricular preferences of students in this major are not well understood. Therefore, a pilot study using a combination of qualitative and quantitative demographic questions was created to better understand career preferences and barriers to timely graduation and academic success. It was designed as a first phase to a larger study to be conducted in Spring 2017 that will examine additional non-cognitive barriers and curricular preferences in this program. The phase 1 research was mixed methods pilot survey that included the following qualitative response questions: 1. If you could have any career or job, what would it be and where? 2. What are your strongest design, business and people skills that will prepare you for your career?

158

D. Satterfield

3. What skills do you still need to develop to get the great job that you want in this field? 4. What can you learn from this class to help you in your career? How can I help you gain this information? 5. What was the best class experience that you have had in college? 6. What made it the best class experience? 7. Describe the best professor, counselor or mentor that you have had in college and what made that person effective and excellent in helping you? 8. What can I do to help you achieve your goals and have a great experience in this class? How can I be a great professor and mentor? Additional demographic data was collected on age, gender, ethnicity, university class status, in-state versus out-of-state status, hours worked on campus and hours worked off campus.

10

Qualitative Survey Responses

Questions 1–4 were based on a personal assessment of job preferences, perceived strengths and skills need to prepare for a future career. These questions were focused on how students view their degree in terms of what they want it to prepare them for and how well the curriculum is doing in preparation for their future careers. Responses (n = 20) to question 1 varied greatly. Also because it was a qualitative question, many students gave more than one response. The following responses given and numbers of responses to question 1 about career aspirations were: 1. 2. 3. 4. 5. 6. 7.

Spatial, Environmental or Set Design-5; Hospitality or Entertainment Design-4; UX or Service Design-4; Product or Furniture Design-3; Game Design-3; Interior Design or Architecture-3; and Graphic Design or General Design-2.

Questions 5–6 were based on a personal assessment of what makes a best class situation. These were also open ended and allowed students to give multiple, free-form qualitative responses to the questions. Responses to question 6 (n = 17) were as follows: 1. 2. 3. 4.

The course gave “real world” knowledge-10; The student had respect or a connection with the professor-4; The course pushed the student to higher levels of learning -3; and The student had positive social friendships in the class-2.

Questions 7–8 were based on an assessment of the qualities of an ideal professor. Responses to question 7 (n = 14) were: 1. The professor encouraged students or showed a genuine concern for helping them-6; 2. The professor answered questions and gave thorough information or directions-6;

Factors Influencing Academic Success for Design Students

3. 4. 5. 6.

11

159

The professor pushed students to achieve their best work-5; The professor showed an open-mindedness-4; The professor was passionate about the subject of the class-1; and The professor made the course fun or enjoyable to students-1.

Demographic Survey Data

Of the students surveyed (n = 22), 15 were female and 7 male. On ethnicity (n = 20), those responding reported: 1-African American, 4-Asian, 10 Hispanic, and 5 Caucasian. Age (n = 22) reported 3–20 to 21, 8–22 to 24, and 11 age 25 and above. On residence, 1 student reported out of state status and 22 reported in state. Off campus work hours (n = 19) were reported at 5 who don’t work, 2 worked 1–10 h per week, 3 worked 11–20, 6 worked 21– 30 h, 3 worked more than 30 h per week. On-campus work (n = 20) was only indicated by 1 student at 11–20 h per week and 2 students at 21–30 h on campus. However, 17 students indicated that they had no work opportunities on campus.

12

Findings and Implications

The pilot study was useful in identifying specific characteristics about students in this design major. Of the 22 students surveyed, 68% were female and 50% were 25 years and older. In this particular degree, students may be more likely to have changed majors and be gender-biased towards females. Some degrees like interior design are widely known to have more females and industrial design often has more males, but the degree program being studied is more ambiguous in its nature so less is known about its age and gender tendencies. In addition, 50% of the students indicated they worked 20 h or more per week and 85% worked entirely off campus. This may indicate that older students have less financial aid available to them and are forced to work in a pay-as-you go fashion as they enter the upper division years of study. In addition, 68% of the students in the pilot study self identified as ethnic minorities, 45% indicating Hispanic and only 22% indicating Caucasian as their ethnicity.

13

Conclusions and Areas for Future Study

Phase two of this research will use the pilot study data to collect more specific infor‐ mation about the daily lives and academic pathways of these students. This will be done to examine how non-cognitive factors influence the academic success and time to grad‐ uation for students in this design major. Specifically, future study will be done to deter‐ mine if these students began their academic career at this university or if they came in as transfer students. Additional study will also be done to determine how many years they have spent in college, average commute times and other factors such as design thinking, availability of computer labs and work spaces on campus, parking, long or difficult commuting, and other physical conditions on campus that might affect their attendance and impact academic success.

160

D. Satterfield

From the literature review, it can be determined that the availability of funding for low income and first generation students is critical to their success. With only 37% of financial aid being targeted to low income students and the larger debt load they carry at gradation, this may have ethical implications with regard to how this impacts their overall academic success by possibly forcing them to take on outside work at levels that negatively impact their academic success and time to graduation beyond that of their peers with less financial need. In addition, non-cognitive factors such as starting at a community college, the concept of hope, and connections with a social group during the first year are critical to academic success. Other factors such as those indicated by students in the qualitative survey indicate that they value real world knowledge and career related content. They also indicated that the relationship with their professors and the classroom environment are very important to their academic success and perceived satisfaction with their educational experience. Specifically they indicated that they prefer to have encouragement from their professors, get questions answered, working with open-minded educators, and to be pushed to do their best work. An underlying ethical question is whether it is acceptable to encourage low income first generation students to begin academic careers in two-year colleges even when they indicate a desire to complete a 4 year degree and research suggests that this pathway has only a 5% success rate with regard to earning that degree. In addition, is it ethical to dismiss the non-cognitive barriers to academic success and timely graduation as simply the student’s problem? If transferring to another university, changing a major, and finding little or no support for demographic or cultural challenges are known to nega‐ tively impact gradation rates, does the university have some responsibility to track these indicators and intervene for better academic outcomes? The second phase of this research will examine alternative teaching methods such as hybrid and online course content delivery as possible ways to mitigate some of the non-cognitive factors that may be contributing to problems faced by transfer students and upper division students.

References 1. Fact Sheet: Focusing Higher Education on Student Success, July 2015. https://www.ed.gov/ news/press-releases/fact-sheet-focusing-higher-education-student-success. Accessed 21 Nov 2016 2. Csikszentmihalyi, M.: Flow: The Psychology of Optimal Experience, p. 43. Harper & Row, New York (1990) 3. Snyder, C.R., Shorey, H.S., Cheavens, J., Pulvers, K.M., Adams, V.H., Wiklund, C.: Hope and academic success in college. J. Educ. Psychol. 94(4), 820–826 (2002). doi: 10.1037//0022-0663.94.4.820 4. Duggan, M.H., Pickering, J.W.: Barriers to transfer student academic success and retention. J. Coll. Stud. Retent. Res. Theor. Pract. 9(4), 437–459 (2007). doi:10.2190/cs.9.4.c 5. Engle, J., Tinto, V.: Moving Beyond Access: College Success for Low-Income, FirstGeneration Students, pp. 2–3. Pell Institute for the Study of Opportunity in Higher Education, Washington, D.C (2008) 6. Federal Student Aid at a Glance (2016). https://studentaid.ed.gov/sa/sites/default/files/aidglance-2017-18.pdf 7. Engle, J., Tinto, V.: pp. 23–24 (2008)

Factors Influencing Academic Success for Design Students

161

8. Fact Sheet: Focusing Higher Education on Student Success, July 2015 9. Fact Sheet: White House Launches New $100 Million Competition to Expand Tuition-Free Community College Programs that Connect Americans to In-Demand Jobs (2016). https:// www.whitehouse.gov/the-press-office/2016/04/25/fact-sheet-white-house-launches-new-100million-competition-expand. Accessed 22 Nov 2016 10. SB-412 Public postsecondary education: The California Promise. https://leginfo.legislature.ca.gov/ faces/billTextClient.xhtml?bill_id=201520160SB412. Accessed 22 Nov 2016 11. Csikszentmihalyi, M.: Creativity: Flow and the Psychology of Discovery and Invention, pp. 25–29. Harper Collins, New York (1996) 12. Bransford, J.: How People Learn: Brain, Mind, Experience, and School, p. 31. National Academy Press, Washington, D.C (2000) 13. LeDoux, J.E.: The Emotional Brain: The Mysterious Underpinnings of Emotional Life, pp. 68–69. Simon & Schuster, New York (1996) 14. Kaufman, J.C., Sternberg, R.J.: The International Handbook of Creativity, p. 77. Cambridge University Press, Cambridge (2006)

Humanity-Embedded Service Science Education Fu-Ren Lin ✉ (

)

Institute of Service Science, National Tsing Hua University, Hsinchu City 300, Taiwan [email protected]

Abstract. The interdisciplinary studies of Service Science have been well recog‐ nized by academics and industries since it was established in 2004–2005. The Institute of Service Science (ISS) in National Tsing Hua University (Taiwan) established in 2008 is the first institute dedicated for Service Science research and education in Taiwan. The master and doctoral programs were designed for inter‐ disciplinary studies of Service Science composed of service management, design, and engineering embedded with humanity as the foundation. This article shares the structure of ISS curriculum and the innovated programs for cultivating Service Science professionals with humanitarian capacity, and the promotion of Service Science by joining the efforts of Service Science Society of Taiwan sponsored by Sayling Wen Cultural and Educational Foundation. The article ends with the reflection and future directions of developing Service Science discipline in Taiwan. Keywords: Service science · Humanity-embedded education · Service management · Service design · Information technology · Internship

1

Introduction

The interdisciplinary studies of Service Science have been well recognized by academics and industries since it was established in 2004–2005. The cultivation of talents in Service Science has been implemented in various practices, such as dedicated institutes or departments, programs in existing institutes or departments, integrated programs across colleges or departments, etc. Among them, there is one common objective; that is, to bring domain knowledge from various existing disciplines to shape up the emerging interdisciplinary field: Service Science. Service Science is an umbrella term to denote the integration of disciples. Originally it was called SSME (Service Science, Management, and Engineering), and later called SSMED or SSMEA (Service Science, Management, Engineering, Design/Arts) [1]. Service Science is defined as a science of studying value co-creation of service systems, in which a service system is defined as a configuration of people, technology, organi‐ zations (internal and external) connected by business processes to achieve good quality of life for societies. Thus, the concept of value co-creation and service system entities are fundamental to Service Science [2].

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_15

Humanity-Embedded Service Science Education

163

Then, Service-Dominant Logic (S-DL) [2] laid the foundation for the study of value co-creation and service systems. The very first axiom and foundational promise of SDL is “service is the fundamental basis of exchange” in which service is the value cocreation phenomenon by exercising operant resources in service systems. Since the customer is always a co-creator of value (axiom 2 and also foundational promise 6 of S-DL), there is no restricted differentiation between provider and customer with S-DL. Instead, all participants of a service system are called actors, and a service system is actor-to-actor (A2A) networks that are configured to co-create value to benefit each other as a whole. Thus, foundational promise 7 of S-DL defines the role of an enterprise in value creation as follows: “the enterprise cannot deliver value, but can only offer value propositions.” Axiom 3 and also the foundational promise 9 of S-DL is that “all economic and social actors are resource integrators,” extending the role playing for entities in service systems, which creates a new perspective for boosting up service innovation. Taking four categories of academic resources from existing disciplines: business and organization, technology, people, and information, at least 19 domains in academics, such as behavioral science, economics and law, human resource management, strategy, sociology and anthropology, marketing, engineering, information and communication technologies, etc., are highly related to emerging Service Science discipline [1]. Thus, it is a choice for any institute to decide the composition of academic domain knowledge in their education systems for cultivating Service Science talents. In this article, I would like to share a case that a post-graduate institute, the Institute of Service Science (ISS), National Tsing Hua University, dedicated to educate Service Science talents by inte‐ grating service management, engineering, and design into the curriculum, and embed‐ ding humanity in learning activities.

2

Service Science Initiatives in Taiwan

Service Science initiatives in Taiwan consist of two types of efforts: individual institu‐ tions and cross organizational collaboration. In Subsect. 2.1, I briefly introduce the launch of the Institute of Service Science in National Tsing Hua University, which is the very first institute dedicated for Service Science research and education in Taiwan. In Subsect. 2.2, I introduce the establishment of “Service Science Society of Taiwan”, the professional community of Service Science in Taiwan by joining efforts across organizations. 2.1 Institute of Service Science (ISS), National Tsing Hua University (NTHU) The trend of interdisciplinary studies of Service Science was first perceived by academics in Taiwan in 2006. That motivated the College of Technology Management, National Tsing Hua University to establish a new post-graduate institute dedicated for interdisciplinary Service Science research and education. After 2-year preparation including the proposal of the establishment of the ISS approved by the university-level committee and the Ministry of Education (Taiwan) and the recruitment of faculty for

164

F.-R. Lin

the institute, the ISS master program was launched in the fall, 2008 (http:// www.iss.nthu.edu.tw). It is the very first educational program in Service Science in Taiwan. The ISS vision is to integrate domain knowledge in science, technology, management, and design to develop the Service Science discipline, to nourish students to become professionals in service innovation, to improve the wellbeing of human society and raise national competitiveness, to cooperate with industries to boost up the service economy, and to contribute to domestic and international academic communities with research outputs. The curriculum development of the ISS, taking the value co-creation of service systems perspective, included professionals from academics in various disciplines and practitioners from industries and governments. As Taiwan’s economy has been moving toward service-oriented economy, the demand of talents to tackle issues in social and business transformation is increasing. The ISS curriculum was developed to integrate three disciplines in service management, information technology, and service design, and to embed humanity into the learning activities, so that the talents of Service Science can demonstrate their competency in human centered service innovation. Figure 1 illus‐ trates the integration of business, information, and design which are embedded with humanity as the core of ISS curriculum. Service

B Information technology ICT service Service Oriented Architecture Cloud service

I

Business strategy, Business process modeling, Service management, Service marketing

D

Service innovation, Service experience design Service interaction design

umanity

Fig. 1. The constituted areas of studies in Service Science in ISS (NTHU)

The curriculum of the ISS master’s degree program has been evolving since 2008 according to the needs of the service economy and the recruitment of faculty. The current curriculum of the ISS master’s degree program is illustrated in Fig. 2. Master’s degree students, usually finishing the program in 2 years, take the Introduction to Service Science, Service Innovation, Research Methodology, Seminars as required courses, and complete master theses to obtain the master degree in Master in Business Administra‐ tion. To fulfill the interdisciplinary studies of Service Science, master’s degree students should take courses from each of four modules: service innovation and design, service marketing and management, service oriented information technology, and business analytics, and can declare their concentrations from one or two modules by taking at least three courses from a module. By taking a deep look at Fig. 2, we can understand the cubic structure of the ISS master’s degree curriculum, in which learning activities are founded on humanistic, social, and economic contexts and are interacting with

Humanity-Embedded Service Science Education

165

cooperating industrial partners. It extends the academic silos across disciplines and moreover co-creates value with social and economic service systems on the way to cultivating Service Science talents. Currently, the ISS admits 21 full time master students each year.

Industrial cooperaƟon

Service Science Interdisciplinary Study Module 1

Module 2

Service InnovaƟon & design

Service markeƟng & management

IntroducƟon to service science

Service innovaƟon

Summer internship program

Module 3 Serviceoriented informaƟon technology

Seminar I & II

Module 4 Business analyƟcs

Research methodology

Humanity, social, economic context

Master thesis

Fig. 2. The curriculum structure of ISS master’s degree program (2017)

There are two learning activities which actualize the intention of integrating the Service Science learning environment in social and economic contexts. One is the required summer internship program for each master student; the other is the 3-day humanity camp and extended 3-month social service projects for students who partici‐ pate in this program. By implementing these two aforementioned programs since ISS was established, ISS graduates have differentiated themselves from graduates who have majored in such disciplines as management, information systems, and design. ISS program has also provided them with maturity for the business world and for human centered service innovation. Moreover, their self-perception of life goals to benefit others while they are strengthening their professional capabilities is also distinguishing them from other graduates. I will highlight the humanity-embedded learning activities in Sect. 3. After establishing the recognition of the master’s degree program in Service Science in Taiwan, the ISS launched its very first doctoral program in Service Science in Taiwan or even in the world in 2014. The doctoral program admits international doctoral students and teaches in English. The doctoral program aims to nourish doctoral students to equip their research capabilities across discipline boundaries. The doctoral students are flexible to select their areas of problem domains related to Service Science as their research foci, by which to orient their selection of course works. The doctoral program requires students to master research methodologies and explore theories related to Service Science as their required courses, besides the goal to complete dissertation to be granted the Ph.D. degree in Service Science. The doctoral program generally takes four to five years to complete. Currently, the ISS admits 2–4 doctoral students including interna‐ tional students per year.

166

F.-R. Lin

2.2 Service Science Society of Taiwan (S3tw) The community for Service Science in academics and industries was emerging since the ISS was established in 2008. The ISS held conferences and workshops to promote Service Science in research and education in Taiwan. IBM in Taiwan also facilitated the engagement of service front runners in the world with Taiwan’s community. The need to form a formal association in Taiwan in order to promote Service Science for academics and industries was emerging when the government of Taiwan was also seeking to strengthen the service-oriented business transformation. The Service Science Society of Taiwan (s3tw) was formally chartered as a juridical association in 2011 to champion the promotion of Service Science in Taiwan (http://www.s3tw.org.tw). The s3tw established task forces as committees to coordinate the development of Service Science communities. For example, the teaching and learning committee called for a joint effort of scholars from disciplines in management, information technology and design to construct a reference master’s degree level curriculum in Service Science. The committee referenced the competency-based master’s degree curriculum for SSME program using Delphi method [4], and concluded the web search of Service Science related academic programs around the world to propose the reference master’s degree curriculum structure for the Service Science community in Taiwan. It is published on the s3tw web site for open access. Figure 3 depicts the journey of studying Service Science starting from taking the Introduction to Service Science until exiting by completing the service case studies. The courses are composed of four categories: service context, service management, service design and service engineering. Recom‐ mended courses for each category are also identified with example syllabi and course modules for the instructors’ reference. Service case studies

IntroducƟon to Service science Law and regulaƟon for service

InformaƟon and service economy

Service context

Service management OrganizaƟonal behavior and leadership

Service markeƟng Service innovaƟon

Service design

Service engineering

Service design

Service process modeling and analysis

Service experience design

Service system engineering

Service interacƟon design

Service oriented Technology

Fig. 3. The reference master’s degree curriculum in Service Science proposed by s3tw (2011)

The s3tw research and development committee organized an International Confer‐ ence on Service Science and Innovation (ICSSI) to create a platform for scholars and practitioners to share their research progress in Service Science related topics. It also

Humanity-Embedded Service Science Education

167

hosts the award competition for doctoral dissertations and master’s theses in Service Science related topics in Taiwan. ICSSI is grateful for the generous sponsorship by Sayling Wen Cultural and Education Foundation. The foundation also sponsored scholars to go abroad to conduct collaborative research activities with foreign scholars, and sponsored scholars to visit Taiwan to connect Service Science academic leaders with the Taiwan community. The s3tw extends the professional network in Service Science by motivating scholars and practitioners in numerous regions to form their regional alliances for knowledge sharing and collaborative projects dealing with local social and economic issues. By exercising the disciplines from various domains, there have been several successful local initiatives in solving social and economic issues in regions. For example, in Hsinchu County, a startup called “Da Shan Bei Yue” (meaning big hill and north moon in Chinese) innovated services to vitalize the campus of a primary school which was disbanded due to decreasing population in the community. The s3tw bid to acquire the rights from the township office to operate the venue. In line with the vision of s3tw to promote innovation and entrepreneurship, Mr. Peter Chuang, an alumnus graduated from the ISS and also the founder of Da Shan Bei Yue, operates the venue as a startup. He acts as a curator to invent a dining service using the local agricultural produce. Moreover, he held events to connect people in the region and to raise the concerns about the sustainability of local service systems such as transportation, education, agriculture, and environment. He inspired many youths to join the efforts, and many serve in its summer intern program. There are two other local projects, performed in the environs of Taipei and Kaohsiung. The projects are demonstrating cross-disciplinary efforts for solving social and economic issues while granting the opportunities for faculty and students to have hands-on experiences in applying service design and innovation methods and tools for solving real world problems.

3

Humanity-Embedded Learning Activities

᭹ົ

The term “service” in Chinese is “ ” (fu wu). There are two intentions for people in Chinese culture to view “service”. One is to connect service with service industries; the other refers to service for a charity or nonprofit organization. These various perceptions ” two characters in Chinese exactly mean. I inspired me to investigate what “ ” (shu wen jie zi) to understand what “ ” and “ ” checked the ancient book “ mean, respectively. It says that “ ” denotes a person rowing a boat, and “ ” means ” as the fast pace to do things for others. As fast pace. We can interpret the term “ we can image the difficulty of keeping the boat floating as smoothly as possible when the water is turbulent, service calls a special attention for the rower to maintain passen‐ gers’ comfort and speed to arrive at the destination. By taking the metaphor of service in rowing a boat in a fast pace, we catch the meaning of service in Chinese culture which is not directly related to service industries or charity/non-profit organizations. It is essential not to mislead students in business school to perceive that Service Science only cares of business in service industries, since it narrows down the potential of Service Science for the service systems based on S-DL. Within a Chinese cultural context, I

᭹ົ 崹ᩥゎᏐ ᭹ ᭹ົ

᭹ ົ ົ

168

F.-R. Lin

㐨ᚫ⥂

brought the philosophy of water from Dao Te Ching ( ) as the philosophy of service. It says that the top virtue is like water benefiting all beings but contending none; it behaves reservedly from people in general pursuing high positions, but it reaches the . By highest virtue taking this philosophy of water, it becomes easier for Chinese society to understand the science of service systems based on S-DL. We hold the annual event of river tracing for ISS students to experience the dynamics of water in nature while tracing a river with a team’s effort. Thus, they are able to be inspired by the hurdle they may face to help each other to reach the destination. That serves as an orientation activity for ISS first year master students to engage with second year students. 3.1 Service Humanity Camp and Service Learning In order to embed humanity in learning Service Science, an experiential learning activity called Service Humanity Camp was created when the ISS master’s degree program was launched in 2008. The camp has been held twice a year in winter and summer breaks. Its vision is to ignite participants’ power of self-change on their life journey of service. As we may know that we cannot achieve a different future if we keep ourselves unchanged. That is, if we want to have better self, we at the moment need to be able to change ourselves. It is essential for learning Service Science that we learn “to be”, besides learning “to know”. The service humanity camp is designed to facilitate individuals to examine their own status of life quality by participating in various experiential learning activities. Individ‐ uals are formed into teams to accomplish team challenges that provide opportunities to empathize with others during the team’s work. Each camp can accommodate around 80 participants, who have mainly been juniors or above in colleges from the universities in Taiwan. Starting from the first camp till the most recent camp, I have been fully engaged in the camp activities. From my observation and participation of activities, I have observed the empowerment of change for many individuals who are able to take their initiatives to change themselves regarding things they mentioned in the camp. I think it is a funda‐ mental value of education to enable individuals to possess the power of change that drives them to do things meaningfully different. After the 3-days of camp activities, teams continue their planned actions to perform social service for the following three months. Then, they reunion to share the outcomes and lessons learned from their endeavors. The 3-month projects were designed for team members to practice individual expertise to accomplish the tasks they committed, and to experience empathy and leadership in the process. The camp is a very unique learning experience for the majority of students in Taiwan. In order to allow individuals to have the fresh experiences in the camp, it is discouraged for participants to release the details of activities to those who haven’t attended the camp. Because of this restriction, it is very different from common courses that share the agenda for prospective attendees, and in turn, the upcoming attendees are usually referred by those who have already attended the camp. Gradually those who attended the camp are forming their communities because of the common experiences in service humanity

Humanity-Embedded Service Science Education

169

camps. They are becoming the task forces to sustain the program by returning to serve as assistants or referring prospective participants. In S-DL, they co-create value and sustain the service system for experiencing service humanity. 3.2 The ISS Internship Program Every ISS master’s degree student is required to spend two months, July and August, working for an organization in the summer. This program element aims to enable the first year master’s degree students to exercise the domain knowledge they obtained from their first year of study. Moreover, they can get feedback from being involved in daily service encounters to examine their willingness and abilities to challenge themselves to complete tasks. In order to engage students with summer intern employers, the ISS office helps students to reach out to potential organizations that might provide summer intern positions. Students are also welcome to identify their dream summer internship and contact potential providers. Two events, a Yes Party and a Celebration Party, before and after summer internships, respectively sustain the value co-creation between intern employers and ISS. The “Yes Party” is an event where the ISS invites summer intern employers to visit the ISS facilities. At the party, students selected by the corresponding intern employers present themselves to highlight their capabilities and expectation of learning from the internship. The employers introduce their organizations and the planned tasks for the interns. From these bi-directional communications, the employers representing different organizations got the chance to know more of the ISS via students’ sharing. Moreover, faculty is also able to meet these employers to establish the channels for communica‐ tions. There are two intentions in naming this the “Yes Party”. One is to thank employers saying “yes” to our students and accepting their intern applications. The other is to ask students to say “yes” when they are asked to take the assignment to accomplish tasks. The Yes Party lays the foundation for successful internships by co-creating value between the ISS students and intern employers. The “Celebration Party” is held in fall semester after students return from their summer internship. Its purpose, mainly, is to show our appreciation for employers’ contributions to students’ learning during summer internship, and for students to share their accomplishments and lessons learned from internship. First year mater’s students also attend the party, so that they can get the sense of the internship experience that they will go through in following summer, and also they can start to anticipate the upcoming requirements. In summary, the ISS summer internship program is unique from existing master’s degree programs in related disciplines. By aligning with industrial resources that are synchronized with the educational activities of Service Science, the summer program realizes the anticipation of Service Science learning from tackling real world problems (outside-in) and utilize academic capacities to solve the problems (inside-out). Moreover, it is a practice of humanity as students can collaborate with others to solve problems, by which they fulfill their life destiny.

170

F.-R. Lin

3.3 Project-Based Learning Activities We adopt the outside-in and inside-out approach to conduct project-based learning activities in many courses, such as the Introduction to Service Science, Service Inno‐ vation, Service Design, Data/Text Mining, Service Oriented Architecture, etc. It takes the real-world problems into the learning contexts, mainly aligning with the design process from discovery, define, design, to development stage. Taking the Introduction to Service Science as an example course, students enrolled in this class need to conduct three projects in a semester, and the team size will grow from 2, 4, to 6, respectively. The first project is to study the real-world service systems and model them using stakeholder diagrams, actor network diagrams based on ActorNetwork Theory, and activity diagrams based on Activity Theory. A team with two members can explore the service systems they are interested in with low coordination cost due to the limited co-workers in a team. Project 2 is designed for two teams in project 1 to form a new team to conduct service design for the target domain they agree with. Besides exercising the methods and tools of service design process, the embedded learning of humanity by cooperating with team members is designed in the process of conducting the second project. Along with the project 2, the philosophy of water in service, and the S-DL are brought to the class for discussion and clarification after they read the provided reference materials. Finally, the third project is designed for a team of six members to revise the service prototypes they developed in the second project from the team they originated, and create a service prototype with market testing. Since the change of team members may cause several issues in team dynamics, the design of the third project provides students oppor‐ tunities to exercise different stages of tribal leadership [5] which was taught in class. In order to grant students the opportunities to access issues faced by local commun‐ ities and industries, the instructor organized field trips for students to visit nearby town‐ ships to experience the living and working contexts of the local residences and incoming tourists. Thus, students exercising their empathy to understand the local contexts and then contributing to the problems’ solutions via service innovation can further strengthen their humanitarian capacity. This endeavor currently becomes a common practice for universities to fulfill their social responsibilities in Taiwan.

4

Reflection

It was a new venture for Taiwan to embrace the interdisciplinary study of Service Science starting from 2006 when Taiwan was facing the industrial transformation from manufacturing oriented type of business to service-oriented economy. Although the original SSME (SSMED/SSMEA) advocator, IBM, gradually shifted its emphases on smarter planet and cognitive computing after successfully creating the awareness of interdisciplinary studies of Service Science around the world, the intention of integrating technology with human centered service is even stronger with advanced artificial intel‐ ligence technologies. Thus, I perceive the spirit and practice of interdisciplinary studies of Service Science are living and even more thriving than ever.

Humanity-Embedded Service Science Education

171

The ISS in NTHU, as the first institute dedicated for Service Science related research and education, bears the responsibility to sustain the growth of the discipline and to contribute to academics and society via talent cultivation and research outputs. The leading master and international doctoral programs grant the opportunities to practice new ways of teaching and learning in order to adapt to the dynamic social and economic environments. There are several initiatives that lay the foundation for Service Science research and education to create unique contributions to talent cultivation and research breakthrough. First, the ISS brought the philosophy of water from Dao Te Ching to raise the anticipation of service as the top virtue to communicate with people in Chinese culture that connects the discipline with the root of wellbeing of society. Second, besides enhancing and expanding students’ knowledge and skills in service management, information tech‐ nology, and service design, the ISS embeds humanity into the courses and activities intentionally to equip students’ humanitarian capacity to contribute to the society during and after their studies of Service Science in ISS. It is very unique in Taiwan to hold the dedicated service humanity camp with three month service learning projects and the required summer internship in post-graduate studies. However, by doing these humanity-embedded programs, ISS graduates have been standing out from their peers in various industries and organizations to demonstrate their professional excellence with humanitarian capacity. Third, the ISS experiences in Service Science support the establishment and growth of Service Science community in Taiwan. The s3tw establishment enlarges the profes‐ sional networks of Service Science to bring government, industries and academics to describe the world using S-DL, and start to model emerging service systems with stake‐ holders for value co-creation. At the end, the community of Service Science will be able to embed human-centered mindset and technology into the process of institutionalizing service systems aligning with social changes. Finally, foreseeing the trend of adopting advanced technologies, such as artificial intelligence, mobile technologies, etc., for service innovation, humanity cannot be over‐ emphasized while facing the dilemma between automated service replacing human servants and the leverage of human intelligence for social wellbeing. Thus, the philos‐ ophy of service humanity should be included in the study of Service Science, so that the innovation of service technology and management achieves the common goods of human societies.

5

Conclusion

Founded in 2008, the ISS is about to move into its tenth anniversary year. By employing the essential competency of service innovation to co-create value with the stakeholders of the emerging service systems of Service Science, the ISS has founded and built a service system for Service Science discipline in Taiwan. With the recruitment of inter‐ national faculty and international students to enroll in the doctoral program, the ISS aims to contribute to the international Service Science communities by leveraging talents around the world to co-create academic value to sustain the Service Science discipline,

172

F.-R. Lin

which will contribute to the paradigm shift of service economy and humanitarian society [6]. The humanity-embedded Service Science education practiced by the ISS has been recognized by the Service Science society in Taiwan; even some emerging disciplines in Taiwan, such as learning science, or existing disciplines, such as information systems, industrial engineering and management, and computer science, show their curiosity and interests in learning from these practices. Thus, it reflects the call for the emphases of humanitarian capacity embedded in professional development. Acknowledgments. The author would like to express the appreciation to the faculty of the Institute of Service Science (National Tsing Hua University), Service Science Society of Taiwan, Sayling Wen Cultural and Educational Foundation, and scholars and practitioners of Service Science in Taiwan for promoting Service Science research and education collaboratively.

References 1. IfM and IBM: Succeeding Through Service Innovation: A Discussion Paper. University of Cambridge Institute of Manufacturing (2007) 2. Spohrer, J.C., Maglio, P.P.: Toward a science of service systems. In: Maglio, P.P., et al. (eds.) Handbook of Service Science: Research and Innovations in the Service Economy (2010) 3. Lusch, R.F., Vargo, S.L.: Service-Dominant Logic: Premises, Perspectives, Possibilities. Cambridge University Press, Cambridge (2014) 4. Choudaha, R.: Competency-based curriculum for a master’s program in Service Science, Management and Engineering (SSME): an online Delphi study. Unpolished dissertation, University of Denver (2008) 5. Logan, D., King, J.: Tribal Leadership: Leveraging Natural Groups to Build a Thriving Organization. Harper Business, New York (2011) 6. Kuhn, S.: The Structure of Scientific Revolution. University of Chicago Press, Chicago (1962)

Developing T-Shaped Civil Engineers Through Involvement in Out-of-Class Activities Denise R. Simmons ✉ (

)

Myers-Lawson School of Construction and Department of Civil and Environmental Engineering, Virginia Tech, 310A Bishop-Favrao Hall, 1345 Perry Street (MC0188), Blacksburg, VA 24061, USA [email protected]

Abstract. The National Academy of Engineering, American Society of Civil Engineering, ABET, Inc. and other entities have described the global vision for the future of engineering and civil engineering specifically. Included in this global vision are the competencies necessary to equip civil engineers to contribute to solving industry-spanning technical, social, financial, political, and cultural chal‐ lenges. While curricular activities have long been the focus of engineering educa‐ tion research and practice, engineering educators have recently begun to recog‐ nize the value of what engineering students do outside of the classroom. The goal of this paper is to introduce and engage the AHFE HSSE community on the workforce challenges specific to the construction industry, and to explore out-ofclass activities as pathways for developing the 21st century competencies neces‐ sary to solve these challenges. Keywords: Civil engineers · Construction industry · Rapid rebuilding · Competency

1

Introduction

The National Academy of Engineering (NAE) is a key stakeholder in identifying engi‐ neering competencies. The book, The Engineer of 2020: Visions of Engineering in the New Century, is a part of the NAE Committee on Engineering Education (CEE) initiative on engineering in the future and educating engineers to meet the needs of the new era. The Engineer of 2020 presents facts, forecasts future conditions, and develops future scenarios of the possible world conditions for the 2020 engineer and the attributes these engineers must have to prepare them to help the public meet these world conditions [1]. The Engineer of 2020 identifies 10 attributes necessary for graduating engineers to possess in the next three years [1]: 1. strong analytical skills; 2. practical ingenuity (skill in planning, combining, and adapting. Manner in which one identifies problems and finds solutions); 3. creativity (invention, innovation, thinking outside the box, art); 4. good communication skills;

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_16

174

D.R. Simmons

5. 6. 7. 8. 9.

principles of business and management; principles of leadership; high ethical standards; strong sense of professionalism; dynamism, agility, resilience, flexibility (the ability to learn new things quickly and the ability to apply knowledge to new problems and new contexts) and 10. lifelong learning. Reviewing the calendar and category of each attribute support the timeliness of this discussion. The engineers of 2020 entered college in the fall of 2016 if a four year time to degree completion is assumed. If a longer time to degree completion is assumed, the engineers of 2020 arrived on campus before fall 2016. Are the institutions that received these future engineers ready to develop their whole person in a way that fulfills the vision of engineering in the new century? Further, take note of the 10 attributes and which can be categorized as deep, discipline based versus broad discipline spanning attributes. Surprisingly, nine of the 10 attributes are broad in nature – the very attributes that have been persistently less developed in graduating engineers.

2

Rating the Quality of US Infrastructure

Civil engineers are charged with solving complex technology, infrastructure and labor challenges. Infrastructure spending in the US represents 3.3% of the Gross National Product (GNP) [2]. Personal consumption is the largest sector contributing to the GNP, approximately 70% of GNP, and it relies heavily on infrastructure to transport goods across the US [2]. Businesses depend on a network of infrastructure for all aspects of daily operations. However, the condition of our infrastructure is in such disrepair in the US that it has received a D+ grade by the American Society of Civil Engineers (ASCE) Infrastructure Score Card for many years including the current year [3]. The score card provides one of the most comprehensive snapshots of where the US is, and is not, spending money, and where the needs are likely to be the greatest in the coming years. Infrastructure spans a wide range of public and private assets, including highways and bridges, airports, ports and inland waterways, electricity plants and transmission lines, information and telecommunication networks and water and sewage facilities.

3

Rapidly Rebuilding from Scratch

Civil engineers have an important role in repairing and replacing infrastructure – an enormous US challenge given the number of infrastructure systems rated below a C grade on the ASCE Infrastructure Score Card. The current US government administra‐ tion promises to devote over $200 billion to infrastructure spending [3]. While consistent investment in infrastructure and planning are essential solutions to raising and sustaining the quality of our infrastructure, competency of civil engineers and other constructors must also be considered. One competency widely discussed as vital to these professio‐ nals is the ability to rapidly rebuild infrastructure from scratch.

Developing T-Shaped Civil Engineers

175

Rapidly rebuilding from scratch is a measure of resilience of systems. Too often civil engineers think about building something that will last forever, but in the age of accel‐ erating technologies it is like trying to build a stone bridge when steel is coming. Eiffel built one of the first steel bridge in Porto, Portugal at a fraction of the time and cost of building a larger stone bridge. In the age of accelerations, or the ever-widening range of innovations that simulta‐ neously impact business, society, and the built environment, the nature of what gets built, how it gets built, as well as how it is recycled and rebuilt are changing more rapidly with broad impacts on costs, business models, safety, and environmental factors [4–7]. In an age of rapid change, people may have to rapidly rebuild careers in new areas too. Can we as a civilization create T-shaped engineers who can rapidly rebuild everything from scratch?

4

Competency Development from Out-of-Class Activities

The engineering education community has dedicated attention and research to curricular changes [8] and more recently to out-of-class support for engineering students [9–11]. Involvement in out-of-class activities has long been assumed to be complementary to educational and developmental processes and specifically undergraduate STEM learning [12, 13] and persistence. In this paper, out-of-class activities are defined as activities that occur outside of the formal classroom and curriculum. These activities are connected to the academic curriculum and educational institution and can include “community service groups, student government, fraternities, athletics, honor societies and religious clubs” [12, p. 4]. Solving rapid rebuilding challenges requires civil engineers to have broad and deep competencies. Spohrer and Kwan (2009) [14], through work at the IBM Corporation, articulate these competencies through a model dubbed “The T”. They explain that far too many of today’s students come out of college with deep competencies in a single discipline, referred to as “I-shaped” professionals, but very little boundary crossing competencies with which to engage in complex interactions. Therefore, they advocate for the development of “T-shaped” students who are deeply skilled in their own area, broadly versed in boundary spanning skills and adept in these skills when they graduate and enter professional practice. Recent studies have begun to examine the influence of what engineers do outside the college classroom to develop technical and professional competencies [11, 15]. In a recent study of undergraduate civil engineering students, the three most prevalent outof- class activities in which civil engineering students participate include off- or oncampus employment (53.4%), sports (37.8%), and student clubs and organizations (32.7%) [16]. The most reported positive outcomes of civil engineering students’ outof- class involvement are personal development, social development, and social engage‐ ment [16]. These findings complement conclusions from research studies conducted with a broad engineering student sample. For example, Flowers (2004) [17] found that student involvement directly affects gains in personal and social development. Similarly, Burt et al. [18] reported that engineering students involved in co-curricular activities

176

D.R. Simmons

exhibit greater leadership skills and practice more ethical decision-making processes— both significant facets of personal and social development. More specifically, Zydney et al. (2002) [19] found that participation in undergraduate research led to greater enhancement of important personal skills, while other researchers likened positive outcomes such as social engagement to learning community participation [20]. The engineer of 2020 attributes can serve as a proxy for the T-shaped skills and expertise. Initial results from an on-going study of civil engineering students (n = 185) to examine their involvement in and outcomes from out-of-class activities indicate that when the 10 engineer of 2020 attributes are grouped as a single outcome, these students point to involvement in an off- or on-campus job (16.7%), design competition teams (10.8%), research (9.2%) and equally to professional experiences and sports (8.7%) as the activities from which the attributes were developed. These results offer insight into the ability of out-of-class activities to help students development technical and profes‐ sional competencies.

5

Conclusion and Future Research

This paper introduces the call for rapid rebuilding as a job competency for civil engineers and constructors and explores competencies that may help civil engineers answer this call. Finally, the paper recommends the role that involvement in out-of-class activities by civil engineering students can play in helping them develop broad and deep attributes while in college. Future research includes answering four questions: (1) what specific competencies indicate and, when developed, enable a civil engineer to rapidly rebuild infrastructure from scratch; (2) what out-of-class activities and specific components within out-of-class activities produce positive academic outcomes leading to a civil engineer’s ability to rapidly rebuild infrastructure from scratch; (3) What influences civil engineering students’ decision to become involved in these out-of-class activities; and (4) What, if any, are the technological and policy barriers to rapidly rebuilding infra‐ structure from scratch? Acknowledgments. The author gratefully acknowledges the contribution of Jim Spohrer, IBM Research, for helping to shape this submission. This material is based upon work supported, in part, by the National Science Foundation, under grant #EEC-1351156. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.

References 1. National Academy of Engineering: The Engineer of 2020: Visions of Engineering in the New Century. The National Academies Press, Washington, D.C. (2004) 2. Bureau of Economic Analysis: Gross Domestic Product. United States Department of Commerce, Economics and Statistics Administration, Bureau of Economic Analysis (2016) 3. ASCE: Infrastructure Report Card (2017). http://www.infrastructurereportcard.org/grades 4. Friedman, T.L.: Thank-You for Being Late: An Optimists Guide to Thriving in the Age of Accelerations. Farrar, Straus and Giroux, New York City (2017)

Developing T-Shaped Civil Engineers

177

5. Handy, S.L., Boarnet, M.G., Ewing, R., Killingsworth, R.E.: How the built environment affects physical activity: views from urban planning. Am. J. Prev. Med. 23(2), 64–73 (2002) 6. Tukker, A.: Product services for a resource-efficient and circular economy–a review. J. Clean. Prod. 15(97), 76–91 (2015) 7. Amaratunga, D., Haigh, R.: Post-disaster reconstruction of the built environment: rebuilding for resilience. Wiley, New York City (2011) 8. Brown, M., Hershock, C., Finelli, C., O’Neal, C.: Teaching for retention in science, engineering, and math disciplines: a guide for faculty, vol. 25, pp. 1–12. The University of Michigan, Ann Arbor (2009) 9. Finelli, C., Matthew, H., Ra, E., Bielby, R., Burt, B., Carpenter, D., Harding, T.S., Sutkus, J.: An assessment of engineering students’ curricular and co-curricular experiences and their ethical development. J. Eng. Educ. 101(3), 469–494 (2012) 10. Wilson, D., Jones, D., Mee Joo, K., Allendoerfer, C., Bates, R., Crawford, J., Floyd-Smith, T., Plett, M., Veilleux, N.: The link between cocurricular activities and academic engagement in engineering education. J. Eng. Educ. 103(4), 515–651 (2014) 11. Simmons, D.R., Ye, Y., Hunsu, N.J., Adesope, O.O.: Engaging engineering students outside of the classroom: initial validity, reliability and instrument development of the Postsecondary Student Engagement (PosSE) Survey. Int. J. Eng. Educ. (2017, accepted) 12. Ahren, C. S.: Disentangling the unique effects of co-curricular engagement on self-reported student learning outcomes. Indiana University, ProQuest, UMI Dissertations Publishing (2009) 13. Thiry, H., Laursen, S., Hunter, A.B.: What experiences help students become scientists? A comparative study of research and other sources of personal and professional gains for STEM undergraduates. J. High. Educ. 82(4), 357–388 (2011) 14. Spohrer, J., Kwan, S.K.: Service science, management, engineering, and design (SSMED): an emerging discipline–outline and references. Int. J. Inf. Syst. Serv. Sect. 1(3), 1–31 (2009) 15. Yu, R., Simmons, D.R.: Synthesis of Engineering Undergraduate Students’ Out-of-Class Involvement Paper presented at 2015 ASEE Annual Conference & Exposition, Seattle, Washington, June 2015. doi:10.18260/p.24787 16. Simmons, D.R., Ye, Y., Ohland, M., Garahan, K.: Academic and professional skill development through out of class involvement by civil engineering students. J. Prof. Issues Eng. Educ. Pract. (under review) 17. Flowers, L.A.: Examining the effects of student involvement on african american college student development. J. Coll. Student Dev. 45(6), 633–654 (2004). doi:10.1353/csd. 2004.0067 18. Burt, B.A., Carpenter, D.D., Finelli, C.J., Harding, T.S., Sutkus, J., Holsapple, M., Bielby, R.M., Ra, E.: Outcomes of Engaging Engineering Undergraduates in Co-curricular Experiences, pp. 00556–00511. American Society of Engineering (2011) 19. Zydney, A.L., Bennett, J.S., Shahid, A., Bauer, K.W.: Impact of undergraduate research experience in engineering. J. Eng. Educ. 91(2), 151–157 (2002) 20. Pike, G.R., Kuh, G.D., McCormick, A.C.: An investigation of the contingent relationships between learning community participation and student engagement. Res. High. Educ. 52(3), 300–322 (2011). doi:10.1007/s11162-010-9192-1

T-Shape Professionals Co-working in Smart Contexts: VEGA(ST) – Venice Gateway for Science and Technology Paolo Piciocchi1, James C. Spohrer2, Linda Martuscelli1, Maria Cristina Pietronudo4, ( ) Michele Scocozza3, and Clara Bassano4 ✉ 1

4

Department of Political, Social and Communication Sciences, University of Salerno, Salerno, Italy 2 IBM Almaden Research Center, San Jose, CA 95120, USA 3 University of Modena and Reggio Emilia, Modena, Italy Department of Management Studies and Quantitative Methods, University of Naples “Parthenope”, Naples, Italy [email protected]

Abstract. The places where people routinely gather to interact and work can be thought of as territories with boundaries that provide a smart context to achieve personal and professional goals. For example, the Venice Gateway for Science and Technology of VEGA(ST) is such a smart context, in the form of a co-working space. In this paper, smart contexts are explored using the integrated framework of Service Science, Management Engineering, Design Arts Public Policy (SSME +DAPP) and the Viable Systems Approach (VSA) research methods, which encourage the development of T-shaped professionals with depth and breadth. We explore the importance of: (1) adopting a win-win logic between the external environment and the resource holders of the wider enterprise-territory system, (2) the creation of smart contexts, as a symbol of sharing, where interaction and actor involvement facilitate the co-creation of value for the collective welfare, and (3) co-working spaces as a way to encourage T-shape professionals to develop their broad dynamic capabilities and their in depth competencies. Keywords: Co-working · T-shaped professionals · SSME+DAPP & VSA · Value Co-Creation

1

Introduction

Driving the next wave of innovation, productivity, and economic growth for regions will likely require rethinking the design of places where people routinely gather to interact and work. Consequently, we will argue that organizations should make better use of new opportunities to optimize performance and contribute to the growth and development of the territory of reference by promoting the mechanisms of value cocreation with a plurality of subjects, integrating internal with external resources and activating virtuous circles of shared growth. In this respect, the involvement of a wide range of professionals within an integrated system guarantees a flow of interesting © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_17

T-Shape Professionals Co-working in Smart Contexts

179

information. In the actual context, competitiveness of new business models is strongly related to the concept of the “Shared Value” [1], defined as policies and operating prac‐ tices that enhance the competitiveness of a company while simultaneously advancing the economic and social conditions in the communities in which it operates [1]. This means creation and sustainability processes of competitive advantage needs to involve economic and social stakeholders. The concept of shared value isn’t about “sharing” the value already created by firms, but is a redistribution approach [1]. For this assumption, it is required to start from a process of spreading “embedded” knowledge acquired both from inter-social relations and from the territory of reference and involve smart conso‐ nant resources in the business development. A smart context is a place where a wide range of professionals can routinely gather to interact and work, and which fosters the creative and productive integration of resources within a specified terrority with boundaries. People flow into and out of places of work, and so they must be designed in harmony with their external environment as well. To achieve such a goal, the design and building of smart contexts, characterised by adequate levels of mutual interaction and co-finalization of the actions is funda‐ mental, as in the case of the “contexts of the so-called sharing economy.” A smart context derives from the “environment” and “resources” capable of facilitating the sharing and co-creation of value and, not least, the equifinality of the entire service system. A smart context is a place (a territory with boundaries, in harmony with an external environment) where professionals gather to interact, and the mental model (self-image) that professionals have of themselves matters as well. Therefore, in our study we attempt to define the structural profile of T-shaped professional, as smart consonant resources needed to guarantee the sustainability of competitive advantage. In this perspective, we will consider the “behaviorist” logic based on the “knowing how to be” concept rather then the “structuralist” logic [2, 3] just based on the “being” concept. These considera‐ tions are consistent with the systemic approach of organizations to rethink and renew their organizational structure having the necessary degree of openness and willingness to cooperate both for their own special interests and at the same time, for those in the collective context of reference. This work presents an integrated framework and practical realization. The first illus‐ trates the methodological configuration of the integrated framework we propose - the Service Science Management and Engineering Design Arts Public Policy (SSME+DAPP) and the Viable Systems Approach (VSA) - through which the relations and collective positive externalities from “law of increasing returns” are explored [4] and contextualized in the reality of co-working and T-Shape Professionals. The theoretical background in the second section deals with the issue of identity and systems character/personality of a terri‐ tory, not in terms of a static object, but from a dynamic view of subject, characterized by inter-actor co-evolutionary strategies for systemic, equifinal goals. The integrated framework finds practical realization in the co-working system in which the contribution of value to collective growth derives from the integration of contributions co-designed by virtue of the professional specificity of the T-shaped type. The VEGA(ST) project is a pertinent example of global development whereby urban spaces, in line with the logic of smart working, are regenerated and made available to institutions, private sector companies and universities, to increase competitiveness not

180

P. Piciocchi et al.

only in individual terms but also in terms of the context as a whole, by acting on the principles of smartness. Finally, the practical implications, namely the potential and desirable structural and systems conditions of future scenarios with highly competitive although not easily replicable content will be illustrated.

2

Integrated Framework: T-Shaped Professionals

The dynamism of the competitive scenario leads companies and institutions to modify their approaches in the direction of smart government [5] or toward collaborative, shared arrangements of governance. From a cohesive service systems viewpoint, according to a systems logic based on the concept of integrated “relational capacities for distributed competences” [6], globalization should be seen from the perspective of wide, synergistic collaborations between open systems such as companies, universities, research centers and institutions. In this regard, the common goal is the co-creation of value consistent with open innovation models that are integrated within the sharing economy. The value co-creation orientation in service systems thus becomes the turning point to guide new logics of local/territorial growth. In other words, system survival and sustainability of lasting competitive advantage, not easily replicated, depends on the capacity to develop syner‐ gies and involvement of several different local actors. According to this setting, SSME +DAPP perspective, summed up as SS (Service Science), is defined by Maglio and Spohrer [7] as the theoretical framework for the study of service systems as integrated dynamic resource configurations, resonant but at the same time autonomous, capable of participating in the shared value co-creation process. Service systems, in fact, survive and evolve through the reciprocal exchange of resources (such as knowledge and skills) with other systems. The constancy of adaptive and incremental interaction is the sine qua non condition for the effective co-creation of value in order to guarantee the survival and socio-economic growth of the context. The Viable Systems Approach (VSA) helps to establish relations between the various sub-systems (actors in general/stakeholders) characterizing the structure of the overall system. The survival of such a system depends on the degree of structural consonance1 (compatibility) required to achieve a suitable gradient of the overall systemic and wide‐ spread resonance2 (equifinality, reciprocal satisfaction of expectations). In other words, the viability of any system (individual or organizational) depends on [8, 9]: (1) openness of each actor/organization; (2) shared growth prospectsin the same territorial context; (3) contribution of each component to the emergent system in the process of value cocreation; 1

2

Consonance measures the capability of the system to achieve mutual benefits (value co-crea‐ tion) based on their structure (accessible resources) and the limits to sharing and coordinating information between different entities (viable systems). Resonance measures the results of interaction in context, producing and sharing value for and with stakeholders or suprasystems.

T-Shape Professionals Co-working in Smart Contexts

181

(4) ability to contribute to building competitive advantage by exploiting the opportu‐ nities for technological innovation. With regard to the last point listed above, Chesbrough [10] identified the “rupture point” for the historical isolation of companies and organizations in general (Open Innovation). The starting point for a synergistic fusion between systems is the willing‐ ness to collaborate and share evolutionary strategies, leveraging new relational struc‐ tures offered by technological innovation. Therefore, the open landscape office areas can be represented as an appropriate architecture for guaranteeing the smart and effective combination of internal and external resources in a specific coordinated context. In a perspective of value co-creation (Service Science perspective), this means offering a firmer connotation in terms of expertise and integration. Through the integrated framework of SSMED+DAPP and VSA it is also possible to appraise the aspects of professionalism employed in integration and sharing processes. To develop a process of value co-creation, many of competences, together with adequate integration of them by virtue of relational and communication skills, is required. The “shape of a professional” defines the capabilities that professionals can apply for problem solving and for communicating with other professionals as part of a project team. Such shape is determined by the depth and breadth of knowledge in different areas of study [11]. In a dynamic context, professionals and/or talented individuals can no longer seek only in depth specialization (academic degree) but should also be equipped, to some

Fig. 1. Characteristics of the T-shaped model, an effective evolution of the I-Shape model due to the allocation of relational and communication skills appropriate to smart integration between competecien and capabilities.

182

P. Piciocchi et al.

extent, with “horizontal” capabilities that enable them to learn and adapt quickly to changes in system needs. A T-shaped professional (T-SP) has breadth as well as depth: in depth, knowledge of one discipline area and, in breadth, the ability and experience necessary to communicate across a range of disciplinary areas. A person characterized with breadth understands the vocabulary of other disciplines, and uses this knowledge to interpret and describe problems that they may be unable to solve, but about which they can reason with experts of other disciplines to solve. Such a configuration defines the adaptive and/or proactive innovator. One proposed type of T-SP that SSME-DAPP as an emerging discipline seeks to graduate is summarized in Fig. 1 below [12–14]. According to the KSA (Knowledge-Skill-Attitude) model [15], the vertical dimension of the T-Shape model refers to the disciplinary specialization (degree/competences), while the horizontal-dimension refers to collaborative and communication skills (soft skills and attitudes). Both dimensions are essential for interpreting collective and territorial growth as such as the co-working, new form of “synergic space of cooperation”. Coworking is the social gathering of a group of people who are still working inde‐ pendently but share the values and are interested in the synergy that can be working closely with talented people; this implies co-working is a new form of cooperation – also based on technology – in dynamic contexts that engages a several actors (profes‐ sionals and organizations) in a shared project, generating virtuous mechanisms of cocreation and value-spreading. Co-working is a working style that creates collaborative processes based on shared environments, services and common values. Co-working facilitates the creation of synergies, physical and virtual sharing of spaces and encour‐ ages exchange of knowledge, enrichment of skills and competencies. It also increases the level of trust and, as a result, the stability of the collaborative network.

3

Integrated Framework: Smart Systems

Coherently with the VSA viewpoint, most “entities” (companies, individuals, districts, nations, customers, markets, communities) are interpreted as sufficiently open systems whose survival and growth imply interpersonal skills and integration with other systems. In fact, viability and competitiveness are the features that enable, within highly complex environments and presence of diverse variables, competitive identity - adaptive and proac‐ tive - as a result of core competencies, appropriately integrated with the systemic resources of other entities. Interaction becomes, therefore, the determining factor for the co-creation of shared value among multiple systems which have the same purpose. New paradigms push companies towards innovation outside the firm’s boundaries of research, consistent with a systemic, shared and participated vision of business; in this sense a “facilitator” role can and should be played by Open Innovation architectures, also in relation to the review of traditional management theories, since because of the practical implications and better performance. Companies realize that they cannot rely solely on innovation capability from within, which is why external actors have become an increasingly crucial part of compa‐ nies’ innovation capability [16].This result is related to the ability to be open (collabora‐ tive availability) and the ability to work synergistically (participatory availability) in coworking projects for collective economic and social growth. If we consider co-working as

T-Shape Professionals Co-working in Smart Contexts

183

a “physical and virtual” area designed to facilitate the exchange and sharing of knowledge among several actors (professionals and/or organizations) therefore we can say that Open innovation is able to play as a facilitator in co-working projects. Co-working, generally, offers various options: from open spaces to desks equipped with computers, to private meeting rooms, depending on the preferences expressed by the co-worker at the time of booking the space. The real added value of this practice lies in the inherent spirit of sharing fostered by the promoters of such spaces. The soul of co-working is, in fact, its “host”, i.e. the one who manages the structure that tends to offer aggregative opportunities and rela‐ tional occasions and exchange [18]. To support social capital, collaboration, and community-building, managers of co-working increasingly deploy synergic management strategies to foster interaction [19] assuming that innovation and knowledge exchange are social practices [20]. In this context, social relations are the main driver by which virtuous processes of collective productivity can be activated and knowledge and unique capabilities exploited in a synergistic way. Meeting new people is a reason for professionals to frequent coworking spaces, to access additional knowledge or skills [17, 21]. Meeting new people can also be of importance for many economic decisions and working outcomes, bridging resources, acquiring information, learning how to do one’s work or solving problems [22–24]. Sharing and contamination are the primary objectives to aspire to in order to benefit from the system of relationships offered by co-working practices. Co-working spaces are also regarded as “serendipity accelerators” [25] and workers are said to seek places to maximize serendipity and potential interaction with peers. Moreover, they access co-working spaces with the purpose of fostering networking practices and highly value the collaborative environment to feed innovation and crea‐ tivity [26]. Therefore, innovation and creativity are the ingredients that, together with a collaborative spirit and purposeful sharing, can guarantee the viability of the territorial systems, characterized by non replicable sustainable competitive advantage. In order to facilitate sharing, spaces and stimulating environments are required. However, it is also necessary that in such contexts, professional profiles have a propen‐ sity for a high degree of specialization and adequate communication and relationship skills; T-Shaped Professionals are a case in point. Careful management of the mix of members can also affect interaction. Handpicking tenants or members helps to increase the chance that they interact, work together and benefit from each other’s presence [19]. Based on this concept, some founders of co-working spaces seek specialization. Special‐ ization helps to enhance a reputation as “the place to” be within a certain industry and it can be a way to lure companies away from other co-working spaces. There is some evidence that specialized knowledge locations grow faster than those with a heteroge‐ neous member mix [27]. In our view systems are characterized by a simple collection and/or set of profes‐ sionals (I-shaped or H-Shaped Professionals) composed of specialists who however, are not able to ensure the circulation of information, to share knowledge and, therefore, to create virtuous processes of widespread value co-creation. In dynamic environments, like the present, the most suited to collaborative reality appears to be the T-Shaped Professional, a profile characterized at the same time as cognitive specialization (contri‐ buting value in value propositions) and relational attitude (mediation contribution and

184

P. Piciocchi et al.

participation in value co-creation and sharing value) in their co-working environments. This means, T-SP are better decision-makers because they are able to manage complexity in breadth and in depth and so are able to live with change (elasticity and flexibility plus efficiency). Several studies [28] suggest that T-shaped skills influence positively the creation of knowledge. We believe, also, that in the absence of an envi‐ ronment that stimulates the creation and use of T-shaped profiles, new knowledge is unlikely to be created. This implies that a crucial element in the successful management of knowledge is not only T-shaped ability as such, but the systematic management of this competence [29]. This takes place in open, collaborative and intelligent environ‐ ments and promotes interaction, co-creating and sharing values in co-working spaces, but especially in smart contexts, such as smart cities. Shared work, smart spaces, compe‐ tition, synergies contribute to the emergence of intelligent bundles in the territory, expanding the objectives and interests of a system of players, seeking the well-being not only of the promoters and participants, but of every citizens. Theoretical assumptions and the Smart City setting unanimously justify, within a collaborative and participatory vision, a new social, ethical and environmental dimension, harbinger of development and economic growth. The application of the theories of co-creation of value to territorial systems shifts the observation from a micro to a macro perspective. Smart Cities reinforce the theory that the territory can no longer be considered a mere container of endowed resources to exploit and/or consume, but a comprehensive service system able to coordinate and manage the available resources and those of system rationale for sustainable growth processes and a distinctive identity in a hyper-competitive environment [30]. Environ‐ ments such as schools, universities, research centers and industries that produce new technologies, if properly supported, thanks to the mediation of the government, can help the urban community in this smart growth, providing tools, knowledge, meeting points, training and learning methods that contribute to the co-creation of value for the entire community. For these reasons co-working could be qualified as component of Smart Local Service System (SLSS). Such local systems conceived as a “service” system of inte‐ grated and dynamic resources (people, technology, organizations) create and offer serv‐ ices for the extension of co-creation value. A SLSS thus conceived is able to provide the structural conditions for collaborative compatibility, sharing the same evolutionary purposes and active participation in proposing and co-creating value for the stakeholders of the context, based on the positive externalities arising from the law of increasing returns and a win-win value distribution logic [31]. Undoubtedly problems may arise in this regard on the governability of these dynamic and highly collaborative environments. The viability of a SLSS requires, in fact, intelligent or Smart government able to mediate the expectations of the various stakeholders and enabling shared projects based on the promotion and coordination of the territory as a network. Active governance (governa‐ mentality) must be guaranteed - according to bottom-up logic - in which the shared paths of government authorities and a new configuration in collaborative decision-making practice exist [30]. In this sense, co-working spaces respond concretely to such a theo‐ retical conception, because, thanks to technological applications, they enable increased

T-Shape Professionals Co-working in Smart Contexts

185

productivity and better relationships that transcend the merely physical boundaries of the aggregate space, ensuring, at the same time, the peculiar context specificity [32].

4

Practical Implications: VEGA(ST) Case Study

The Venice Gateway for Science and Technology is a Science and Technology Park network that involves universities, research centers and the productive sector. Its goals are to promote and develop scientific research initiatives and to facilitate the transfer of knowledge for technological growth and business competitiveness. It operates in the areas of technological innovation: Nanotechnology, ICT, and Green Economy. VEGA(ST) was created to provide networking opportunities and to set in motion, not only the progress of a specific territory by enhancing its peculiarities or individual companies, but also the progress of the country as a whole. VEGA(ST) is an example of a system that places interest in the prospect of sharing and searching for co-creation processes to achieve continuous and integrated develop‐ ment, typical of a Smart Local Service System. As a private company with a public majority (Municipality of Venice, Eni Syndial, Veneto Inn Veneto Region), VEGA(ST) is made up of 34 public and private partners to promote the development of the industrial area of Porto Marghera through the creation of research centers, business and services. Currently the structure consists of 80,000 m2 of offices and services more than 200 companies. In total VEGA(ST) comprises more than 1,700 employees counting users and the “condominium Venice Gateway”. An incubator facility with 20 start-ups, “Vegaincube”, works in the technology park, the incubator is also integrated in “FabLab Venice”. Co-working spaces, such as FabLab and the incubators, are the open innovation fulcrum, since, also in ASV and SSME+D, they are combined with the need to establish relational exchanges which organizations need in order to survive and to foster competitiveness. These spaces, therefore, are where partnerships are the key to the new way of co-creating value. The technology park is made up of 85 companies, 44% operate in ICT, 17% Green Economy, 13% cultural heritage and tourism development, 18% other professions, 18% of turnover being invested in R&D. The annual revenue amount to 250 million of EUR, and 18% of the tournover is invested in R&D. Every 1.5 companies boast patents. The promoting and development of scientific research stimulate the establishment of new initiatives and the development of innovative entrepreneurial ideas. This shows how the convergence of resources and expertise in a finite, but open local context generates highly creative and innovative results that favor synergies, relations and projects between entrepreneurs operating in the Park. The goal of VEGA(ST) is to facilitate relationships between local actors (individuals and organizations) and supporting their technological and competi‐ tive development through applied research activity carried out by the laboratories of innovation. By Service Science Approach, a service system could co-create value if resources are properly organized for value propositions which define the desired outcome. In fact, they argue that the foundations of a service system are:

186

P. Piciocchi et al.

– a dynamic configuration of resources; – a set of value co-creation mechanisms between suitable entities; – an application of competencies-skills-knowledge any person(s) in job or stakeholder roles; – an adaptive internal organization responding to the dynamic external environment; – learning and feedback to ensure mutual benefits or value co-creation outcomes. Therefore, value is considered to be co-created through a combined effort of several entities across all service sectors and value is determined by the beneficiary, for instance the local territory. What is important is to learn how the relational structure assembled to generate service value and how this is impacted by technology. This perspective should explain how to invest (internal, external, and interface) in exploration and exploi‐ tation. Investment is required to attain higher value-creation, value-capture, and oppor‐ tunity share future states. So as to use these Service Science concepts in a new paradigm shift in order to innovate through knowledge creation and transfer, we can make these acknowledgments converging into the study of a new skilled profile of professionals in a new efficient way of working. The dual challenge of creating T-shaped professionals and smarter systems can be summarized as follows: (1) A change in human capital is required to make collaborative innovation truly successful, then the challenge is that educating T-shaped people; (2) educationT-shaped people is fundamental for creating synergies as such as in collabo‐ rative systems of co-working. In fact, people with T-shaped skills are extremely valuable for creating knowledge because they can integrate diverse knowledge assets. The T-shaped professionals/organizations are characterized by deep, expert, contrib‐ utory expertise and broad complex communications in interactions: they have the ability both to combine theoretical and practical knowledge and to see how their branch of knowledge interacts with other branches. Therefore, they can expand their competence across several functional branch areas, and thus create new knowledge for themselves and the others. There is a positive relationship between the presence of the organizational members with T-shaped skills and knowledge creation process, so to say, it is easier that the SECI spiral [21] improves on with T-shaped people. T-shaped knowledge resources have at least two advantages: lower communication and collaboration costs; lower learning and adaptation costs. The disadvantage is that they imply major cost for the initial investment but the positive externalities of their use - communication, collaboration, learning, and adaptation - are suitable to recoup the design and system costs according to scale and scope economy laws. So, integration of specialized figures is crucial also in this context. Such profiles should be capable of facilitating interaction between various resource holders to provide for: (1) fostering a culture of innovation; (2) generating and intercepting good business ideas, products, services, mainly based on innovative technologies, the exploitation of research results, methods of contam‐ ination of knowledge and skill levels and different experience;

T-Shape Professionals Co-working in Smart Contexts

187

(3) favoring Start-ups of new local entrepreneurial, innovative and/or traditional players, promoting comparison and exchanges among the carriers of ideas or more established businesses, to start motivational growth paths of business culture capable of raising the quality and competitiveness of local production systems and creating business networks; (4) developing existing enterprises, encouraging consolidation processes on the market, innovation, networking and internationalization. The VEGA(ST) project is therefore an example of glocal development in which urban spaces, according to smart logic principles, are thus regenerated and put to full use by institutions or companies, the private sector, universities and individuals with high performances in specialization, communication and sharing. In this way competi‐ tiveness increases competitiveness not only of each involved actors, but of stakeholders as a whole. The challenge is to avoid the misalignment among the different actors, that would make VEGA(ST) a mere place of “hospitality”. The aim is to create non-random nature of connections. However, at present, there is no business model linked to the parks to facilitate the interconnection, even sometimes lack the basics right (conventions, standard), on which the actors can develop cooperative work. Such parks should also promote the skill of “doing business” that startups often do not have. In addition, they should strengthen their capacity to relate with other inoovators in other parks, univer‐ sities, businesses, administrative contexts. Moreover, socio-political problems are taken into consideration as they are strictly connected to the typical character of Italian terri‐ tory. The instability of governments, the fragility of their programs for innovation, the marginal role given to universities and poor economic sustainability [33] characterize the socio-political climate.

5

Conclusions and Research Limitation

The issues addressed albeit only in part, characterize the status of the sharing economy scenario. In a time of crisis like the current, a collaborative approach, rather than struc‐ tural approach, can be considered for the interpretation of new management practices. Indeed, from a systems and open innovation perspective it becomes crucial the involve‐ ment and the circulation of shared knowledge. To support this process we consider that T-shape professionals can contribute to the value co-creation processes within the socalled “smart context of co-working”. This means, in the search for models for intelligent (Smart) management of the territory, the often merely speculative, logical top-down governance pathway based on a vision of individual and unilateral power, should be rejected in favor of a reticular rationale promoting sharing and collaborative governance by all impacted stakeholders [34–36]. This approach integrates the form and personality of a SLSS with resources, both in terms of value proposition processes (distinct and sustainable competitive advantage) and of value co-creation and value-distribution (positive externalities of the Whole Service). To achieve such processes and promote the transformation and/or the emergence of new models of collaborative processes, the perspective of analysis should

188

P. Piciocchi et al.

go in the direction of consonant and resonant interactions of sharing. The new wide‐ spread systemic and active innovation logic, whereby the exchange of information and resources also enables the community to take advantage of the context of evolutionary paths becomes a must for the development of the entire local community. Undoubtedly, this work reveals various limits especially in terms of analysis of more specific instances. However, it does represent an attempt to give value to the so-called Smart perspective, where the primary objective is to circulate information and knowledge, reconfiguring territorial spaces according to a perspective of complex systemic subjectivity (SLSS). In this respect, the “VEGA(ST)” project provides useful indications in support of the new competitive collaborative arrangements, adequate to respond to the dynamics and variability of contexts. In a future scenario, more empirical evidence will enable the analysis in depth of the processes of collaboration and co-creation of value in a glocal perspective, however, this must involve new training programs to enable future profes‐ sionals to adapt to the dynamic and growing specialization of open and innovative contexts.

References 1. Porter, M.E., Kramer, M.R.: The big idea: creating shared value. Harvard Bus. Rev. 89(1), 2 (2011) 2. Bain, J.S.: Relation of profit rate to industry concentration. Q. J. Econ. 65, 293–324 (1951) 3. Mason, E.S.: Price and production policies of large-scale enterprise. Am. Econ. Rev. 29, 61– 74 (1939) 4. Arthur, W.B.,: Why a Population Converges to Stability. IIASA Working Paper. IIASA, Laxenburg, Austria, WP 79-085 (1979) 5. Piciocchi, P., Spohrer, J., Bassano C., Giuiusa, A.: Smart Governance to mediate human expectations and systems context interactions. In: Spohrer, J.C., Freund, L.E. (ed.) 18 Advances in the Human Side of Service Engineering, pp. 319–328. Taylor & Francis Ltd. (CRC Press) (2012). ISBN: 9781439870266 6. Spohrer, J., Piciocchi, P., Bassano, C.: Three frameworks for service research: exploring multilevel governance in nested, networked systems. Serv. Sci. 4, 147–160 (2012) 7. Maglio, P.P., Spohrer, J.: Special issue on Service Science, Management, and Engineering. IBM Syst. J. 47(1) (2008) 8. Golinelli, G.M., Gatti, M.: L’impresa sistema vitale. Il governo dei rapporti inter-sistemici, p. 53 (2000–2001) 9. Cinquini, L., Di Minin, A., Varaldo, R.: Nuovi modelli di business e creazione di valore: la Scienza dei Servizi. Springer, Milano (2011) 10. Chesbrough, H.: The era of open innovation. R&D Manage. (2003). Special Issue: Open R&D and Open Innovation 11. Donofrio, N., Sanchez, C., Spohrer, J.: Collaborative innovation and service systems: implications for institutions and disciplines. In: Grasso, D. (ed.) Holistic Engineering Education (2010) 12. Spohrer, J., Anderson, L., Pass, N., Ager, T., Gruhl, D.: Service science. J. Grid Comput. (2007). Special Issue on Grid Economics and Business Models 13. Spohrer, J., Kwan, S.K.: Service Science, Management, Engineering, and Design (SSMED): an emerging discipline - outline & references. Int. J. Inf. Syst. Serv. Sect. 1(3) 1–31 (2009)

T-Shape Professionals Co-working in Smart Contexts

189

14. Spohrer, J., Maglio, P.P.: Toward a science of service systems: value and symbols. In: Maglio, P.P., Kieliszewski, C.A., Spohrer, J.C. (eds.) Handbook of Service Science. Springer, New York (2010) 15. Kraiger, K., Ford, J., Salas, E.: Application of cognitive, skill-based, and affective theories of learning outcomes to new methods of training evaluation. J. Appl. Psychol. 78, 311–328 (1993) 16. Freeman, C.: Networks of innovators: a synthesis of research issues. Res. Policy 20(5), 499– 514 (1991) 17. Senoo, D., Magnier-Watanabe, R., Salmador, M.P.: Workplace reformation, active ba and knowledge creation: from a conceptual to a practical framework. Eur. J. Innov. Manage. 10(3), 296–315 (2007) 18. Ravicchio, F., Repetto, M., Trentin, G.: Formazione in rete, teleworking e inclusione lavorativa. Franco Angeli, Milano (2015) 19. Van Winden, W., De Carvalho, L., Van Tuijl, E., Van Haaren, J., Van den Berg, L.: Creating Knowledge Locations in Cities. Innovation and Integration Challenges (2012) 20. Amin, A., Roberts, J.: Knowing in action: beyond communities of practice. Res. Policy 37(2), 353–369 (2008) 21. Nonaka, I., Takeuchi, H.: The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. Oxford University Press, Oxford (1995) 22. Ellison, N.B., Steinfield, C., Lampe, C.: The benefits of Facebook “friends:” social capital and college students’ use of online social network sites. J. Comput. Mediated Commun. 12(4), 1143–1168 (2007) 23. Hutchins, E.: Organizing work by adaptation. Organ. Sci. 2(1), 14–39 (1991) 24. Lave, J., Wenger, E.: Situated learning: Legitimate peripheral participation. Cambridge University Press, Cambridge (1991) 25. Moriset, B.: Building New Places of the Creative Economy. The Rise of Coworking Spaces (2013) 26. Leforestier, A.: The co-working space concept. CINE Term Project. Indian Institute of Management (IIMAHD), Ahmedabad (2009) 27. Link, A.N., Scott, J.T.: US university research parks. J. Prod. Anal. 25(1–2), 43–55 (2006) 28. Johannenssen, J.A., Olsen, B., Olaisen, J.: Aspects of innovation theory based on knowledge management. Int. J. Inf. Manage. 19(2), 121–139 (1999) 29. Lee, H., Choi, B.: Knowledge management enablers, processes, and organizational performance: an integrative view and empirical examination. J. Manage. Inf. Syst. 20(1), 179– 228 (2003) 30. Piciocchi, P., Siano, A., Bassano, C., Conte, F., Smart Local Service System. “Governamentalità intelligente” per la competitività del territorio, p. 232, Conference proceedings: XXIV Sinergie Conference (2012) 31. Gummesson, E.: Marketing as Networks: The Birth of Many-to-Many Marketing. Publishing House Djursholm, Stockholm (2009) 32. Piciocchi, P., Saviano M., Bassano, C.: Network creativity to reduce strategic ambiguity in turbulent environments: a viable systems approach (VSa). In: Proceedings of the 11th International Conference on Global Business & amp; Economic Development. University of Bratislava, Slovak Republic, vol. 3, pp. 27–30, May 2009 33. Cozza M.: I parchi scientifici-tecnologici: da strutture insediative a infrastrutture connettive, POLISpóliV, XXVIII, 3, pp. 393–416, December 2014 34. Triantafillou, P.: Conceiving “Network Governance”. The Potential of the Concepts of Governmentality and Normalization. Working paper 2004/4. Centre for Democratic Network Governance, Roskilde, May 2004

190

P. Piciocchi et al.

35. Piciocchi, P., Bassano, C.: Governance and viability of franchising networks from a Viable Systems Approach (VSA). In: Proceedings of the 2009 Naples Forum on Service. Service Dominant Logic, Service Science and Network Theory, Capri, 16–19 June, Giannini Editore, Napoli (2009). ISBN13: 978-88-7431-452 36. Trunfio, M.: Governance turistica e sistemi turistici locali. Modelli teorici ed evidenze empiriche in Italia. Giappichelli. Torino (2008)

Worker and Consumer Factors in Service Engineering

Work-Related Factors Affecting Sustained Alert State Among Bank Security Personnel in the Philippines Yoshiki B. Kurata1,2 ✉ and Aura C. Matias1 (

)

1

2

Department of Industrial Engineering and Operations Research, University of the Philippines Diliman, Quezon City, Philippines {ybkurata,acmatias}@up.edu.ph Department of Industrial Engineering, Technological Institute of the Philippines, Quezon City, Cubao, Philippines

Abstract. Around-the-clock operations in public safety and security are perva‐ sive in modern society as crimes are increasing in rate. This drive security service operatives to work for long hours especially those deployed in high risk industries like banks. These working states create significant physiologic challenges in providing safe and productive operations. Alertness levels vary due to numerous factors inherent to individual, psychological, and working condition. This study is intended to investigate the correlation between all identified variables and crit‐ ical factors affecting the sustained high alertness level for effective and maintained surveillance among bank security personnel for increased operational efficiency through ergonomic assessment. Overall, results show that additional tasks, hour of the work day, and smoker/non-smoker are useful predictors associated with alertness level of security bank personnel. Recommendations were geared to minimize the unfavorable outcomes of decreasing alertness in providing sustained attention. Keywords: Alert state maintenance · Sustained attention · Security · Banks · Cognitive ergonomics

1

Introduction

Around-the-clock operations in public safety and security are pervasive in modern society as crimes i.e. theft and robbery are increasing in rate [1]. Security services in high risk industries like banks commonly practice shift work and long work hours as mandated in the Department Order No. 141 of Department of Labor and Employment [2] which defines a compulsory 12 h per shift schedule for security personnel. This shift length requirement creates significant physiologic challenges in sustaining high alert‐ ness level. In studies regarding long work hours, personnel stated getting less than enough sleep and rest which caused declining alertness levels on their respective duties [3]. Alertness is a state in which a person can think clearly and do a given task safely, 1

Guidelines Governing the Employment and Working Conditions of Security Guards and Similar Personnel in the Private Security Industry (DOLE, series of 2001).

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_18

194

Y.B. Kurata and A.C. Matias

effectively and efficiently. Thus, alertness in any work environment is an important factor to reinforce as it is regarded to be directly proportional to productivity and safety. Alertness of workers becomes a basic concern to any work environment but critical to safety-sensitive settings such as transportation, health care, and public safety including security services. This study will give emphasis on the factors that affect alertness and safety. Approaches to resolve or minimize issues concerning these risk factors must occur in all levels, from the individuals to organizational. 1.1 Bank Security Industry Any bank is a highly -secured place due to the sensitive transactions it handles. Knowing this, security personnel are commonly stationed on these places but occurrences of undesirable events which concern security still occur e.g. bank robbery, theft, etc. Security personnel are trained to manage this kind of critical situation with the goal of keeping everyone in the premises safe and avoiding casualties. Alertness and a good state of mind greatly affect the ability of security personnel to attend to these situations. Thus, maintaining a sustained good alertness level all throughout the shift will ensure effective surveillance and intuitive thinking in case of emergencies. Alertness levels vary due to numerous factors inherent to the individual, immediate environment and working condition. Examining the relationship of these factors can provide valuable insights on how policies can be improved for improving the welfare of security personnel and to increase their operational efficiency. 1.2 Statement of the Problem According to researches, long work hours lead to decrease performance due to accu‐ mulated fatigue, as well as insomnia and individual’s social life disorders [4]. In addi‐ tion, working for long hours may result to short-term and long-term effects among the people, which includes changes in the sleeping pattern and circadian rhythm, perform‐ ance, and safety. Long-term effects include gastrointestinal, cardiovascular, and muscu‐ loskeletal problems [5]. It is also known that poor sleep quality leads to difficulty in concentration which causes errors and increases the risk of accidents [6]. To date, research about shift work among security officers is well established in many countries worldwide, yet it is not yet an area of research attention in the Philippines. Based from the Department Order No. 14 of the Department of Labor and Employment [2], security officers are mandated to work for 12 h in a day. Due to the work consider‐ ations among security personnel working in high risk industries like banks, it is a common concern if the work schedule is optimal with respect to the nature of their work. With this data in mind, the researchers see the need for the study to be conducted. 1.3 Objectives of the Study In tackling issues concerning the welfare of bank security personnel, assessment of critical factors affecting alertness during their work shift is vital to provide safe and productive operations. This study aims to identify the critical factors affecting the

Work-Related Factors Affecting Sustained Alert State

195

sustained high alertness level for effective and maintained surveillance among security personnel during their work shift. The general objective can be achieved by the following specific objectives: 1. To statistically describe the demographics and determine significant trends inherent to the target population. 2. To assess individual factors and psychological factors of security personnel. 3. To measure the work condition brought by work schedule and tasks done by the workers. 4. To measure alertness level through Psychomotor Vigilance Task (PVT). 5. To create a framework that shows the relationship between individual factors, psychological factors, work condition, and alertness using Logistic Regression. 6. To identify the critical factor among the variables in the framework through Binary Logistic Regression. 7. To prepare recommendations based on the identified risk factors which can be used as a reference on policy modification among security personnel’s labor laws. 1.4 Scope and Limitations This paper only considered the individual factors, psychological factors and work condi‐ tion as factors affecting alertness. This study was also limited to security personnel rendering their service to banks in day shift or any similar establishments that keep safe tangible assets or money in the National Capital Region. The respondents recruited were categorized in two groups: 1) regular employees and 2) contractual employees on a certain security agency working on one specific bank. Also, since only one security agency of a certain bank agreed to participate on the study, the researchers only consid‐ ered one cluster area of the bank, specifically the bank’s branches in Quezon City, which the participating security agency is handling. In addition, this study was limited to gathering psychophysical measurements of alertness in their present work time schedules. Measuring alertness through physiolog‐ ical indicators is beyond the scope of this research as it requires specialized equipment and laboratory tests. In lieu to this, the researchers will utilize the Psychomotor Vigilance Test (PVT) as a means of measuring the alertness level objectively. Also, variables that will be considered in the study are those factors that are present during the actual time of work, where alertness is an utmost concern to be maintained. Security personnel are tasked to do surveillance of the bank premises alongside with other tasks given to them i.e. assisting customers which can further divide their attention. Sustaining alertness at this level poses a challenge to work designers as physical and psychological limitations varies from one person to another. Reviewing previous researches gives light to the factors affecting work performance in general but only few have studied on an industry specific basis. There are reports describing security person‐ nel’s work definition and scope but limited have been documented regarding critical factors affecting their work performance. No study has previously tackled direct meas‐ urements of aspects detrimental to the job i.e. alertness and attention. Moreover, Posner and Boies [7] recommended three major modules of attention where two of which can

196

Y.B. Kurata and A.C. Matias

be integrated to security personnel’s training. Despite the presence of these modules, methods on alert state maintenance had been identified to be a gap in the literatures. Alert state maintenance is still an untapped area in the Philippines’ security service industry (Fig. 1). Psychological Factors 1. Fatigue Perception 2. Employee Satisfaction 3. Mood 4. Anxiety Level Work Condition Individual Factors 1. Age 2. Sex 3. Body Mass Index (BMI) 4. Smoker / Non-smoker 5. Alcohol Consumption 6. Physical Fitness 7. Sleeping Habits

Performance Measurement

Alertness Level

1. Sitting vs. Standing 2. Temperature / Humidity 3. Illumination 4. Number of Bank Customers Present 5. Day of the Work Week 6. Hour of the Day 7. Additional Tasks 8. Sound Level 9. Work Location

Fig. 1. Conceptual framework

2

Methodologies

In determining the sample size, the study utilized Cochran’s [8] sample size formula used in categorical data. The initial calculated minimum sample size was 384 respond‐ ents. Since the total number of the population is 128 security personnel which is lesser than the computed sample size, Cochran’s correction formula was used to obtain the correct sample since only one security agency agreed to be the participating agency. Alpha level used in this study was 0.05 while the acceptable margin of error was set to 3% as this percentage is deemed acceptable for continuous data [9]. Estimation of variance, the study used a 5-point scale with six standard deviations to capture 98% of all respondents. There were five measurement tools used in the study Fatigue Risk Assessment (FRA) [10], Generalized Anxiety Disorder Assessment (GAD-7 Scale) [11], Brunel Mood Scale (BMS) [12], Berkeley Lab 2010 Employee Survey [13] and Psychomotor Vigilance Task (PVT) [14]. The PVT is a reaction time test commonly used by researchers to assess person’s alertness. The researchers distributed the required number of survey questionnaires and provided an alertness measurement test to over 128 security personnel working in banks in Quezon City. In order for the researchers to gather relevant results, the survey questionnaires and alertness measurement test admin‐ istered were bounded into the following criterion: a) the selection of the respondents was done in a randomly manner working in a certain bank company in Quezon City, b) the security personnel who can answer the survey questionnaires and can be measured by the alertness tool must be between 18 to 60 years of age and can be of any gender, and c) the security personnel before qualifying as a respondent must be at least six

Work-Related Factors Affecting Sustained Alert State

197

months working in the bank. This study utilized PVT to measure each employee’s alert‐ ness through the response time elicited on the application. All 120 samples were gathered based on the security personnel assigned with in the bank’s operating hours. The researchers collected three samples for each hour for 8 h. During the test, a visual stim‐ ulus is shown at random intervals, and the user presses a single physical button in response to each stimulus over a 5–10-minute period. Reaction times are reported in milliseconds, and various statistical summaries of these times have been shown to correlate well with wakefulness. Demographic data as well as information from their current job were based from a face-to-face interview with the representative of the security agency who manages the daily manning operation. This research considered necessary research protocols for human subjects. The aim of this research is to investigate on the critical hours of the day when alertness decreases to help the personnel be more effective and meet expected performance. There are no implications on the personnel’s employment conditions. For the fatigue aspect of the sample security personnel, the researchers utilized the Fatigue Risk Assessment to assess the risk arising from the fatigue associated with work patterns for safety critical workers. To measure perceived anxiety level brought by work schedule and long work hours, Generalized Anxiety Disorder Assessment was applied as the tool for measuring the security personnel’s anxiety level based on the factors. For the mood encountered by the security personnel be described, the researchers used the Brunel Mood Scale to obtain the necessary data. In assessing the employee satisfaction of the security personnel, the researchers used the Berkeley Lab 2010 Employee Survey assessing the different factors. For the individual factors, and work condition, data was obtained from a validated construct where questions were answered based on the subjec‐ tive perception of the security personnel. In treating the data gathered, this study utilized Minitab 17 statistical tool for the development of models and analyses. Summary of the results will be presented through descriptive statistics and cross tabulation to see inherent trends. Correlation analysis among the factors of the independent variables will be performed in order to test the strength and direction of the inter-variable relationship. Then, factors that will emanate from each independent variable will be further processed to logistic regression to model a relationship between the useful predictors to alertness and predict the effect of the series of variables on a binary response variable.

3

Results and Discussion

Using Cochran’s [8] sample size formula, computed sample size for this study corre‐ sponds to 384 samples. This makes the computed sample size greater than the actual popu‐ lation of 128 security personnel. With this, the researchers opted to use the Cochran’s [8] correction formula to determine the corrected minimum sample size of 96 samples. In addition, based from the designed experiment, three (3) security personnel will be the respondents on a specific day and hour of banking; instead of using 96 samples only, the researchers were able to compute to have 120 samples (Three (3) security personnel per hour of the day × 8 working hours per day × 5 working days in a week).

198

Y.B. Kurata and A.C. Matias

Descriptive statistics of the surveyed security personnel is summarized in Table 1. Psychological factors are also included in Table 1 because values were derived from the computation of the tool used. Table 1. Description of continuous data from survey Variable

Unit

Fatigue perception Employee satisfaction

Fatigue risk index Employee satisfaction index Mood index Anxiety level index Years

Mood Anxiety level Age BMI

kg/m2 Sticks/day

Nicotine consumption ml/day Alcohol consumptiona Hours Sleeping habits (average hour per sleep) Temperature oC Illumination Lux (inside bank) Illumination Lux (outside bank) Noise Decibel Number of Number of bank customers customer present Number of Number of additional tasks tasks

Minimum value 4.6

Maximum value 8

Mean

169

288

229.0583

26.86

0.1 0

6.7 28

2.1283 11.1833

1.36 4.62

23 16.5

57 31.2

37.0667 23.5147

7.77 3.55

0

20

4.0250

5.70

0

57.4

4

10

5.2580

53.2500 7.0750

26.7833

Standard deviation 0.46

35.5 1.30

17

35

300

380

341.3

26.22

6.63

1002

1971

1529.84

292.73

41 0

95 11

75.4750 3.7500

16.32 2.67

1

4

2.0917

0.83

a Alcohol consumption computation = average percentage of alcohol by volume (ABV) in millilitres x average volume consumed a day before the PVT tapping

In relation to alcohol consumption, the computation in identifying the alcohol level of the security personnel was as follows: (1) identifying the alcohol concentration percentage (% concentration) of the liquor consumed, (2) multiply the percent concen‐ tration to the volume of the liquor per bottle, (3) multiply to the average number of

Work-Related Factors Affecting Sustained Alert State

199

bottles consumed per day. Generally, security personnel’s average fatigue is considered moderate, employee satisfaction has an average of somewhat satisfied, and the level of anxiety is considered moderate anxiety. High variability in the illumination outside the bank can be observed due to the hour of the day and weather condition when the data were gathered (Fig. 2). Probability Plot of Response Time Normal

99.9

Mean StDev N AD P-Value

99 95

Percent

90

355.4 58.72 120 0.475 0.237

80 70 60 50 40 30 20 10 5 1 0.1

200

250

300

350

400

450

500

550

Response Time

Fig. 2. Probability plot of PVT response time

Assessing the normality of data is a prerequisite before running statistical tests because sample data must be drawn from a normally distributed population. After subjecting the PVT response times to normality test in Minitab, the test generated a pvalue of 0.237 which indicates that at 0.05 level, the sample data was significantly drawn from a normally distributed population. To better understand the connection between the variables, this study used correla‐ tion analysis to test the degree or strength of relationship through Pearson’s correlation coefficient for continuous data while Spearman Rank correlation for categorical data and the corresponding p-value. To run the correlation analysis, this study utilized Minitab 17 to compute the correlation coefficient and p-value. P-value was used in order to obtain a more reliable basis in determining possibility of correlation. Fatigue has a weak positive correlation to Mood while it has a weak negative corre‐ lation to sleep. This implies that as sleep decreases fatigue increases. Mood on the same hand has a weak positive correlation to fatigue while has a weak negative correlation to BMI. It can also be seen there is a weak negative correlation between BMI and smoking while smoking has a positive moderate correlation to alcohol. There is a strong positive correlation between illumination, sound level and temperature. Results generated by the Spearman’s rho correlation between categorical data shows that correlation significant at the 0.05 level were between employee satisfaction and

200

Y.B. Kurata and A.C. Matias

work location with a positive but weak correlation. Anxiety level has a positive but weak correlation to physical fitness. 3.1 Binary Logistic Regression Binary logistic regression is an appropriate test since only two discrete outcomes, being alert or not alert, are involved in the study where value “0” pertains to PVT response indicating low alertness level while “1” indicates a satisfactory to high alertness level adequate for the security personnel to do their intended tasks. This test used the 95% confidence interval with an alpha of 0.05 to be the basis in determining which of the predictors are associated to low or high alertness level. The Hosmer-Lemeshow test also showed that the model fitted the data well, with pvalue of 0.208. In addition, the said methodology also provided which predicting factors are statistically significantly predicting variables affecting alertness level. These part of the data is summarized in the Tables 2 and 3. Table 2. Work condition predictors affecting alertness level in security personnel Dependent variable Alertness (Y1)

Work condition as predictors Additional tasks (X18) Hour of the day (X17)

P-value 0.000 0.000

Odds ratio 3.1898 *

Table 3. Individual factor predictors affecting alertness level in security personnel Dependent variable Alertness (Y1)

Individual factor as predictors Smoker/Non-Smoker (X8)

P-value 0.036

Odds ratio 0.9143

Based on the results generated by the test through logistic regression, additional tasks and hour of the day given to the security personnel mostly affect the alertness level of the security personnel as indicated by the odds ratio. Odds ratio compare the odds of two events, in this case the odds of being in a low alertness state or being in a high alertness state. In logistic regression, odds ratios compare the odds of each level of the response variable (alertness level i.e. alert or not alert). The ratios quantify how each predictor variables affects the probabilities of each response level. While odds ratios for the categorical data tells that during time of the day and additional tasks most likely affect alertness level the most. In the individual factors, smoker/non-smoker obtains an odds ratio 0.9143 which is less than the odds ratio of additional tasks in the work condi‐ tion category. Thus, odds ratio presented can be a basis for the magnitude of various risks factors of an outcome [15]. Based on the odds ratio results, it can be interpreted that for every increase in additional task given to the security personnel, the model predicts an increase of 3.1898 in the odds of being alert. While for smoker/non-smoker, the negative coefficient of -0.0896 and an odds ratio of 0.9143 indicates that the prob‐ ability of non-smokers to be not alert is 91.43% compared to security personnel who smokes. Results also show that none of the psychological factors come out as predictors of alertness level in security personnel.

Work-Related Factors Affecting Sustained Alert State

201

The variance inflation factors (VIF) are very good predictors if multicollinearity exists. The larger the variance inflation factors, the higher the multicollinearity present. In addition, it has been noted from other authors that variance inflation factors should not exceed a value of 5. Since the highest value of the variance inflation factor is 2.10 which is less than 5, the researchers conclude that there is no existing multicollinearity in the data.

4

Conclusions and Recommendations

Hour of the day deemed to be a strong predictor as validated by logistic regression due to the nature of human body to have different wakefulness state and internal body processes in different times of the day. As hour of the day progresses, trend of being not alert increases. Most of the recorded occurrences of security personnel to be not alert can be observed at 1:00 PM. It can be linked to the body’s circadian rhythm. It is also important to note that by studying the relationship of alertness and time/hour of the day can give job designers insights on how to maximize productivity or surveillance capabilities. As the number of additional tasks increases, chances of a security personnel not to be alert decreases. With regards to alertness of security personnel and additional tasks assigned, result shows that additional task is another strong predictor to alertness. Other tasks besides surveillance can continuously stimulate the senses thus preventing drows‐ iness and eventually decline alertness level. It is also important to note that increasing the number of other tasks more than the capacity of a certain personnel can also lead to too much divided attention, thus, negatively affecting performance. Factors related to work condition are controllable and can be mitigated by the management. Based on the research findings, most of the factors affecting alertness are associated to how the security agency and bank institution design work for their security personnel. Additional tasks, hour of the day and smoker/non-smoker must be given attention as these are the most critical factors identified affecting alertness. Results show that additional task is directly proportional to the alertness level; as additional tasks increases, alertness level among security personnel increases as well. In addition, hour of the day is indirectly proportional to the alertness level meaning as hour of the day progresses, alertness level among security personnel decreases. Recommendations made in this paper were based on the results of the statistical analysis of the individual, psychological and work condition factors of security personnel. The researchers also utilized survey questionnaires, direct observations and personal interviews. Recommendations given are generally simple and inexpensive to implement. These are: 1. Workspace area must allow unrestricted movement of the security personnel. 2. Conduct seminars or trainings on how to avoid day time sleepiness and increase alertness. 3. Lunch break and/or short breaks are vital for security personnel to rest and recover from fatigue. 4. Management can schedule position shifting (i.e. inside and outside) with respect to the time of the day.

202

Y.B. Kurata and A.C. Matias

5. Provide caffeinated drinks i.e. coffee, tea and cola to security personnel to keep them alert during the work shift most especially after lunch (1:00 PM). 6. Promote frequent hydration among security personnel to maintain good blood flow. 7. Short breaks of 15 min must be scheduled and followed strictly as it can benefit the health of the security personnel. Occurrence of absences and tardiness together with their causes must be strictly and carefully monitored. 8. More frequent movement or surveillance (promote walking or changing positions) improve blood flow to the head thus increasing alertness. 9. Alternate the assigned post (inside or outside the bank) of the security personnel more frequent especially in summer season.

References 1. National Statistics Office (2014). https://psa.gov.ph/sites/default/files/2014%20PIF.pdf 2. Guidelines Governing the Employment and Working Conditions of Security Guards and Similar Personnel in the Private Security Industry (DOLE, series of 2001) 3. Perlo-Freeman, S., Fleurant, A., Wezeman, P., Wezeman, S.: Trends in world military expenditure, 2014. In: SIPRI (2015) 4. Lehtimäki, A., Helander, M., Pessi, A.M.: Circadian periodicity of airborne pollen and spores; significance of sampling height. Aerobiologia 7(2), 129–135 (1991) 5. Spengos, K., Tsivgoulis, E., Tsivgoulis, A., Zakopoulos, N., Vemmos, K.N.: Circadian and seasonal distribution of cardioembolic strokes due to atrial fibrillation. Hellenic J. Cardiol. 45, 234–241 (2004) 6. Soehner, A.M., Kennedy, K.S., Monk, T.H.: Personality correlates with sleep-wake variables. Chronobiol. Int. 24(5), 889–903 (2007) 7. Posner, M.I., Boies, S.J.: Components of attention. Psychol. Rev. 78(5), 391 (1971) 8. Cochran, W.G.: Sampling Techniques, 3rd edn. Wiley, New York (1977) 9. Krejcie, R.V., Morgan, D.W.: Determining sample size for research activities. Educ. Psychol. Meas. 30, 607–610 (1970) 10. The development of a fatigue/risk index for shiftworkers. Health and Safety Executive. http:// www.hse.gov.uk/research/rrhtm/rr446.htm 11. Spitzer, R.L., Kroenke, K., Williams, J.B., Löwe, B.: A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch. Intern. Med. 166(10), 1092–1097 (2006) 12. Soos, I., Lane, A.M., Leibinger, E., Karsai, I., Hamar, P.: Validity of the brunel mood scale for use with UK, Italian and Hungarian athletes (2007) 13. Berkeley Lab 2010 Employee Survey. www2.lbl.gov/Ops/assets/docs/surveys/BLES-2010.pdf 14. Kay, M., Rector, K., Consolvo, S., Greenstein, B., Wobbrock, J.O., Watson, N.F., Kientz, J.A.: PVT-touch: adapting a reaction time test for touchscreen devices. In: 2013 7th International Conference on Pervasive Computing Technologies for Healthcare (Pervasive Health), pp. 248–251. IEEE, May 2013 15. Szumilas, M.: Explaining Odds Ratios. National Center for Biotechnology Information, U.S. National Library of Medicine, August, 2010. www.ncbi.nlm.nih.gov/pmc/articles/ PMC2938757/www.ncbi.nlm.nih.gov/pmc/articles/PMC2938757/. Accessed 6 Mar 2014

The Negative Effect that Benefit Delay in Service Consumption Causes to a Customer and the Countermeasure Against It Kazuhiro Fujimura ✉ (

)

Faculty of Economics, Kagawa University, 2-1 Saiwai-cho, Takamatsu, Kagawa 760-8523, Japan [email protected]

Abstract. In the consumption of services such as education and medical care, there exists a time gap between when service activities to produce benefits are actually performed and when, as a result, a customer can really enjoy benefits generated by that service. We have conceptualized such time gap as Benefit Delay and categorized it into three types: Occurrence Delay, Perception Delay, and Achievement Delay. In cases where any or all of three types of delay occur after the service delivery process, Benefit Delay has serious negative effects on customer satisfaction and customer participation. Thus, the objective of this research is to develop a framework to systematically assess Benefit Delay. Furthermore, our research aims to uncover the negative effects that Benefit Delay may cause for customers, and to seek ways to reduce these effects. By literature review and pre-survey, we posit three types of benefits associated with services, which we label Functional Benefit, Emotional Benefit, and Perspective Benefit. We conducted the survey by questionnaire for patients with chronic diseases in Japan to discover which types of delay occurs in the enjoyment of each benefit, and how these delays in turn affect customer satisfaction and customer partici‐ pation. We found that when any or all types of delay occur in delivery of Func‐ tional Benefit, customer satisfaction and customer participation can be still main‐ tained by providing Perspective Benefit. Keywords: Benefit Delay · Functional Benefit · Emotional Benefit · Perspective Benefit · Customer participation · Customer satisfaction

1

Introduction

1.1 The Concept of Benefit Delay Service is generally produced and consumed simultaneously, but customers cannot always enjoy the benefit of the provided service immediately on the spot. In service areas such as education and medicine, for example, where enhancement or improvement of customers’ ability or physical condition is expected, there exists a time gap between the point of time when service is provided and the point when customers can actually enjoy benefits of the provided service. We conceptualized such time gap as Benefit Delay [1].

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_19

204

K. Fujimura

However, Benefit Delay does not necessarily occur in educational and medical serv‐ ices. The reason is that occurrence of Benefit Delay highly depends on the ultimate goal level of the benefit that customer expects. By this we mean that the higher the level of expectation of the customer on the ultimate goal is, the larger the delay tends to be; and, the lower the goal is, the smaller the delay would be [2]. In the medical arena, for example, if the expected ultimate goal of a post-operative cancer patient is to become able to eat or walk on his own without aids, such a level of ability might be achievable relatively easily while they are hospitalized, and so there may not exist Benefit Delay. However, if an ultimate goal of the same patient is to become able to lead the same level of daily living or working activity as he enjoyed before his hospitalization for cancer treatment, the patient may need to go through longer period of rehabilitation and as a result, the degree of Benefit Delay will be much longer than in the case of easily achiev‐ able goals. Likewise, in educational service, if the ultimate goal of a student is to obtain required credits from courses for graduation, such goal can be achieved upon the completion of curriculum and by passing the exams, so there may exist no Benefit Delay when the final course is passed. However, if ultimate goal of a student is more difficult like to become a Nobel prize level of scholar, it will require much longer period of time after graduation from the university, if not impossible, and so the degree of Benefit Delay will be extended. In the case of Services that are expected to bring positive change on human ability or body such as educational and medical services, there is a hierarchy of the goal - means to achieve the ultimate goal that a customer expects. In addition, if there are many phases to go through, as a result, the degree of Benefit Delay becomes larger. Learning or medical treatment are step-by-step processes of smaller goals aimed at achievement of ultimate goal. As shown in Fig. 1, assuming that there is means a1 to achieve ultimate goal Z, there should be a middle goal A1. Furthermore, assuming that there are other means as b1 and b2 set to achieve middle goals A1, there should also be middle goals B1 and B2 respectively. The lower the phases are, the more concrete and easier the goals

Fig. 1. Hierarchy of the goal-means to determine the degree of Benefit Delay

The Negative Effect that Benefit Delay in Service Consumption

205

are likely to be. If the ultimate goal is set at lower level, there may be less middle goals, and as a result, the degree of Benefit Delay is less. In this context, whether Benefit Delay occurs or not and if it occurs, the degree of the delay, depends on the level of the ultimate goal that a customer expects to achieve. Thus, if we view Benefit Delay from the perspective of the time required to achieve the ultimate goals of customers, we can call it Achievement Delay. The process of achieving the ultimate goals that customers expect can be segmented into two phases, i.e., from the start point of benefit occurrence, and the point when customer can perceive benefit. This indicates that Benefit Delay occurs at each phase. For this reason, in this research, we call the time required from the start point of service delivery process until the point of occurrence of benefit as Occurrence Delay, and the time required from occurrence of benefit until the point when customer can perceive that they enjoy benefit of that service as Perception Delay. Thus, we define Benefit Delay to be composed of three types of delay: Occurrence Delay, Perception Delay and Achievement Delay as in Fig. 2. The reason why these types of delay occur is that service requires accumulation of various effects produced by activities of service employees and customers themselves, and by functions of equip‐ ment or devices used in the service delivery. In this sense, Benefit Delay would not occur when service activities or functions themselves directly and instantly bring benefits to their customers on the spot. We can call such type of services as Immediate Benefit Services (hereinafter “IBSs”). When the three types of delay occur where services require the accumulation of effects, they are referred to as Delayed Benefit Services (hereinafter “DBSs”).

Fig. 2. Three types of delay that constitute Benefit Delay

If not only the delivery of DBSs requires a long term such as several months or years, but also any or all types of delay continue to exist after the service delivery process, Benefit Delay may cause significant negative effects on their customer satisfaction and customer participation. This is because there is no appropriate understanding about degree and reason of Benefit Delay, nor any kind of countermeasures prepared to assist

206

K. Fujimura

customers to achieve their ultimate goals. If the customer starts to feel that the choice of the service was a failure after having consumed that service for some time, it is not easy for customers to stop using the service or to re-select alternative service. A discus‐ sion of this negative effect is presented in Sect. 1.3. 1.2 Benefits and Delays When Receiving Services We have stated that concept of Benefit Delay consists of three types of delay. If so, what are the delayed benefits? Benefit assumed in this concept is benefit that contributes to solve the problem that triggered customer to use the service, and is related to the funda‐ mental function and ability of the service. Hence, we can call it Functional Benefit. If we assume that customers evaluate service quality in view of output and process, Func‐ tional Benefit is related to evaluation of output quality. This indicates that the benefits related to process quality also exist. We would like to propose the concept of Emotional Benefit and Perspective Benefit in relation to process quality. In realizing Functional Benefits, three types of delay seem to occur. Functional Benefit is defined as benefit that generate positive changes on customer’s ability, knowl‐ edge, skill, psychological or physical health conditions as a result of fulfillment of fundamental needs that motivated customers to use that service and various relevant needs [3]. In the medical service, Functional Benefit refers to recovery from deteriorated health condition to the level that the patient aspires to reach. As the favorable effects generated by service activities accumulate, Functional Benefit positively changes. To realize this change, the service providers need to perform appropriately and also the customer need to participate positively and properly in the delivery process, in other words, to invest their consumable resources1 when performing activities required for the achievement of their ultimate goals. Emotional Benefit is a byproduct (added value) of the activities to produce Func‐ tional Benefit. It contributes to change customers’ emotional state toward positive direc‐ tion by removing causes that triggered the fundamental needs on service, and by suppressing negative emotions created in the achievement process [3]. In medical service, Emotional Benefit eliminates or reduces psychological anxiety caused by deter‐ iorated physical condition due to diseases, and also drives patients to experience pleasant emotion and hope of recovery in the medical treatment process. As represented by Fig. 3, positive changes that Emotional Benefit can bring is volatile, and emotional state of customers tends to change quickly at various events they encounter during the service delivery process. Emotions can be influenced by each event, so emotional state repeats short-term fluctuation. However, as a whole, Emotional Benefit is benefit that can enhance customers’ emotional state toward upper right direction as time elapses. Perspective Benefit is also associated with generating Functional Benefit. It produces positive changes in recognition, attitude, and priority of the fundamental needs that triggered customers to consume the services, and leads them in an appropriate direction. 1

Consumable resources mean resources that consumer themselves have to possess and invest in order to obtain benefit in consumption of goods and services, such as money, time, physical or psychological energy, knowledge, skills, supplementary good, spaces.

The Negative Effect that Benefit Delay in Service Consumption

207

Fig. 3. Positive change of customer’s emotional state by Emotional Benefit

In the medical area, current treatment techniques cannot cure all physical impairments, and there are certain numbers of incurable or partially curable diseases. Furthermore, the residuals from disease or operation scars are sometimes inevitable. To cope with such unwanted, undesirable conditions, Perspective Benefit can bring positive changes to patients by encouraging them to have positive thoughts, attitudes or concerns. Such positive changes are expected to have positive impacts in motivating patient participa‐ tion and maintaining patient satisfaction. Consequently, co-production with medical workers and patients can be performed smoothly and effectively. We suppose that this Perspective Benefit brings positive change in customers’ perspective or value. Such positive changes are expected to play an important role in managing ultimate goal of each customer, namely, qualitative change on expectation. If we follow a hypothesis that customer satisfaction is determined in comparison with the level of expectation against the offered performance, one way to increase customer satisfaction would be to enhance performance. Another way would be to manage the level of expectation on ultimate goals. However, lowering the level of the customer’s goal would not necessarily increase satisfaction. In the study on customer satisfaction, it has been evidenced that the level of expectation itself would influence the level of customer satisfaction, and when there is lower expectation, customer satisfaction would not increase even though performance exceeds such expectation [4]. In the educational and medical service, managing the ultimate goal of customer to a lower level would produce negative impact not only on the customers and service providers but also on the society. When the level of the customer’s goal is low and the achievement is rela‐ tively easy, the service provider may not be motivated to improve or innovate their services, and so motivation for learning by their employees or growth of the company may be deteriorated. In the education area, potential ability of the learners may not be sufficiently developed, so the society cannot secure highly talented people as workers, and furthermore, the economic growth may be suppressed. In medical area, if the patients do not recover, they cannot go back to their normal social activities to contribute to the society, and the society has to bear their medical expenses. Therefore, it is important to raise the level of expectation on ultimate goal rather than to lower it. Hence, if Perspec‐ tive Benefit produces perspective change toward positive direction and changes

208

K. Fujimura

customer expectation on ultimate goal properly, customer satisfaction can be enhanced while the level of expectation is maintained as desired. As shown above, DBSs are composed of Functional Benefit, Emotional Benefit and Perspective Benefit, and while three types of delay are occurring when receiving Func‐ tional Benefit, Emotional Benefit and Perspective Benefit are considered to supplement Functional Benefit in promoting customer participation and in maintaining or improving customer satisfaction. 1.3 Significant Issues Brought up by Characteristics of DBSs Apart from Benefit Delay, DBSs have two important characteristics. One is discomfort with customer participation in the service delivery process, and another is uncertainty regarding achievement of ultimate goals. Discomfort Held by Customers when Participating in the Delivery Process. In the delivery of Services, especially DBSs, how co-production is conducted has a significant influence on the benefits that the customers will be able to enjoy and those delays. To improve benefits received so as to achieve the expected level of the customer’s ultimate goals, or to improve customer satisfaction effectively and efficiently, it is not sufficient to improve only the quantity and quality of production resources or capability on the service provider side. It is also necessary to improve customer participation in the service delivery process, i.e., appropriate and active investment of consumable resources by customers themselves. For example, in the delivery of the medical service, medical workers have to motivate their patients to invest consumable resources such as times and efforts properly for rehabilitation. Customers may experience positive or negative emotions when, for co-production, they invest their consumable resources such as times, activities, and efforts. In consump‐ tion of service at amusement parks like at Disney Land, customers actively participate in the delivery process by investing their consumable resources of time and physical activity because such an investment generate pleasant and joyful emotions on them. On the other hand, in medical service, patients tend to arouse negative emotion such as unpleasantness or fear because they may experience physical and/or psychological pain during the medical treatment process. In educational service, learning processes like attending lectures and preparation or review for lectures are lonely activities for students that force them to sacrifice other joyful activities. Consequently, they experience unpleasant emotions. In the consumption of DBSs, more active customer participation and more investment of consumable resources are required in the service delivery process than in other type of services. However, customer participation in delivery process may be restricted because it not only generates negative emotions but also forces sacrifice of other joyful activities. Uncertainty regarding Achievement of Anticipated Ultimate Goals. As the delivery of DBSs requires the accumulation of various effects produced by activities, services may not always be able to produce benefit resulting in achievement of the customers’ ultimate goals. Furthermore, if it could achieve the ultimate goals, the time until

The Negative Effect that Benefit Delay in Service Consumption

209

achievement may be more prolonged than estimated at the beginning of the delivery unless these effects effectively and efficiently accumulate. For example, in medical services, treatment effects may vary depending on health and physical conditions. The effects may also differ by self-effort for recovery of individual patient even though same medication is offered. In educational services, the same thing can happen, i.e., the achievable level of knowledge or ability varies even though students attend the same lectures, as the outcomes depend on ability, experience or on the learning efforts of individual students. These two characteristics together with DBSs, may have two various serious impacts not only on customers but also on the service providers: (1) a demotivation in customer participation in the service delivery process and (2) a decline of customer satisfaction. The customers’ motivation for participation in consumption of medical or educational services tends to be suppressed when they feel uncertain whether expected benefits can be achieved at the cost of invested consumable resources and psychological and/or physical pains. It is assumed that as the degree of decline in participation motivation becomes larger when service delivery time is longer, and when the three types of delay are serious, the uncertainty whether or not customers can achieve the anticipated ultimate goal increases. In addition, the more serious the degree of three types of delay are, and longer the duration of delay becomes, customers’ motivation to invest consumable resource proactively into the service delivery process tends to decline. As such, the attempt to deliver service effectively and efficiently by the service providers will be prevented by such demotivated customers, and the possibility of achieving the ultimate goal diminishes. Customer satisfaction is also presumed to decline when any type of delay of Func‐ tional Benefit occurs. So, DBSs providers not only need to minimize the degree of delays in three types of benefit but also simultaneously improve or maintain motivation for participation in order to increase the probability of achievement of benefit customer expects. It is desirable that Emotional Benefit and Perspective Benefit play supplemen‐ tary role when any or all of three types of delay in offered Functional Benefit are occur‐ ring. The reason is that positive changes in customers’ emotion and perspective brought by Emotional Benefit and Perspective Benefit are expected to influence their behavior and evaluation favorably.

2

Methods

In 2014, we conducted questionnaire surveys on outpatients at four general hospitals (Hospital. A–D)2 located at Osaka Prefecture, Kagawa Prefecture, and Chiba Prefecture, in Japan. Generally, service delivery processes related to treatment tend to be longer for chronic diseases than for acute diseases; consequently, it can be difficult for chronic 2

Hospital A is a hospital with 565 beds in 35 departments in Osaka Prefecture, which is a member of the Japan Community Health Care Organization; Hospital B is a municipal hospital with 216 beds in 18 departments in Kagawa Prefecture, Hospital C is a hospital operated by a stock company with 300 beds in 21 departments in Osaka Prefecture; and Hospital D is a private hospital with 917 beds in 34 departments in Chiba Prefecture.

210

K. Fujimura

disease patients to receive Functional Benefit of medical service. This paper analyzes patients with chronic diseases, aged 20 years or older. Table 1 shows a summary of the samples (n = 1984). Table 1. Characteristics of all samples

All samples from the four hospitals (Hospital. A–D) were categorized into six groups based on the phase of medical treatment, in order to examine when three types of benefit are enjoyed, and what impact they have on customer satisfaction and participation. The six groups (Phase 1–6) were as follows: initial consultation (Phase 1), in preparation for treatment (Phase 2), undergoing treatment (Phase 3), in recovery period (Phase 4), regularly visiting hospital(s) (Phase 5), and nearly completely cured (Phase 6).

3

Results and Discussion

3.1 Difference in Delays Among Three Types of Benefit Delay, as represented by the concept of Benefit Delay, basically occurs when receiving Functional Benefit. However, as Emotional Benefit and Perspective Benefit are related to quality of process, there will also be delays influenced by the structure of delivery process, the ability, enthusiasm, ethical view, and style of participation of service employees, and the level of expected ultimate goal of customers. The sequential line graph in Fig. 4 indicates how the patient perception changes in different phases in receiving each benefit in percentage ratio. Here, the answers to following questions are set with five-point scale, with ‘not at all’ and ‘very much so’ as two extremes, and the indicated ratio of patients represents those who replied 4 or 5 at each phase.

The Negative Effect that Benefit Delay in Service Consumption

211

• Functional Benefit: My subjective symptoms have reduced more than expected. • Emotional Benefit: I have fewer anxieties about disease than before. • Perspective Benefit: By receiving medical service at this hospital, I have learned to live with my disease.

Fig. 4. Change in the percentage of the patient who perceived the enjoyment of each benefit

In all types of benefits, the number of patients who recognize benefit increases as medical treatment process progress, and suggests that delays occur while receiving any type of benefits. However, the largest difference between Phase 1 and 6 is on Functional Benefit (42.1 point) and delay is most significant. It also suggests that recognition of Perspective Benefit precedes Emotional Benefit, then after that, Functional Benefit follows. 3.2 Supplementing Effects Among Each Benefit Figure 5 indicates how the proportion of patients who replied ‘satisfied’ changes at each phase. Here, the question asks about the level of Overall Satisfaction with “not satisfied at all” and “very much satisfied” as two extremes, and then, Expectancy Disconfirmation with “far below expectation” and “far beyond expectation” as two extremes. Both

Fig. 5. Change in the percentage of the patient satisfied with medical service

212

K. Fujimura

answers are set with five-point scale, and the ratio of patients represents those who replied 4 or 5 at each phase. In either question, the ratio of patients who replied “satisfactory” increases as phases of medical treatment process advances. The level of satisfaction is higher in every phase when asked in Overall Satisfaction than Expectancy Disconfirmation. At Phase 6, 87.5% of patients answered they are satisfied when asked in Overall Satisfaction, but only 54% of patients are satisfied when asked in Expectancy disconfirmation. This indicates that Perspective Benefit and/or Emotional Benefit may supplement negative effect caused by Achievement Delay in Functional Benefit. The fact that some patients seem to be satisfied even at Phase 1 or 2, where no Functional Benefit can be experienced, also implies it. This result suggests that preceding Perspective Benefit and/or Emotional Benefit may contribute to forming customer satisfaction. Table 2 is the result of analysis by enjoyment pattern of three types of benefit in relation to the levels of Overall Satisfaction. The ratio of very satisfied patients is highest in Pattern 1 (44.0%), where all three types of benefits are received. The next highest is Pattern 3 (33.6%), Pattern 7 (28.0%) and Pattern 5 (27.7%), where Perspective Benefit is received regardless of the enjoyment of Functional Benefit. On the other hand, the least satisfied patients are Pattern 8 (5.8%) where none of benefit is received, and the next, Pattern 6 (8.3%) where only Emotional Benefit is received. From this result, it is assumed that Perspective Benefit has more important role in improving of customer satisfaction than Emotional Benefit does. Table 2. Relationship with enjoyment pattern of three types of benefit and overall satisfaction Enjoyment pattern of three sub-benefits Pattern

1 2 3 4 5 6 7 8 Subtotal

Overall Satisfaction

Functional Benefit

Emotional Benefit

Perspective Benefit

○ ○ ○ ○ × × × ×

○ ○ × × ○ ○ × ×

○ × ○ × ○ × ○ ×

Not very satisfied

Subtotal Very satisfied

Neither satisfied nor satisfied

3 (0.5)

11 (1.7)

48 (7.5)

297 (46.3)

282 (44.0)

0 (-)

3 (4.0)

18 (24.0)

43 (57.3)

11 (14.7)

641 (100.0) 75 (100.0)

0 (-)

3 (2.2)

12 (8.8)

76 (55.5)

46 (33.6)

137 (100.0)

1 (1.6)

5 (7.9)

20 (31.7)

27 (42.9)

10 (15.9)

63 (100.0)

1 (0.6)

5 (3.0)

25 (15.1)

89 (53.6)

46 (27.7)

166 (100.0)

1 (2.8)

1 (2.8)

8 (22.2)

23(63.9)

3 (8.3)

36 (100.0)

1 (0.6)

2 (1.2)

33 (20.1)

82 (50.0)

46 (28.0)

164 (100.0)

11 (3.7)

16 (5.4)

123 (41.7)

128 (43.4)

17 (5.8)

295 (100.0)

18 (1.1)

46 (2.9)

287 (18.2)

765 (48.5)

461 (29.2)

1577 (100.0)

Chi-square: 314.146*** ***p < 0.001 Notes a ○ means that a patient enjoyed each benefit, and × means that a patient does not enjoy it.

Based on the results, we conducted a structural equation modeling on the relationship between Functional Benefit, Emotional Benefit, Perspective Benefit, customer satisfac‐ tion, and customer participation. We have been able to detect the existence of relation‐ ship as indicated in Fig. 6, and amongst all, the following findings deserve special atten‐ tion. 1. Three types of benefit do not affect customer satisfaction independently.

The Negative Effect that Benefit Delay in Service Consumption

213

2. Perspective Benefit gives positive effect on Emotional Benefit that relieve negative emotion of customers. 3. Perspective Benefit and Emotional Benefit may influence the Perception of Func‐ tional Benefit. 4. Perspective Benefit gives positive effect on customer participation.

Fig. 6. The effects of three types of benefits on customer satisfaction and customer participation

Such results suggest that service providers of DBSs not only need to minimize three types of delay in delivering Functional Benefit but also to improve customer participa‐ tion and satisfaction by providing Perspective Benefit when any or all types of delay in delivering Functional Benefit is occurring.

4

Conclusion and Future Study

As service delivery is to be realized by service providers together with their own customers, receivable benefits and satisfaction that customer can enjoy depend not only on the quality and quantity of service providers’ production resources and their capa‐ bility to optimize those but also on consumable resources that customers can invest. However, we noted that delays in DBSs bring discomfort to customers through participation in the service delivery process and uncertainty regarding achievement of expected ultimate goal, and these factors may inhibit customer participation. Lower customer participation increases the degree of Benefit Delay and this might inflict a great loss on not only customers themselves but also on the society. Therefore, DBSs providers have to prepare means to decrease such negative effects. Our analysis suggests that offering Perspective Benefit may be effective in maintaining customer participation and customer satisfaction during the period when any or all of three types of delay in offered Functional Benefit are occurring. As Perspective Benefit may play important role to maintain and increase customer satisfaction and participation, research on what change in customers’ perspective influences their participation and satisfaction, and what means are effective when such positive change is required. Finally, we are in the process of analyzing the details of the result obtained by customer satisfaction survey in medical service, and are preparing our logic and model

214

K. Fujimura

to examine the concept of Benefit Delay for educational services. We would like to conduct a similar survey in educational service, and to build systematic logic and model on customer satisfaction and participation in the consumption of DBSs by comparing these results to those obtained for medical service. Acknowledgments. This work was supported by Center for Research and Development Strategy, Japan Science and Technology Agency.

Appendix: Observation Parameters of the Model Functional Benefit • My subjective symptoms have been reduced more than expected. • I feel less physical pain than before. Emotional Benefit • I have fewer anxieties about disease than before. • I feel less fear than before. • I have fewer anxieties about wellness in post-treatment everyday life than before. Perspective Benefit • I have the will to live positively in everyday life. • By receiving medical service at this hospital, a positive change has occurred in my outlook on life. • By receiving medical service at this hospital, I have learned to live with my disease. Customer Satisfaction • How are you satisfied overall with the clinical examination and/or treatment you receive at this hospital? • Was it good for you to have received treatment at this hospital? • Would you like to visit this hospital again for medical treatment in the future? Customer Participation • I took a lot of exercise that was suitable for my health condition. • I tried to form good relationships with physicians and nurses. • By receiving medical service at this hospital, I am now more proactive in doing activities that is required for my treatment than before.

References 1. Fujimura, K.: The issues related to customer satisfaction in the consumption of the professional service of the benefit delayed type: considering medical service as a case. Kagawa Univ. Econ. Rev. 81(1), 1–62 (2008) 2. Fujimura, K.: Reconsideration of “the Benefit Delay” as the key concept: toward the extension to education service. Kagawa Univ. Econ. Rev. 89(4), 1–60 (2017) 3. Fujimura, K.: Attempting to build the work motivation model of the staffs of the care service industry by introducing “Benefit Enjoyment” and “Benefit Delay” as key concepts. Kagawa Univ. Econ. Rev. 89(1), 15–85 (2016) 4. Oliver, R.L.: A cognitive model of the antecedents and consequences of satisfaction decisions. J. Mark. Res. 17, 460–469 (1980)

A Task Analysis of Small-Scale Jewelry Craft Workers to Investigate the Effects of Work System Elements and Sitting Mobility on Body Discomfort Hazel Caparas ✉ and Aura Matias (

)

University of the Philippines Diliman, Quezon City, Philippines [email protected], [email protected]

Abstract. Jewelry making is a labor-intensive process for which Filipino craft workers are very well known. Jewelry makers are considered as inactive workers and they must sustain contorted sitting posture for long periods. This study aims to investigate the factors of body discomfort by analyzing various ergonomics risk factors present in the workplace. Moreover, the task analysis intends to have a comparative assessment between symptomatic and asymptomatic workers. The methodology consists of Revised Oswestry Disability Questionnaire, Body Discomfort Assessment using Borg’s CR-10 Scale, Video Based Assessment, Rapid Upper Limb Assessment (RULA), Anthropometric and Workplace Design Components direct measurements. In addition, Multiple Linear Regression and Logistic Regression are used for data analysis. The result primarily shows that working height mismatch, sitting posture, working duration and sitting mobility significantly affect the perceived body discomfort. Furthermore, recommenda‐ tions are made to reduce the risk of developing work-related disorders. Keywords: Human factors · Jewelry making · Task analysis

1

Introduction

The Philippines has long been recognized to be a country with large natural gold deposits [1]. As a result, this immense natural resource eventually lead to the emergence of the jewelry industry in the country. Manufacturers of jewelry in the Philippines are mostly small enterprises, which up to now perform in the traditional ways of jewelry making. But, along with rich natural resources, labor skilled in design and production is also essential in the jewelry industry. Filipino craft workers are well known for their expertise in producing fine and costume jewelry by merely using tools controlled by hand. In fact, the Guild of Philippine Jewelers Inc. (GPJI) commends their artistry in this craft [2]. The group also gives emphasis on the availability of cheap but skilled labor resource in the country [3]. But, despite the emergence and growth of the jewelry industry in the country, an aging artist workforce and reduced numbers of new artists being trained poses a great threat to the needed human resource of expert Filipino jewelry makers. Younger generations prefer high-status jobs, thus foregoing the craft selected by their fathers. Moreover, the nature of work industry work practices and environmental

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_20

216

H. Caparas and A. Matias

conditions are probable contributing factors in ceasing the decline in newly trained jewelry workers. As reported by the National Profile on Occupational Safety and Health for the Philippines on 2006, less than 10% of 32.4 million workers benefit from effective occupational safety and health programs. This proportion is generally occupied by workers at medium to large-scale businesses. Taking into consideration these opportu‐ nities, the need to support the Filipino jewelry makers particularly those covered by small and medium scale industry is seen clearly. Ensuring the occupational safety and health of these craft workers can help in preserving the dying breed of traditional artists in jewelry making. The study aimed to investigate the causes of body discomfort by analyzing various ergonomics risk factors present in the work systems and environment of jewelry makers. In general, the current study considered the following factors: (1) Sitting mobility; (2) Sitting Posture; (3) Individual Factors; (4) Workplace Design Components; and (5) Environmental Factors. This work also identified which among these factors were corre‐ lated variables. It planned to create an interrelationship model to clearly understand the interdependency of level of body discomfort, frequency of sitting mobility and key risk factors in the jewelry making work system. There was a comparative assessment of the outcomes between symptomatic and asymptomatic jewelry craft workers. Finally, an acceptable range of values for identified ergonomics risk factors were recommended in the form of policy or work instructions. At the time of the current study, there was no available literature which focused on analyzing the work systems of jewelry craft workers. The current study was not a complete replication of any previous work, since most of the available related literature concerning measurement of body discomfort and postural changes were performed under controlled environment. Accordingly, the researcher validated the findings in natural set-up of jewelry making workplace. Relevant work explicitly indicated that collecting data from large population would improve the ability to generalize from these findings. The current study covered substan‐ tial number of craft workers to significantly represent the small-scale jewelry making industry. In addition, most reported studies included only pain-free and physically fit subjects and so these findings may not be applicable when compared with findings pertaining to Philip‐ pine jewelry craft workers in their natural workplace settings. The work of P. Waongenn‐ garm, et al. (2015) stated that the results may differ between symptomatic and asympto‐ matic individuals [4]. According to Dunk and Callaghan (2010), people without low-back pain have higher level of perception to discomfort than those already experiencing low-back pains [5]. The current study involved symptomatic and asymptomatic craft workers in order to investigate and compare the response of both groups. The majority of the respondents on previous studies were almost stationary while being observed. This was true with Vergara and Page’s (2002) study, which investigated the relationship of comfort and back posture and sitting posture mobility [6]. Similarly in the experiment of H. Bae, et al. (2015), the participants performed simple task types such as memorizing words, playing tetris, reading articles and relaxing [7]. In the current study, the participants performed actual jewelry making tasks.

A Task Analysis of Small-Scale Jewelry Craft Workers

2

217

Development of Hypothesis and Conceptual Model

The study intended to determine the interrelationship of level of body discomfort, sitting posture mobility and various ergonomics risk factors present on small-scale jewelry making work environment (Fig. 1).

Work Design Components

Sitting Posture

Environment Factors

Sitting Mobility

Individual Factors

Level of Body Discomfort

Fig. 1. A conceptual framework of the relationship between work system elements, sitting posture, mobility and body discomfort.

3

Methodology

3.1 Measurement and Instrumentation of Variables Initially, a Revised Oswestry Disability Index (ODI) Questionnaire (O’Brien, 1976) was used to categorize the participants into symptomatic and asymptomatic individuals. An ODI score of at least 21% indicated that the craft worker was under symptomatic individuals or those who experienced more pain however there activities were not completely affected. The Body Discomfort Chart by Corlett and Bishop (1976) was utilized to evaluate the level of body discomfort among jewelry craft workers [8]. In this survey method, craft workers rated the degree of body discomfort for different body segments using a scale from 0 (I do not experience discomfort) to 10 (I experience extremely strong discomfort). The scale-based approach was based on Borg’s (1990) category ratio scale (CR-10). The craft workers were to be surveyed before and after the observation to check the level of body discomfort as well as locate the potential musculoskeletal problems. A video-based observation method was chosen to record the frequency of sitting posture mobility while performing jewelry making tasks. At the same time, the video record aided in determining the type of sitting posture for every postural change. The motion-capturing device was positioned laterally and perpendicularly to the craft work‐ er’s position. During analysis of the video records, when a craft worker maintained a sitting posture for at least 30 s, it was considered to be one postural change [7].

218

H. Caparas and A. Matias

In terms of observation duration, the current study adopted the work of N. Owen, et al. (2015) which focused on the sedentary behavior [9]. Based on the study, it was recommended to limit discretionary sitting time to a maximum of 2 h per day, and to stand up and move after 30 min of uninterrupted sitting as a break in sitting time. Sitting postures for every postural change were analyzed using the Rapid Upper Limb Assessment (RULA) methodology [10]. The body segments included in the postural analysis was upper arm, lower arm, wrist, neck and trunk. In the current study, trunk movement signified global change in sitting position. This indicator may exhibit trunk flexion, extension and lateral bending. Data concerning individual factors were collected by means of a survey method and direct body specifics measurement. A survey questionnaire was designed so as to gather information such as age, length of work experience of craft workers and to classify them as right or left handed. Anthropometric devices such as weighing scale and flexible tape were utilized to measure the workers weight and stature, respectively. Work Design Components mainly consisted of the anthropometric mismatch between the craft worker’s body dimensions and the physical elements of their work setting and equipment such as tool grip, seat design and work surface design. Different design parameters were be measured using caliper, flexible tape and clinometer protractor. Environmental factors were measured using a Luxmeter for illumination level, a Temperature and Humidity Meter for thermal condition and humidity (Fig. 2).

Fig. 2. Methodological framework.

A Task Analysis of Small-Scale Jewelry Craft Workers

219

3.2 Sampling As of 2016, there were approximately 103 Platero and Stone Setters from registered jewelry repair shops in Meycauayan, Bulacan (Source: Business Permit and Licensing Office of Meycauayan, Bulacan). Using the Slovin’s formula (Slovin, 1960) to calculate the sample size using ±0.05 margin of error, it was estimated that 82 jewelry craft workers were required in the observation. Nevertheless, the study covered 91 jewelry makers in Meycauayan, Bulacan thus giving a margin of error of ±0.0358. Data were collected in Meycauayan, Bulacan given that the nation’s small-scale jewelry industry was largely concentrated in this area. 3.3 Statistical Analysis The analysis plan used different statistical analyses depending on the suitability to the variables considered in the test. The effect on level of body discomfort of various factors such as occurrence of sitting mobility, sitting postures, work design components, envi‐ ronmental factors and working person’s characteristics were investigated using Logistic Regression. Similarly, analysis of the level of body discomfort utilized a Logistic Regression to predict the likelihood of having a certain degree of variation. Accordingly, this was linked to the exposure of the individual to ergonomic risk associated with upper extremity MSD due to the influence of work design components, environment factors and individual factors. Moreover, the current study intended to determine the frequency of sitting mobility with respect to various factors in the work system of jewelry making. There were several variables presumed to be needed to predict the frequency of sitting mobility and the time interval between macro-repositioning movements. Thus, to analyze the relationship of various predictor variables and sitting mobility, a multiple linear regression model was developed. A Stepwise Regression modelling approach simplified establishing the final regres‐ sion model by progressively removing the insignificant variables. The process aided in finding a model reduced to the most important variables that most easily predict the response variable. In quantifying the degree of relationship between two variables, appropriate measures of association were employed. The study made use of the Pearson correlation coefficient for continuous data and Spearman rank correlation coefficient for two ordinal variables, whichever was applicable. Both tests provide coefficients, which hold similar interpretation concerning the strength of relationship between two variables. A Chi-square Goodness-of-fit test was used to determine whether the statistical model adequately described the data. A large p-value is desirable and it means the model fits the data under investigation. Moreover, a test of hypothesis was used to investigate the significant difference of Asymptomatic and Symptomatic groups of jewelry makers. Particularly, a two-sample t-test and confidence interval were conducted to establish whether the responses of both groups were statistically the same or not.

220

4

H. Caparas and A. Matias

Results and Discussion

The participants of the study were adult males, with an average age of 44 years. The mean weight and stature of these jewelry craft workers was 62.98 kg (SD 12.79 kg) and 157.22 cm (16.27 cm), respectively. They were in the craft of jewelry making for an average of 18 years. A total of 81% of the participants self-reported to be right-handed. 4.1 Comparative Analysis of Symptomatic and Asymptomatic Subjects The work sitting postures maintained by Asymptomatic and Symptomatic jewelry makers were significantly the same (P-Value = 0.909). Both tasks involved high degree of meticulousness in order to achieve various intricate designs of fine jewelry. In view of that, we concluded that work sitting postures maintained by craft workers were prob‐ ably affected by the requirements of their job activities. The outcome could be considered as a positive feedback loop to working posture of an individual based on cybernetic model by Marmaras, et al. (2008) [11]. Likewise, Dowell, et al. (2001) verified that workers unconsciously and consciously lean forward when performing intensive tasks [7]. The frequency of postural changes of Asymptomatic and Symptomatic jewelry makers were significantly different (P-Value = 0.000). Also, Asymptomatic subjects have been recorded to exhibit less movements than Symptomatic subjects. The results verified the findings of K. O’Sullivan, et al. (2012) indicating that subjects experiencing low back pain have greater frequency of postural changes [12]. Also, the study of F. Maradei, et al. (2015) which focused on driving tasks validated the findings of the current study [13]. Consis‐ tently, our results indicated that there was higher number of occurrences of postural changes for subjects with low back pain. The level of overall body discomfort perceived by Asymptomatic jewelry makers was significantly different from that perceived by Symptomatic jewelry makers (PValue = 0.000). Correspondingly, Asymptomatic subjects perceived less overall body discomfort compared to Symptomatic subjects (P-Value = 0.000). These results validated the classification method used in categorizing jewelry craft workers into Asymptomatic and Symptomatic groups. The outcome of body discomfort assessment was consistent with the response of the craft workers to Revised Oswestry Disability survey. Pearson’s r (0.9115) and Spearman’s rho (0.9065) values suggested a strong positive correlation between the Revised Oswestry Index and the Rating of General Body Discomfort. The coefficients indi‐ cated that as the rating for perceived overall body discomfort increases, there was a higher chance of being in the symptomatic category. 4.2 Work Design Component Mismatch, Environmental Factors and Individual Factors as Explanatory Variables to Work Sitting Posture Given that work sitting posture scores provided binary data as response variables (WSP = 2 means deviated from neutral position; WSP = 3 means highly deviated from neutral position), the study used Binary Logistics Regression to determine the relation‐ ship of various work system elements to the work sitting postures maintained by jewelry

A Task Analysis of Small-Scale Jewelry Craft Workers

221

makers. The results indicated that at 90% confidence level, mismatch on working height (WDC_M5: p-value = 0.001) and thermal conditions (ENVF_T: p-value = 0.062; ENVF_H: p-value = 0.023) have significant effects on work sitting posture. Moreover, the result showed that a one unit increase in working height mismatch resulted in 22% increase in the odds that the craft worker deviated from a neutral position (OR = 1.22) with other variables held constant; a one unit increase in temperature level resulted in a 62% increase in the odds that the craft worker deviated from a neutral position (OR = 1.62) with other variables held constant; and a one unit increase in room relative humidity resulted in 15% increase in the odds that the craft worker deviated from neutral position (OR = 1.15) with other variables held constant. The statistic G = 37.626 with a p-value of 0.000 indicated that there was sufficient evidence to conclude that at least one of the estimated coefficients was different from zero. Also, the goodness-of-fit tests with p-values ranging from 0.016 to 0.355 designates that there was insufficient evidence to claim that the model did not fit the data adequately. Using the identified significant factors for work sitting posture, an experiment was conducted to further investigate the main and interaction effect. The design of the experi‐ ment used a high and low range of values for the factors of interest. The result of this 2-level Factorial Design for proportion of work sitting posture score of 3 showed that temperature level (P-value = 0.000) and working height mismatch (P-value = 0.037) have significant effects on the proportion of workers exhibiting a work sitting posture score of 3. Addition‐ ally, there was no interaction effect between these variables. 4.3 Work Design Component Mismatch, Environmental Factors, Individual Factors and Work Sitting Posture as Explanatory Variables to Work Sitting Mobility A General Linear Regression was used to determine the relationship between various work system elements and work sitting posture and frequency of work sitting mobility and postural changes exhibited by jewelry makers. The results showed that at the 95% confidence level, a mismatch on working height (WDC_M5: p-value = 0.000), tool grip design mismatch (WDC_M7: p-value = 0.015), individual factor particularly righthandedness (INDF_H: p-value = 0.050), subject classification (ODI: p-value = 0.000) and work sitting posture (WSP: p-value = 0.000) had significant effects on the frequency of sitting mobility of jewelry makers. The results of regression analysis showed that 78.92% of the total variability in the frequency of sitting mobility was explained by identified predictor variables. Using a Chi-Square Goodness-of-fit Test for frequency of work sitting mobility, the p-value (1.000) is greater than a given alpha-level of 0.05. Therefore, we do not reject the null hypothesis indicating that the model adequately described the data. Reduced Regression Models were used to predict the frequency of sitting mobility on a given work sitting posture, subject classification and predetermined workplace design mismatch. The result showed that sitting mobility was more frequent for settings having working height mismatch of 45 cm. The results showed that as the mismatch on working height increased, the frequency of sitting mobility also increased. Based on the result, sitting mobility was seen to be a function of changes in work sitting posture and

222

H. Caparas and A. Matias

mismatch between workers anthropometric measurements and working height designs. Mismatch on working height influenced the occurrence of sitting mobility. These factors resulted in frequent postural changes which were seen to be a mode of reducing the body strain and way of adjustment for a more relaxed working position. Another reasonable cause was the craft worker’s reflex to meet the visual requirements of the task while being exposed to injuries in a workplace with incompatible design. 4.4 Work Design Component Mismatch, Environmental Factors, Individual Factors, Work Sitting Posture and Mobility as Explanatory Variables to General Body Discomfort The effects of various work system elements, work sitting posture and frequency of work sitting mobility to overall body discomfort was analyzed using an Ordinal Logistic Regression. The result revealed that at a 95% confidence level, mismatch on working height (WDC_M5: p-value = 0.033), work sitting posture (WSP: p-value = 0.006), frequency of sitting mobility (WSM_F: p-value = 0.000) and sitting duration (WSM_D: p-value = 0.008) have significant effect on the level of perceived body discomfort of jewelry makers. Moreover, the odds of experiencing low body discomfort in a workplace with greater working height design mismatch was 15% less likely than in a workplace with small working height design mismatch, (OR = 0.85) with other variables held constant. The odds of experiencing low body discomfort for workers in a highly contorted and awkward sitting posture was less likely than for workers who were working with a sitting posture close to neutral position (OR = 0.0042) (with other vari‐ ables held constant). The odds ratio for the work sitting posture indicated a very strong degree of association with general body discomfort level. The odds of experiencing low body discomfort given a prolonged work sitting duration was 49.59% less likely than when working for a shorter sitting duration (OR = 0.50), (with other variables held constant). Lastly, the odds of experiencing low body discomfort by workers with high frequency of sitting mobility was less likely than for workers with low frequency of sitting mobility (OR = 0.02), (with other variables held constant). The odds ratio for the sitting mobility indicated a very strong degree of association with general body discom‐ fort level. The statistic G = 176.837 with a p-value of 0.000 indicated that there was sufficient evidence to conclude that at least one of the estimated coefficients was different from zero. Also, the goodness-of-fit tests with p-values ranging from 0.000 to 1.000 indicated that there was a mixed result in relation to adequacy of the model. Cox and Snell R2 indicated that 85.7% of variation in general body discomfort was explained by the logistic model. Similarly, Nagelkerke R2 showed that 93.10% of the variation on general body discomfort was due to identified predictor variables. ] [ P (GBD ≤ 1|WSP = 2) = 10.954 + 0.162WCD_M5 + 0.685WSM_D + 4.158WSM_F − 5.473WSP logit P (GBD > 1|WSP = 2) [ logit

] P (GBD ≤ 1|WSP = 3) = 10.954 + 0.162WCD_M5 + 0.685WSM_D + 4.158WSM_F P (GBD > 1|WSP = 3)

(1) (2)

[ logit

[ logit

A Task Analysis of Small-Scale Jewelry Craft Workers

223

] P (GBD ≤ 2|WSP = 2) = 21.607 + 0.162WCD_M5 + 0.685WSM_D + 4.158WSM_F − 5.473WSP P (GBD > 2|WSP = 2)

(3)

] [ P (GBD ≤ 2|WSP = 3) = 21.607 + 0.162WCD_M5 + 0.685WSM_D + 4.158WSM_F logit P (GBD > 2|WSP = 3)

(4)

] P (GBD ≤ |3WSP = 2) = 31.586 + 0.162WCD_M5 + 0.685WSM_D + 4.158WSM_F − 5.473WSP P (GBD > 3|WSP = 2)

(5)

] [ P (GBD ≤ |3WSP = 2) = 31.586 + 0.162WCD_M5 + 0.685WSM_D + 4.158WSM_F logit P (GBD > 3|WSP = 2)

(6)

An experiment was conducted to further investigate the main and interaction effect of significant factors impacting general body discomfort. The result of 2-level Factorial Design for proportion of workers having general body discomfort rating of at least 2 showed that working height mismatch (P-value = 0.000), sitting duration (Pvalue = 0.003) and frequency of sitting mobility (P-value = 0.000) have significant effect on the proportion of workers who experienced a general body discomfort level of at least 2. Additionally, there was an interaction effect for work sitting posture and working height mismatch (P-value = 0.000), work sitting posture and sitting duration (Pvalue = 0.009), working height mismatch and frequency of sitting mobility (Pvalue = 0.000), sitting duration and frequency of sitting mobility (P-value = 0.000), sitting posture, working height mismatch and sitting duration (P-value = 0.001) and interaction of four factors (P-value = 0.000) (Fig. 3).

Fig. 3. Predicted proportion of workers with body discomfort score of at least 2

4.5 General Body Discomfort as Explanatory Variable of Work Sitting Mobility Conversely, the effect on work sitting mobility by various work system elements, work sitting posture and overall rating of body discomfort was examined using a General Linear Regression Model. The result specified that at 95% confidence, general body

224

H. Caparas and A. Matias

discomfort (GBD: p-value = 0.000) had a significant effect on the frequency of sitting mobility of jewelry makers. The result was used to predict the frequency of sitting mobility on a given level of general body discomfort. It showed that the frequency of sitting mobility was exponentially increasing as the level of general body discomfort increased.

5

Conclusion

Seeing the capability of contributing to economic development and providing social bene‐ fits through creating job opportunities, the study primarily focused on exploring the work systems supporting jewelry making. Specifically, it tried to investigate the causes of body discomfort, work sitting posture and mobility by analyzing different risk factors present in the work systems of jewelry making. Craft workers in small-scale jewelry repair shops were included in the study, which considered how handmade designs, completed by means of traditional ways, could make the fine jewelry more appealing and attractive to customers. The premise of the study involved several factors such as individual factors, workplace design components and environmental factors. Moreover, the study involved a comparative analysis of the response of Symptomatic and Asymptomatic customers. Thus, the task anal‐ ysis arises to the following summary of all findings: • The work sitting postures maintained by Asymptomatic and Symptomatic jewelry makers were significantly the same (P-Value = 0.909). • The frequency of postural changes of Asymptomatic and Symptomatic jewelry makers were significantly different (P-Value = 0.000). Also, Asymptomatic subjects have been recorded and were seen to exhibit less movement than Symptomatic subjects. • The level of overall body discomfort perceived by Asymptomatic and Symptomatic jewelry makers were significantly different (P-Value = 0.000). Correspondingly, Asymptomatic subjects perceived less overall body discomfort compared to Symp‐ tomatic subjects. • Working height design mismatch and thermal conditions of the workplace signifi‐ cantly influenced the work-sitting posture of jewelry craft workers. • Working height design mismatch, tool grip design mismatch, right-handedness, subject classification and work sitting posture significantly influence the frequency of work sitting mobility of jewelry craft workers. • Working height design mismatch, work sitting posture, working duration and frequency of work sitting mobility significantly affect the level of perceived body discomfort of jewelry craft workers. • In the same way, the level of body discomfort has significant effect on the frequency of sitting mobility. The results revealed several factors affecting work sitting posture, sitting mobility and perceived body discomfort of jewelry craft workers. An ergonomics intervention could minimize the risk associated with contorted and awkward work sitting posture as influenced by existing working height design specifics. Anthropometric intervention

A Task Analysis of Small-Scale Jewelry Craft Workers

225

based on the craft workers’ body segment measurements aided in determining the recommended dimensions of working height and sitting height. The proposed heights of these workplace components were set at 95th percentile of anthropometric measure‐ ments (Recommended Working Height = 106 cm; Recommended Work Sitting Height = 55 cm). Also, slanting the workbench by 15° from the horizontal surface from a reference point of 95 cm from the floor intuitively satisfies the anthropometric meas‐ urements of craft workers and may help to bring the work piece nearer and more visible to the optimum light of sight of the workers [11]. For thermal conditions, it is essential to adjust the ventilation system of the workplace properly. Since the average number of workers per repair shop is 2 to 3 craft workers, local ventilation is more economical than general ventilation for the whole facility. In order to avoid static working posture, we recommend that craft workers have alternate sitting and standing positions. Based on a study by N. Owen, et al. (2011), 2 h per day is the maximum time for voluntary sitting and the workers have to stand up and move after 30 min of constant sitting position [7]. Additionally, it is recommended that management promote active sitting among jewelry craft workers. According to the study, this method increases the trunk motion thus giving favorable effects to the lowback area [12]. In the long run, these alternatives may prevent the occurrence of occu‐ pational disorders and diseases. Alternatively, the researcher suggests conducting a replication of the current study with the same subject population using objective methods in measuring body discomfort of different body regions. Application of the same methodology to craft workers in large and medium scale-scale jewelry enterprises is also recommended for future study. It is important to determine who, among the craft workers from large, medium and smallscale businesses and industries, are highly vulnerable to the risk of work-related diseases based on the result of the task analysis.

References 1. Florencio, M.: The Conference of Philippine Jewelers, Inc. (2014) 2. Mantro-Beltran, L.: The Business of Jewelry in the Philippines. The Manila Times (2015) 3. Remo, A.: PH Jewelry Companies Call for Government Support. Philippine Daily Inquirer (2014) 4. Waongenngarm, P., Rajaratnam, B., Janwantanakul, P.: Perceived body discomfort and trunk muscle activity in three prolonged sitting posture. J. Phys. Ther. Sci. 27, 2183–2187 (2015) 5. Dunk, N.M., Callaghan, J.P.: Lumbar spine movement patterns during prolonged sitting differentiate low back pain developers from matched asymptomatic controls. PubMed 35, 3– 14 (2010) 6. Vergara, M., Page, A.: Relationship between comfort and back posture and mobility in sittingposture. Appl. Ergon. 33, 1–8 (2002) 7. Bae, H., Kim, H., Suk, H.: Sitting posture-based lighting system to enhance the desired mood. J. Ergon. Soc. Korea 34, 191–198 (2015) 8. McAtamney, L., Corlett, E.N.: RULA a survey method for the investigation of work related upper limb disorders. Appl. Ergon. 24, 91–99 (1993) 9. Corlett, E.N., Bishop, R.P.: A technique for assessing postural discomfort. Pub Med 19, 175– 182 (1976)

226

H. Caparas and A. Matias

10. Owen, N., Takemi, S., Eakin, E.: Adults sedentary behavior: determinants and interventions. Am. J. Prev. Med. 41, 189–196 (2011) 11. Marmaras, N., Nathanael, D.: Workplace design. In: Gavriel, S. (ed.) Handbook of Human Factors and Ergonomics, 4th edn. Wiley, New Jersey (2012). Chapter 21 12. O’Sullivan, K., O’Keeffe, M., O’Sullivan, L., O’Sullivan, P., Dankaerts, W.: The effect of dynamic sitting on the prevention and management of low back pain and low back discomfort: a systematic review, pp. 37–41. Taylor & Francis Informa Ltd Registered in England and Wales (2012). Registered Number: 1072954 Registered office: Mortimer House 13. Maradei, F., Quintana, L., Castellanos, J.: Macro repositioning movements of pelvis and trunk during simulated driving tasks. Int. J. Ind. Ergon. 46, 19–28 (2015)

The Consequences of Lean Six Sigma on Banking Improvement: A Study at a Front-Line Unit of a Bank Company in Indonesia Nilda Tri Putri1(&), Andi Gunawan1, and Agus Sutanto2 1

Faculty of Engineering, Department of Industrial Engineering, Andalas University, Padang, Indonesia [email protected], [email protected] 2 Department of Mechanical Engineering, Andalas University, Padang, Indonesia [email protected] Abstract. The purpose of this paper is to present research about how lean Six Sigma delivers recommendations to minimize the seven wastes which is strongly associated with human aspects in banking sector. The research was performed using Define, Measure, Analyze, Improve (DMAI) method, whereas the seven wastes are identified and will be minimized by using different tools. A research questionnaire delivered the sigma level of the defects in services that performed by teller, customer service, and security person are 3.01, 2.91, and 3.89 respectively. Lists of actual activities of delay/waiting and unnecessary movements is delivered by using Closed Circuit Television (CCTV) records and Pro Time Estimation software. The wastes of duplication and over processing, unclear communication, incorrect inventory, and lost opportunity are found using brainstorming. There are six recommendations of improvement obtained from brainstorming within the 5-Whys, FMEA, and improve phases. The bank can achieve significant efficiency without extorting employees’ energy with low cost operations and flexibility. Keywords: Banking sector
Brainstorming
Lean six sigma estimation SERVQUAL
Sigma level
Waste




Pro time

1 Introduction 1.1

Lean

The term “Lean thinking” was first proposed by Womack, et al. (1990) [1] and it encompasses a set of lean practices. The lean philosophy were developed by Toyota which is known as Toyota Production System in the mid 50 s and is nowadays worldwide used by companies. Lean Thinking book by Womack and Jones (1996) [2] commented: “Lean thinking can be summarized in five principles: precisely specify value by specific product (value specification), identify the value stream for each product (value stream mapping), make value flow without interruptions (flow optimization), let the customer pull value from the producer, and pursue perfection or continuous © Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2_21

228

N.T. Putri et al.

improvement”. The fundamental principle behind lean is the elimination of waste from a process to promote using the least amount of capital or other investments to produce the most amount of gain. Lean is focusing on eliminating “muda” which is a Japanese word for waste. Lean focused on only seven types of waste. There are seven wastes in the manufacturing sector such as defective units, inappropriate processing, unnecessary motions, excessive transportation, over production, waiting, and excessive inventory [3]. Earlier applications of lean concept only focused on manufacturing companies. There is a movement to apply these very same concepts to the service industry in places like hospitals, office buildings, and even banking sector. The manufacturing industry has some differences with service industry, so there are some revisions of the seven wastes concept for service industry. The seven wastes of the service industry are [4]: Delay/Waiting; Duplication and over processing; Unnecessary movement; Unclear communication; Incorrect inventory; Lost opportunity; and Defects in service. 1.2

Six Sigma

Introduction of the Six Sigma concept for the manufacturing industry on 1980 by Motorola brought about a revolution of the scope and use of quality systems in this era. The goal of Six Sigma is to “reduce variation in organization processes by using improvement specialists, a structured method, and performance metric with the aim of achieving strategic objectives [5]. As a Metric, when a process is operating at Six Sigma level, it will produce nonconformance (i.e. defects or errors) at a rate of not more than 3.4 defects per one million opportunities. As a methodology and quality management philosophy, Six Sigma leads to improve business process by focusing on managing customer requirements. It is also a statistical term used to measure process variations, i.e., how far a given process deviates from perfection, which causes defects. There are two methods to improve the six sigma target, such as: (1) Six sigma DMAIC (Define, Measure, Analyze, Improve, Control) dan (2) Design For Six sigma DMADV (Define, Measure, Analyze, Design, Verify). DMAIC is used to improve current business process, whereas DMADV is used to design the process of new product to deliver zero defects performance [6]. 1.3

Lean Six Sigma

Lean six sigma is an application which combined methodology of lean in six sigma methodology. In other words, techniques in lean integrated within DMAIC or DMADV methodology [7]. There are a lot of scientific publications which discussed the application of lean six sigma in manufacturing sector. Bubshait, et al. (2014) [8], Hassan (2013) [9], Enoch (2013) [10], Mandahawi, et al. (2012) [11], and Panat, et al. (2014) [12] concluded that the lean six sigma as a powerful tool to improve the efficiency and effectiveness of the manufacturing system. Scientific publications which discussed the applications of lean six sigma in services sector is also growing but with different subconcepts. Service sector is very different from manufacture sector, where the

The Consequences of Lean Six Sigma on Banking Improvement

229

definition of “defects” is very dynamic and the way to measure the “defects” is abstract. One of the types of services that is complex to implementing the lean six sigma method is banking or financial service. The benefits of implementing lean Six Sigma in Banking or financial service are highly dependent on the human aspect for example in reducing internal and external customer complaints, increasing customer satisfaction, improving process efficiency, and reducing cycle time. This paper presents a case study of Lean Six Sigma application at a front-line unit of a bank company in Indonesia.

2 Method The purpose of this paper is to deliver the improvement recommendations for banking sector by using Lean Six Sigma. A preliminary questionnaire based on SERVQUAL model was used to identify gap between customer perception and customer expectations. The dimensions are tangibles, responsiveness, reliability, assurance, and empathy. The dimension was explored to the respondent by using 21 attributes. The respondent was asked to give the rating scale between “Strongly disagree” until “Strongly agree” of service dimension for customer perceptions and customer expectations. The preliminary study was done by distributing the questionnaire into 30 respondents. The result of rating scale is tested by validity, reliability, and significance test. An interview with branch vice manager was done to strengthen the gap that was identified. Additionally, a research questionnaire was designed to explore the response of the customers for the services. The purpose of this questionnaire was to measure DPMO and the sigma level of Critical to Quality (CTQ) of the research objects. There were total 16 CTQ’s across the three services i.e. bank teller, bank security, and bank customer service. The respondent was asked to give the rating scale between “very bad” until “very good” of the current services of the company. The questionnaire used a 5 level Likert scale. Scale 5 and 4 are expressions of the desired value, so that the Lower Specification Limit (LSL) is the scale 4. Scale 1, 2, and 3 are expressed as a defects. The sample size was set to 30 respondents as the baseline. If none of the 30 sampling are found the defects value, then the sampling should be continued until at least 1 defects found. This questionnaire system describe by the tabulation sheet as shown in Table 1. The result scale also tested by validity and reliability test. Table 1. The tabulation sheet of customer’s perception of teller’s CTQ Respondents

R1 R2 … Total of defects

Stand up and raise right hand when mentioning the queue number 4 3 …

Say greetings and asking the customers’ needs

Serve the needs of customers well

Asks again whether there could be assisted

Say thanks

Defect or Not

5 5 …

4 4 …

4 4 …

5 4 …

√ … …

230

N.T. Putri et al.

Other instruments are Pro Time Estimation software and CCTV records that used to identify waste by type of delay/waiting and unnecessary movement in bank unit. The actual activity of waste in the company can be detected easier by using CCTV records. CCTV records used two days with seven hours of service each day. There are four research objects (respondents) i.e. functional teller, two customer services and security. It is important that the observations can describe the actual conditions and illustrated at least 90% of the population then validation was done by doing brainstorming section with two experienced managers representing the bank. Brainstorming with the branch vice manager is also done when practice 5-Whys, Failure Mode and Effect Analysis (FMEA), and formulating the recommendations for improvement. The entire instruments are blended in a DMAI methodology.

3 Results Table 2 shown results of testing the questionnaire which included validity, reliability and significance test. The test showed the gap of each quality attributes are significant at 96.6% significance level. The bank can know that the customer is still not satisfied in every dimension of services. Quality gap of responsiveness dimension is the lowest with a value of −0.78 and gap of empathy dimension also has a very low value that is −0.63. Based on the results, it can be concluded that the depth investigation was needed to carry out in terms of non value added activities or defects (wastes) occured in bank front-line unit. The investigation of defects in services conducted through research questionnaire which design to measure Defects per Million Opportunity (DPMO) and sigma level. DPMO and sigma level is used to describe performance baseline of Critical to Quality (CTQ). DPMO and sigma level of teller services obtain by calculating using the equation below: 1; 000; 000  number of defects number of units  number of opportunities per unit 1; 000; 000  10 ¼ ¼ 66; 666:67 30 sample  5 CTQ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rlevel ¼ 0:8406 þ 29:37  2:221  ln ðdpmoÞ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ 0:8406 þ 29:37  2:221  ln ð66; 666:67Þ ¼ 3:01 DPMO ¼

ð1Þ

ð2Þ

Based on customer perceptions, the bank can find out how much the performance baseline of their employees with values of 66,666.67 Defects per Million Opportunity (DPMO) or sigma level of 3.01 to a teller’s services. It should be noted that the value of DPMO and sigma level are presented in accordance with the value issued by Motorola which has shifted 1.5 sigma. Based on the same way, the bank knew the position of customer service performance at 80,952.38 dpmo or 2.91 sigma and security of 8,333.33 dpmo or 3.89 sigma.

The Consequences of Lean Six Sigma on Banking Improvement

231

Table 2. The gap between customer perceptions and customer expectations on service dimension of the bank’s front-line unit Dimension

Attribute Mean of Mean of Attribute perception expectation gap

Responsiveness X10 Xll X12 X13 Tangibles X1 X2 X3 X4 Reliability X5 X6 X7 X8 X9 Empathy X17 X18 X19 X20 X21 Assurance X14 X15 X16

4,2 3,93 4,3 7 4,10 4,20 3,87 4,50 4,00 4,03 4,03 4,47 3,93 4,60 4,20 4,17 4,30 4,33 4,40 4,53 4,73 4,50

4,90 4,93 4,97 4,93 4,93 4,87 4,90 4,90 4,90 4,93 4.97 4,97 4,97 4,83 4,90 4,93 4,90 4,97 4,97 4,93 4,93

−0,70 −1,00 −0,60 −0,83 −0,73 −1,00 −0,40 −0,90 −0,87 −0,90 −0,50 −1,03 −0,37 −0,63 −0,73 −0,63 −0,57 −0,57 −0,43 −0,20 −0,43

RUz mean hypothesis test

Significance ða ¼ 3; 4%Þ

Dimension gap

−3,80 −6,05 −4,43 −4,77 −5,30 −4,85 −3,38 −4,96 −5,76 −4,63 −2,87 −6,38 −2,87 −2,92 −4,98 −4,38 −3,62 −3,40 −3,62 −2,12 −3,07

Significant

−0,78

Significant

−0,76

Significant

−0,73

Significant

−0,63

Significant

−0,36

The bank knew that the customers is still not satisfied in every dimension of services, and knew the amount of defects in their services, the bank also can find out how much the waste existed manifold by delay/waiting and unnecessary movement confiscated employee’s work hours. The investigation was obtained after observing the CCTV records that assisted by Pro Time Estimation software. The results obtained shows that there are two types of activities which caused teller’s job delayed or should waiting. Delay/waiting confiscated 1.05% of teller working hours. Waste type of unnecessary movement divided into five types of activities which confiscated 3.51% of teller working hours. Ten types of activities as shown in Table 3 confiscated 11.51% of customer service 1 working hours. Twelve types of activities confiscated 15.79% of customer service 2 working hours plus 9.41% of customer service 2 works done in locations that are not supposed to. Seventeen types of activities as shown in Table 4 confiscated 35.66% of security working hours.

232

N.T. Putri et al. Table 3. Observations result on unnecessary movements of customer services 1

No. Unnecessary movement

NVA/SVA Frequency Average time (s)

1 Chatting with other employee NVA Total time of NVA 2 Pick up the documents from SVA storage or from account officer 3 Move to discuss the customer documents with customer service 2 4 Move to discuss the customer documents with teller 5 Move towards printer 6 Move towards account officer’s room 7 Pick up the documents from teller desk 8 Deliver and pick up the documents from storage or from account officer 9 Deliver the documents from storage or from account officer 10 Move to ask something to teller Total time of SVA Individual needs Total

1

94,82

4

282,23

Total Percentage time (s) to work hours 94,82 0,22% 94,82 0,22% 1128,9 2,61%

1

10,69

10,69

0,02%

3

285,28

855,83

1,98%

5 2

26,02 362,16

130,09 724,31

0,30% 1,68%

2

52,43

104,86

0,24%

5

206,39

1031,96 2,39%

1

190,42

190,42

0,44%

2

21,81

43,62

0,10%

4220,68 9,77% 658,56 1,52% 4974,06 11,51%

Another four types of waste is difficult to identify. So, an alternative is identified by brainstorming and measured directly with a risk priority number. Two experienced managements who participated in a brainstorming has stated waste manifold by duplication/over processing existed were “waste of paper on the account registration form and not have a tax ID number letter” and “twice queuing of customers who open new accounts”. Waste of unclear communication types existed as “interruptions from other customers” and “cancel the transaction because lack of transaction requirement”. Waste of incorrect inventory type existed as “the files piled up on the service table when customer service serving the customers” and “storing documents should be made less”. Waste of lost opportunity existed as “customer lamented, even cancel the transaction because queuing in teller took a long time”, “more than one customer in a service by customer service” and “uncertainty for customers who wait for customer service”.

The Consequences of Lean Six Sigma on Banking Improvement

233

Table 4. Observations result on unnecessary movements of security No. Unnecessary movement 1 2

NVA/SVA Frequency Average time (s)

Chatting with customer service 1 NVA Chatting with customer service 1 and 2 3 Chatting with customer service 2 4 Chatting with customer service 2 and teller 5 Chatting with customer in customer service 2 desk 6 Chatting with customer in teller desk 7 Chatting with other employee in teller desk 8 Chatting with customer who queuing 9 Chatting with teller 10 Chatting with other three employees 11 Sitting on waiting chairs 12 Sitting on customer service 1 chair 13 Sitting on customer service 2 chair 14 Out from office Total time of NVA 15 Individual needs SVA 16 Perform customer service 1 activities 17 Fetching other employee’s documents Total time of SVA Total

Total time (s)

Percentage to work hours

6 1

204,30 146,77

1225,77 146,77

2,84% 0,34%

6 1

60,93 278,02

365,55 278,02

0,85% 0,64%

3

108,87

326,60

0,76%

1

192,56

192,56

0,45%

1

72,97

72,97

0,17%

1

67,11

67,11

0,16%

1 1

90,71 245,50

90,71 245,50

0,21% 0,57%

2 2

86,51 1480,94

173,01 2961,88

0,40% 6,86%

2

668,03

1336,06

3,09%

4

528,83

18 1

309,73 104,45

2115,32 9597,83 5575,16 104,45

4,90% 22,22% 12,91% 0,24%

1

127,49

127,49

0,30%

5807,10 13,44% 15404,93 35,66%

An analysis was done by approaching 5-Whys and FMEA. Companies can find out the root of the problem (failure mode) that existed in the front-line unit and how much each failure mode should get attention as shown in Table 5. As the most important result, there are six improvement recommendations for upgrade the bank’s unit efficiency. Recommendations were obtained from brainstorming in improve phase are: 3.1

Using Smart IP Camera to Evaluate the Employees Performance

Use of Smart IP camera to evaluate the performance of employees is the proposed solution to overcome the failure mode “Inadequate tools used in evaluating defects of

234

N.T. Putri et al. Table 5. Results of analyze phase

Failure mode obtain from approaching 5 why Inadequate tools used in evaluating defects of services Space and content of the registration form is not optimal Still using the numbering queueing system Customer Service is also doing other work while serving customers The layout is not optimal Customer Services desk capacity is not optimal Security stand too long and doing repetitive work Services at customer service desk does not use a queue number but do not have queue line Do not use a queuing system which customer’s lined up Lack of notice to prepare the document before customers arrived in teller Customers who open new accounts have to sit down again when the file is in the process from customer services to teller The lack of visual aid

RPN obtain from approaching FMEA 486 432 405 315 245 243 216 216 216 144 144 144

services” and “customer service is also doing other work while serving customers”. Recommended Smart IP camera is capable of storing CCTV records nonstop for several days, recording sound and should still give a good and futuristic impression if used at the office. CCTV can be installed respectively near teller, customer service and security. CCTV records can be used as an employee performance evaluation by the management of the bank unit. The next CTQ performance baseline calculations quite done by observing the CCTV records, so that it will becoming an alternatives besides the questionnaire explained. The employees evaluation results can be briefed periodically in the morning or evening. Smart IP camera is the right choice to be used because the cost is around Rp. 1,500,000 or USD 111.23 for the four cameras as very cheap compared to the benefit that is much higher. The benefit is a vision to achieve defect-free services (the good image of the eyes of ±75,000 customers/3 years) through services sigma periodically evaluation using CCTV records. 3.2

Reducing Paper of New Account Registration for Customer Who Doesn’t Have Tax ID Number

Failure modes “space and content of the registration form is not optimal” can be overcome by reducing some space in the account registration form and insert the certificate of lack of tax ID number contents. This recommendation can provide enormous benefit because it can reduce the cost of making the form from two sheets into one sheet of ±Rp. 135,000/year or USD 10.01/year only from one unit of the bank, reduce holding costs of the letter of does not have tax ID number which must saved for 30 years, and reduces the registration time ±1 min/customer.

The Consequences of Lean Six Sigma on Banking Improvement

3.3

235

Using Small Table as the Temporary Place of Customer Service’s Documents

Job specification of customer service indeed requires customer service does some work on office documents when they are not serving the customers. Human- machine systems of customer service working area is not optimal, caused failure mode “customer service is also doing other work while serving customers” and “customer service desk capacity is not optimal”. Besides add smart IP camera for evaluating the performance of customer service, wisely the bank should be able to provide job satisfaction to customer service. Good reversible reaction between the company and its employees will make the company better. The bank can using small table as the temporary place of customer service’s documents to add more function of an existing table. So, when customers are coming, customer service can put a work-in-processes file at the small table and then stood to greet customers and so on. It can be helpful for customer service to concentrating when serving customers. Customers also gain comfort with the customer service desk when it was neat and they will feel privileged with that empathetic treatment. If the bank issuing about Rp. 1,000,000 or USD 74.16 for two tables, then the benefit is customer and employees comfortable that have an impact on customer satisfaction and bank productivities where the value is greater than the cost. 3.4

Changes the Layout and Queuing System

Changes to the layout and queuing system can be used to overcome many failure modes specifically “still using the numbering queuing system”, “the layout is not optimal”, “services at the customer service desk does not use a queue number but do not have the queue line”, “do not use a queuing system which customer’s lined up”, “customers who open new accounts have to sit down again when the file is in the process from customer service to teller”. Queuing system which expected to provide positive change is accompanied with queue barrier rope so that customers can line up neatly. However, parallel to the barrier rope need put a minimalist chairs so that when long queues or existed a bad communication network of the bank, customers who do not want to stand can sit and remain in the queue. Layout changes design created by activity relationship chart and change the concept of “push” to “pull” and create a more efficient flow of value. Changes to the layout and the queuing system can provide enormous benefit in terms of bank productivity and customer conveniences with cost incurred only form of activities to change the layout by unit managements mutual cooperation or contracted and the cost of a queue barrier procurement around Rp. 6.4 million or USD 474.6. 3.5

Using Chair for Security

Failure mode “Security stand too long and doing repetitive work” can be overcome by allowing security to sit while no customers coming with an height adjustable chair. It is intended that security can perform duties that are repetitive standing and sitting

236

N.T. Putri et al.

comfortably. This recommendation can provide benefits that is employees convenience where impact on productivity and customer satisfaction is more valuable than the cost is around 500,000 or USD 37.08 for a single chair. 3.6

Using Visual Aid

Failure mode “lack of notice to prepare the document before customers arrived in teller” and “the lack of visual aid” can be overcome by actively using the visual aid. Visual aid was needed is an appeal for customers to prepare the documents prior to teller desk installed at the site will be queued. In addition, the requirements of transactions should be displayed visually in the location that is often seen by customers. One of the locations is around the transaction attribute table.

4 Conclusion and Discussion The six recommendations cost which can used for more than three years is around Rp. 11,000,000 or USD 815.72 with currency exchange rate Rp. 13,485 equal USD 1. That cost is around 0.036% of the bank unit net profit for three years of Rp. 30,548,294,741 or USD 2,265,353.71. The value of 0.036% can give a vision to achieve defect-free services without reducing job satisfaction of employees, USD 10.01/unit of the bank/year saving form convert two sheets into one sheet, reduce the certificate of lack of tax ID number holding costs which a certificate must keep for 30 years, reduces the registration time ±1 min/customer, the convenience of customer service employees and their customers, increasing security job satisfaction, and significantly increasing system efficiency and customer loyalty. So, lean six sigma is the pursuit of system perfection which not merely about extort more employees sweat, but work with them to make continuous innovations and improvements so that the balance between the two parties obtained not at high cost but at a cost that may be very cheap.

References 1. Womack, J.P., Jones, D.T., Roos, D.: The Machine that Changed the World: The Story of Lean Production. Rawson Associates, New York (1990) 2. Womack, J., Jones, D.T.: Lean Thinking: Banish Waste and Create Wealth for Your Corporation. Simon and Schuster, New York (1996) 3. Neha, S., Singh, M.G., Simran, K., Pramod, G.: Lean manufacturing tool and techniques in process industry. Int. J. Sci. Res. Rev. 2, 54–63 (2013) 4. Arfmann, D., Federico, G.T.P.: The value of lean in the service sector: a critique of theory and practice. Int. J. Bus. Soc. Sci. 5(2), 18–24 (2014) 5. Schroeder, R.G., Linderman, K., Liedtke, C., Choo, A.S.: Six sigma: definition and underlying theory. J. Oper. Manag. 21(2) (2008) 6. Selvi, K., Majumdar, R.: Six sigma-overview of DMAIC and DMADV. Int. J. Innovative Sci. Mod. Eng. (IJISME) 2(5), 16–19 (2014)

The Consequences of Lean Six Sigma on Banking Improvement

237

7. de Koning, H., Does, R.J.M.M., Bisgaard, S.: Lean six sigma in financial services. Int. J. Six Sigma Competitive Advantage 4(1), 1–17 (2008) 8. Bubshait, A.A., Al-Dosary, A.A.: Application of lean six sigma methodology to reduce the failure rate valves at oil field. In: Proceedings of the World Congress on Engineering and Computer Science (2014) 9. Hasssan, M.K.: Applying lean six sigma for waste reduction in a manufacturing environment. Am. J. Ind. Eng. 1(2), 28–35 (2013) 10. Enoch, O.K.: Lean six sigma methodologies and organizational profitability: a review of manufacturing SMEs in Nigeria. Am. J. Ind. Bus. Manag. 2013(3), 573–582 (2013) 11. Mandahawi, N., Fouad, R.H.: An application of customized lean six sigma to enhance productivity at a paper manufacturing company. Jordan J. Mech. Ind. Eng. 6, 103–109 (2012) 12. Panat, R., Dimitrova, V., Selvamuniandy, T.S., Ishiko, K., Sun, D.: The application of lean six sigma to the configuration control in Intel’s manufacturing R&D environment. Int. J. Lean Six Sigma 5(4), 444–459 (2014)

Author Index

B Bassano, Clara, 178 Berrezueta-Guzman, Jonnathan, 73 Bodendorf, Freimut, 50, 128 Börsig, Julian, 37 Brauner, Philipp, 11 Brenner, Florian, 37 Brown, Carolyn Hamasaki, 63 C Calero Valdez, André, 11 Caparas, Hazel, 215 F Fujimura, Kazuhiro, 203 G Ganz, Walter, 103 Gunawan, Andi, 227 H Haeske, Julia, 128 Haurat, Adrian, 140 Hidaka, Kazuyoshi, 115 I Iwano, Kazuo, 115 J Janneck, Monique, 3 Jent, Sophie, 3 K Kohda, Youji, 115 Korn, Oliver, 37 Kurata, Yoshiki B., 193

L Lalli, Fabio, 37 Lin, Fu-Ren, 162 Liu, Li, 140 M Mariano, Lizbeth A., 82 Martuscelli, Linda, 178 Matias, Aura C., 82, 90, 193, 215 Mattmüller, Maik, 37 Müller, Andrea, 37 N Neuhüttler, Jens, 103 Niels, Adelka, 24 P Petrov, Yoan, 50 Philipsen, Ralf, 11 Piazza, Alexander, 50, 128 Piciocchi, Paolo, 178 Pietronudo, Maria Cristina, 178 Putri, Nilda Tri, 227 R Robles-Bykbaev, Vladimir, 73 S Satterfield, Debra, 153 Scocozza, Michele, 178 Sebastian, Vinaya, 63 Sedilla, Keneth, 90

© Springer International Publishing AG 2018 L.E. Freund and W. Cellary (eds.), Advances in The Human Side of Service Engineering, Advances in Intelligent Systems and Computing 601, DOI 10.1007/978-3-319-60486-2

239

240 Serpa-Andrade, Luis, 73 Siddike, Md. Abul Kalam, 115 Simmons, Denise R., 173 Spohrer, James C., 178 Spohrer, Jim, 115 Sutanto, Agus, 227

Author Index W Woyke, Inka C., 103 Z Zagel, Christian, 24, 50, 128 Ziefle, Martina, 11