Knowledge Management And Organizations : Process, System And Strategy 9781845444556, 9780861768288

Citation preview

Business Process Management Journal

ISSN 1463-7154 Volume 9 Number 3 2003

Knowledge management and organizations: process, system and strategy Guest Editors Young-Gul Kim, Heeseok Lee and Ingoo Han

Access this journal online __________________________ 251 Advisory group and editorial board _________________ 252 Abstracts and keywords ___________________________ 253 Guest editorial ____________________________________ 256 Invited viewpoint: the effective knowledge organization William R. King ________________________________________________

259

Visual navigation of distributed knowledge structures in groupware-based organizational memories Stefan Smolnik and Ingo Erdmann _________________________________

261

A framework for designing a workflow-based knowledge map Intae Kang, Yongtae Park and Yeongho Kim ________________________

281

Do we practise what we preach? Are knowledge management systems in practice truly reflective of knowledge management systems in theory? Nilmini Wickramasinghe _________________________________________

Access this journal electronically The current and past volumes of this journal are available at

www.emeraldinsight.com/ft You can access over 100 additional Emerald journals, each with a comprehensive searchable archive of articles (many dating back to 1989), a detailed classification system and links to referenced material. See page following contents for full details of what your access includes.

295

CONTENTS

CONTENTS continued

Knowledge conversion with information technology of Korean companies Hyun-Soo Lee and Yung-Ho Suh___________________________________

317

Interaction management: the next (and necessary) step beyond knowledge management Stephen W. Smoliar _____________________________________________

337

Experiences from knowledge management implementations in companies of the software sector Dimitris Apostolou and Gregoris Mentzas____________________________

354

The potential use of knowledge management for training: a review and directions for future research Joseph R. Muscatello _____________________________________________

382

www.emeraldinsight.com/bpmj.htm As a subscriber to this journal, you can benefit from instant, electronic access to the content of this title. Your access includes a variety of features that increase the value of your journal subscription.

E-mail Services Emerald’s range of free e-mail alerting services is designed to deliver personal notification of news and features in a number of different interest areas.

Automatic permission to make up to 25 copies of individual articles

Emerald WIRE (World Independent Reviews) Emerald WIRE is a fully searchable, subject-specific database brought to you by Emerald Management Reviews. It provides article reviews from the world’s top management journals. The database is updated monthly and gives users details of how to obtain the full text of original articles.

This facility can be used for teaching purposes, training courses, meetings or seminars. This only applies to articles of which Emerald owns copyright. For further details visit www.emeraldinsight.com/copyright

How to access this journal electronically To benefit from electronic access to this journal you first need to register on the Internet. Registration is simple and full instructions are available online at www.emeraldinsight.com/register Once completed, your institution will have instant access to the journal content from www.emeraldinsight.com or via the journal homepage www.emeraldinsight.com/bpmj.htm

Research Register Located at www.emeraldinsight.com/researchregister, the Emerald Research Register is an Internet research forum where you and your readers can identify information on research activity world-wide. This feature is invaluable if you are seeking information or wish to disseminate information about your own research.

Our liberal institution-wide licence allows everyone within your institution to access your journal electronically, thus making your subscription more cost-effective. Our Web site has been designed to provide you with a comprehensive, simple system that needs only minimum administration. Access is available via IP authentication or username and password.

Support Resources Comprehensive librarian and user toolkits have been created to help you get the most from your journal subscription. For further information about what is available visit www.emeraldinsight.com/usertoolkit

Key features of Emerald electronic journals

Electronic access to this journal is available via a number of channels, enabling end users and libraries to reach the content through their preferred delivery system. The Emerald Fulltext Web site – www.emeraldinsight.com/ft – is the recommended means of electronic access, as it provides fully searchable and high value-added access to the complete content of the journal.

Online Publishing and Archiving You can gain access to past volumes as well as new material from this journal on the Internet via Emerald Fulltext. You can browse or search the database for relevant articles. Key Reading This feature provides abstracts of articles chosen by the journal editor on the basis of their subject-matter. These are selected to provide readers with current awareness of interesting articles from other publications within the same field. Reference Linking Direct links are provided from the journal article references to abstracts of the most influential articles cited (where possible this is to the full text of the article). E-mail an Article This facility allows you to e-mail relevant and interesting articles in PDF format to another PC for later use, reference or printing purposes. Non-Article Content The non-article content of our journals (product and materials information, industry trends, company news, etc.) is available online so that it can be accessed by users. It is possible to browse or search non-article content by keyword.

Additional Complementary Services Available Your access includes a variety of features that add to the functionality and value of your journal subscription:

Choice of Access

Subscribers can also access and search the article content of this journal through the following journal delivery services: EBSCOhost EJS journals.ebsco.com Huber E-Journals e-journals.hanshuber.com/english/index.htm Minerva www.minerva.at OCLC Firstsearch Electronic Collections Online www.uk.oclc.org/oclc/menu/eco.htm RoweCom’s Information Quest www.informationquest.com SilverPlatter www.silverplatter.com SwetsBlackwell’s SwetsWise www.swetswise.com

Emerald Customer Support Services For customer service and technical help, contact: E-mail [email protected] Tel +44 (0) 1274 785278 Fax +44 (0) 1274 785204

BPMJ 9,3

ADVISORY GROUP Professor Thomas H. Davenport Partner and Director, Accenture Institute for Strategic Change, Cambridge, MA, USA Professor Varun Grover William S. Lee Distinguished Professor of Information Systems, Clemson University, USA

252

Dr H. James Harrington Principal, Ernst & Young, California, USA Professor Gopal Kanji Dept of Applied Statistics & Operational Research, Sheffield Hallam University, Sheffield, UK Dr John Peters Emerald, 60/62 Toller Lane, Bradford, UK Professor N. Venkatraman David J. McGrath Jr Professor of Management, Boston University, USA EDITORIAL BOARD Professor W.M.P. van der Aalst Department of Information and Technology, Eindhoven University of Technology, Eindhoven, The Netherlands Professor Hassan Abdalla De Montfort University, Leicester, UK Professor Zaitun Abu Bakar Faculty of Computer Science and Information Technology, University of Malaya, Kuala Lumpur, Malaysia Dr Asamir Abuznaid Hebron University, Palestinian Authority Professor Adel M. Aladwani Department of Quantitative Methods & Information Systems, College of Business Administration, Kuwait University, Kuwait

Professor Walter W.C. Chung Department of Industrial and Systems Engineering, Hong Kong Polytechnic University, Hung Hom, Hong Kong Professor Barrie Dale United Utilities Professor of Quality Management, Manchester School of Management, UMIST, Manchester, UK Professor Georgios I. Doukidis Department of Management Science and Technology, Athens University of Economics and Business, Athens, Greece Professor A. Sharaf Eldin Faculty of Computers and Information, Helwan University, Cairo, Egypt Professor A. Gunasekaran University of Massachusetts, USA Professor Suliman Hawamdeh School of Communication and Information, Nanyang Technological University, Singapore Professor Kai Jakobs Computer Science Department, Technical University of Aachen, Germany Chan Meng Khoong Managing Principal, Atos Origin, Singapore Dr Peter Kueng Credit Suisse, Zurich, Switzerland Professor Binshan Lin Department of Management & Marketing, Louisiana State University in Shreveport, USA Professor Pericles Loucopoulos Chair of Information Systems, Department of Computation, UMIST, Manchester, UK

Professor Mustafa Alshawi Faculty of Business and Informatics, University of Salford, UK

Professor Michael Rosemann Director of the Centre for IT Innovation, Queensland University of Technology, Brisbane, Australia

Professor Atta Badii Department of Information Systems, University College Northampton, Northampton, UK

Professor John Sharp Faculty of Business and Informatics, University of Salford, Salford, UK

Professor Saad Haj Bakry College of Engineering, King Saud University, Saudi Arabia

Professor Namchul Shin Pace University, School of CSIS, New York, USA

Professor David Bennett Aston Business School, Birmingham, UK Business Process Management Journal Vol. 9 No. 3, 2003 p. 252 # MCB UP Limited 1463-7154

Dr Ilia Bider Director R&D, IbisSoft AB, Stockholm, Sweden

Professor Sergio Beretta SDA Bocconi, Milan, Italy

Professor Anthony Wensley Rotman School of Business, Ontario, Canada

Visual navigation of distributed knowledge structures in groupware-based organizational memories Stefan Smolnik and Ingo Erdmann Keywords Knowledge management, Groupware, Vision Many of today’s organizations already have a strong integration of groupware systems within their IT-infrastructure. The shared databases of these groupware systems form organizational memories, which comprise the complete knowledge of an organization collected over the time of its existence. One key problem is how to find relevant knowledge or information in continuously growing and distributed organizational memories. In many cases, the basic functionalities and mechanisms of groupware systems are not sufficient to support users in finding required knowledge or information. Topic maps provide strong paradigms and concepts for the semantic structuring of link networks and, therefore, they are a considerable solution for organizing and navigating large and continuously growing organizational memories. The K-Discovery project suggests applying topic maps to groupware systems to address the mentioned challenges. Thus, the K-Discovery project introduces a conceptual framework, an architecture, and an implementation approach to create knowledge structures by generating topic maps from organizational memories and offers navigation tools to exploit the created structures.

A framework for designing a workflow-based knowledge map Intae Kang, Yongtae Park and Yeongho Kim Keywords Knowledge mapping, Workflow software, Integration, Prototyping A complete form of knowledge management system comprises both process management and contents management. Process management is concerned with handling activities to generate and utilize knowledge,

whereas contents management deals with knowledge contents themselves. Workflow, considered as a core component of process management, is to define and administer business processes automatically. A knowledge map is a representation tool to visualize knowledge sources and relationships among knowledge artifacts. Noting the crucial needs to integrate process management and contents management and recognizing that previous research has paid little attention to this issue, we propose a framework for developing a workflow-based knowledge map. The proposed process-perspective knowledge map takes the structure of processes and tasks defined in workflow into account. We then materialize the framework by developing a prototype and applying it to the car seat design process of the automobile industry. The integration represents an exploratory effort to combine process management and contents management.

Do we practise what we preach? Are knowledge management systems in practice truly reflective of knowledge management systems in theory? Nilmini Wickramasinghe Keywords Knowledge management, Knowledge management systems, Knowledge workers, Organization, Memory, Competitive advantage Knowledge management systems are predominant in both theory and practice. However, are the same systems discussed in theory actualized in practice? By comparing and contrasting knowledge management systems in theory and practice, this paper demonstrates that they are indeed dissimilar. In theory, they have both subjective and objective components. In practice, only the objective component appears to be actualized; hence, these systems in practice are essentially organizational memory systems at best and not knowledge management systems at all. By unravelling the mystique of knowledge management systems, this paper exposes a fundamental anomaly. Further, an apparent

Abstracts and keywords

253

Business Process Management Journal Vol. 9 No. 3, 2003 Abstracts and keywords q MCB UP Limited 1463-7154

BPMJ 9,3

254

void currently in practice is highlighted; namely, the lack of the subjective component of knowledge management systems in practice. They are being heralded as key systems that are vital for organizations to survive and thrive in the intense competitive environment of the information age. Surely then, a system that in practice supports not only the objective component, but also the subjective component of knowledge management, would indeed be a truly powerful system.

Knowledge conversion with information technology of Korean companies Hyun-Soo Lee and Yung-Ho Suh Keywords Knowledge management, Knowledge-based organizations, Information technology This research aims at finding out how the efforts of Korean enterprises to convert data and information into knowledge are related to the use of information technology. It also aims to come up with plans to efficiently introduce external knowledge by applying information technology. This research was conducted focusing on the cases of enterprises that are carrying out knowledge management by applying information technology and the result shows that Korean enterprises place high value on the application of information technology in introducing knowledge management. However, the use of information technology is not common to all enterprises in four kinds of knowledge conversion. For an organization to efficiently acquire and apply valuable knowledge, the boundary of an enterprise should be extended, and for this, the enterprise should be changed into an extended knowledge enterprise, which is founded on the trust and compensation of all the interested parties. In an extended knowledge enterprise, its first consideration is to acquire knowledge that can efficiently manage and control the interests of customers, suppliers, public institutions, and competitors that exist in the surroundings and to acquire the core knowledge that can make those

people and institutions participate. In addition, the organization should possess knowledge base and information communication technology that can smoothly convert data and information into knowledge, by efficiently connecting them.

Interaction management: the next (and necessary) step beyond knowledge management Stephen W. Smoliar Keywords Knowledge management, Information exchange, Crisis management Purveyors of knowledge management software have a disconcerting tendency to promote the myth that all problems may be solved by more powerful tools for the exchange of information in the workplace. This fallacy is based on the faulty assumption that knowledge management is about the management of knowledge (as if knowledge were a commodity that could be managed), as opposed to the management of people whose work depends critically on what they know. The origins of knowledge management are far more firmly rooted in the psychological legacy of organizational communication than they are in the technological legacy of information management systems. However, even organizational communication is an inadequate foundation, since various schools of thought in social theory, particularly the structuration theory of Anthony Giddens, inform us that interaction (in the workplace or in any other social setting) is not strictly limited to communication. Knowledge management thus requires moving beyond simplistic models of information exchange to more challenging problems of leveraging social interaction to the advantage of the enterprise. This paper focuses on the claim of structuration theory that the dimension of communication should be supplemented with additional dimensions of power and sanction. This perspective is then examined in light of a case study of crisis management practices, and the case study provides a basis for

addressing implications for technological support.

Experiences from knowledge management implementations in companies of the software sector Dimitris Apostolou and Gregoris Mentzas Keywords Knowledge management, Software, Knowledge workers, Networks The Know-Net knowledge management solution, that includes a theoretical framework, a consulting method and a software tool, is based on a knowledge asset-centric design that innovatively fuses the process or human approach with the product or content approach of knowledge management. This paper describes how Know-Net was applied in four companies of the software sector. It outlines how specific business areas such as an R&D unit, the bid management process and collaboration between geographically-dispersed teams can benefit from knowledge management. It outlines the role specific Know-Net components played within the transformation of existing business processes and structures, and provides key recommendations based on this experience.

The potential use of knowledge management for training: a review and directions for future research Joseph R. Muscatello Keywords Knowledge management, Training, Information exchange The powerful competitive traits of a learning organization are well documented. Companies in the future will have to provide new and innovative techniques if they hope to survive and prosper. These techniques will revolve around knowledge management. However, research on using knowledge management as a tool to increase learning is minimal. Knowledge management systems provide a new vehicle to effectively and efficiently disseminate knowledge, skill and expertise throughout an organization. Knowledge management is an outgrowth of the “information society”. Currently, research and implementation of knowledge has centered on the outside customer service and outside customer relationships. Little has been written about the use of knowledge to serve the training needs of the firm. This study combines the powerful competitive traits of a learning organization with knowledge management traits to potentially increase the positive effects of both.

Abstracts and keywords

255

BPMJ 9,3

256

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 256-258 q MCB UP Limited 1463-7154

Guest editorial About the Guest Editors Young-Gul Kim is a professor at the Graduate School of Management, Korea Advanced Institute of Science and Technology, Seoul. He received his BS and MS degree in Industrial Engineering from the Seoul National University, Korea and PhD in MIS from the University of Minnesota. His active research areas are: knowledge management, IT innovation, and customer relationship management. His publications have appeared in various journals such as Communications of the ACM, Journal of MIS, Journal of Strategic Information Systems, Decision Support Systems, Information & Management, IEEE Transactions on Engineering Management, and others. He has also presented several papers at ICIS, HICSS, and DSI conferences. E-mail: [email protected] Heeseok Lee is a professor of MIS at the Korea Advanced Institute of Science and Technology, Seoul. He received his BS degree in Industrial Engineering from the Seoul National University, Korea, his MS degree in Industrial Engineering from the Korea Advanced Institute of Science and Technology, Korea, and his PhD in Management Information Systems from the University of Arizona, USA. He taught MIS at the University of Nebraska, Omaha from 1992-1994. His research interests include knowledge management, databases and information strategy. His papers have been published in Journal of Management Information System, Journal of Database Management, Information and Management, Omega, Information Systems, Journal of Systems and Software, Information Processing and Management, and Annals of Operations Research. E-mail: [email protected] Ingoo Han is a professor at the Graduate School of Management, Korea Advanced Institute of Science and Technology, Seoul. He received his BS degree in International Economics from the Seoul National University, Korea, his MS degree in Industrial Engineering from the Korea Advanced Institute of Science and Technology, Korea, and his PhD from the University of Illinois at Urbana-Champain. His research interests are information systems audit and security, information system evaluation, and AI applications in accounting and finance. His recent research issues include the audit and control under EC, prediction of stock price using neural network, and integration of AI techniques. E-mail: [email protected]

Over the past decade, understanding, identifying, and strategic management of organizational knowledge resource has become a competitive necessity. Microsoft, whose asset value is one fifth of General Motors, commands a market value ten times as big as that of General Motors. Knowledge-based organizations such as Microsoft and Intel are fast replacing the giant asset-based organizations of the past as the front-runner for the 21C knowledge-based economy. From a conceptual point of view, knowledge management can be interpreted in two ways: knowledge-based management and knowledge resource management. Knowledge-based management is a strategy to achieve the vision of becoming a knowledge-based organization. It triggers a series of changes in its business processes, organizations, people, and business practices. Functional units with iron walls change to cross-unit communities of practice. Decision making changes from intuition-based to analysis-based. Notinvented-here (NIH) syndrome turns into a boundaryless knowledge echo system with suppliers, customers, and outside experts.

Knowledge resource management, which frequently has been the focus of organizational knowledge management initiatives, involves the implementation of various architectures, methods and tools to support knowledge-based management. Just as information strategy planning helps development and management of enterprise-wide information resources, knowledge strategy planning helps organizations in creating and managing their knowledge resources. Knowledge management systems as well as knowledge maps, repositories, search engines, analytic models, video conferencing, etc. comprise the toolbox for effective and efficient knowledge sharing and utilization over the time and space constraints, owing to geographical dispersion of knowledge resources. In addition to tools, successful knowledge resource management also requires a set of diverse policies and processes regarding evaluation, incentives, role design, and resource allocation. In this special issue on knowledge management, we received 19 papers from all over the world: six from the USA, six from Europe, five from Asia, and two from Australia. Of the 19 submissions, we selected seven papers, based on their academic rigor, industry relevance, and potential to ignite a new stream of future studies in knowledge management research and practices. The paper by Nilmini Wickramasinghe, “Do we practice what we preach: are knowledge management systems in practice truly reflective of knowledge management systems in theory?” challenges the purpose, role, and capability of the currently implemented knowledge management systems through in-depth case studies in the consulting industry. The author not only identifies the limitations of the current KMSs but also suggests a future direction and strategies to overcome such limitations. The paper by Stephen Smoliar, “Interaction management: the next (and necessary) step beyond knowledge management” argues that knowledge management needs to move beyond simplistic models of information exchange, to more challenging problems of leveraging social interaction to the advantage of the enterprise. To illustrate its argument, the author examines a case study of crisis management practices and suggests implications for technological support. The paper by Intae Kang, Yongtae Park, and Yeongho Kim, “A framework for designing a workflow-based knowledge map” introduces a framework for developing a workflow-based knowledge map in order to integrate process management and contents management, two areas of crucial importance to enterprise-wide knowledge management. The authors also developed a prototype and applied it to the car seat design process of the automobile industry. The paper by Stefan Smolnik and Ingo Erdmann, “Visual navigation of distributed knowledge structures in groupware-based organizational memories” introduces the topic map as a tool for semantically structuring link networks to organize and navigate large and continuously growing organizational memories.

Guest editorial

257

BPMJ 9,3

258

The paper by Hyun-Soo Lee and Yung-Ho Suh, “Knowledge conversion with information technology of Korean companies” discusses the viability of “extended knowledge enterprise” where, through using information technologies, not only internal units, but also external knowledge sources such as customers, suppliers, public institutions, and even competitors can participate. The paper by Dimitris Apostolou and Gregoris Mentzas, “Experiences from knowledge management implementations in companies of the software sector” provides a useful knowledge management solution, called Know-Net, that includes a theoretical framework, a consulting method and a software tool. It outlines how specific business areas such as R&D, the bid management process, collaboration between geographically dispersed teams, can benefit from knowledge management Finally, the paper by Joseph R. Muscatello (“The potential use of knowledge management for training”) discusses the use of knowledge management to serve the training needs of firms. This study combines the powerful competitive traits of a learning organization with knowledge of management traits to potentially increase the positive effects of both. We hope that the audience of this special issue can grasp the theoretical underpinnings, implementation pitfalls, and available methods and tools for successful knowledge management. We also expect that, unlike the fleeting management techniques, knowledge management will survive the test of time to become one of the most important management paradigms for the twentyfirst century. Young-Gul Kim, Heeseok Lee and Ingoo Han

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

Invited viewpoint

Invited viewpoint

The effective knowledge organization William R. King Katz Graduate School of Business, University of Pittsburgh, Pittsburgh, Pennsylvania, USA Knowledge management (KM) is currently very popular in corporations and among academics. Recently, a colleague of mine suggested that KM is merely a fad that will not exist in a few years. Although I think that KM is very important, I agree with him at one level. Just as the term “total quality management” (TQM) was popular in the 1980s and early 1990s, and now is seldom used by managers, I would not be surprised if the term “KM” also falls into disuse. The reason that this is so with TQM, and will likely be so with KM, is that these activities have and will become integrated into the normal operations of business. Just as there were many people with titles such as “VP of quality” in the 1980s, there are now many “chief knowledge officers”. The former title has all but disappeared, as will the latter one, but not nearly as rapidly. “Knowledge officers” tend to focus on knowledge per se rather than on the applications and consequences of knowledge. Nonetheless, these positions will exist and be useful for some time if firms recognize and organize for their various knowledge-related activities to create what I designate as an “effective knowledge organization” (EKO) – an organization that creates, explicates and communicates knowledge and applies it to create improved decision making and organizational behaviors that positively influence overall performance. An EKO is made up of a number of elements, each distinct because the activities and the required underlying cultures are so different, but which, taken collectively, can make the overall organization more effective. These knowledge-related components of an EKO are: . An individual learning component which focuses on developing human intellectual capital by enhancing the knowledge of individuals through such mechanisms as formal class work, e-learning and mentoring. . An organizational development component which utilizes adaptive learning to create social capital through such devices as the use of teamwork and partnerships and other strategic alliances with other firms. . An intellectual property management component that deals with knowledge that has already been converted into explicit form such as patents, research reports and trademarks, with the objective of generating greater revenues for the firm through licensing, brand extensions and the creation of new products.

259

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 259-260 q MCB UP Limited 1463-7154

BPMJ 9,3

.

.

260

.

An innovation component which focuses on generative learning to create new products and processes through creative thinking and which manifests itself in forms such as “idea fairs”, and small financial grants for exploring new ideas. A knowledge management component which focuses on the explication of the tacit knowledge that exists in the minds of organizational participants and that knowledge which is embedded and in the process and relationships of the organization. Once made explicit, such knowledge can be stored and communicated throughout the organization. An information/communication system infrastructure which enables these components to operate and to be integrated into an overall supporting structure for the EKO.

An organization that develops these knowledge-related capabilities and which implements them in an integrated manner, can become an EKO which uses and enhances its knowledge, implements improved behaviors and decision making, and improves its overall performance. Organizations that do not develop these capabilities will, in the long term, become unable to compete with those that have done so. At some point, just as was the case with quality, these capabilities will become an integral part of the various business activities that the EKO conducts, at which point there will be little need for executives who hold the title of “knowledge officer”.

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

Visual navigation of distributed knowledge structures in groupware-based organizational memories

Visual navigation

261

Stefan Smolnik and Ingo Erdmann Department of Business Computing, University of Paderborn, Paderborn, Germany Keywords Knowledge management, Groupware, Vision Abstract Many of today’s organizations already have a strong integration of groupware systems within their IT-infrastructure. The shared databases of these groupware systems form organizational memories, which comprise the complete knowledge of an organization collected over the time of its existence. One key problem is how to find relevant knowledge or information in continuously growing and distributed organizational memories. In many cases, the basic functionalities and mechanisms of groupware systems are not sufficient to support users in finding required knowledge or information. Topic maps provide strong paradigms and concepts for the semantic structuring of link networks and therefore, they are a considerable solution for organizing and navigating large and, continuously growing organizational memories. The K-Discovery project suggests applying topic maps to groupware systems to address the mentioned challenges. Thus, the K-Discovery project introduces a conceptual framework, an architecture, and an implementation approach to create knowledge structures by generating topic maps from organizational memories and offers navigation tools to exploit the created structures.

1. Introduction and overview Within many organizations, groupware-based environments are the basis for communication and information management. In many cases the groupware infrastructure is used to support teams in communication, coordination, and collaboration. (The delimitation of these three terms is not homogeneous in literature. For an overview of differing controversial viewpoints please consult Bornschein-Grass (1995). We are referring to the definition by Lotus Development Corporation (1995).) Effective access mechanisms to knowledge objects are needed because of their increasing amount in various shared information spaces of a groupware-based infrastructure. The following short case illustrates the need for these mechanisms: within the groupware competence center (GCC) of the University of Paderborn, the groupware-based system environment Lotus Notes/Domino is used in almost every part of operative work. There are several core applications in different contexts, such as daily office work, Web publishing, teaching and examinations, projects and cooperation, research, and literature maintenance. These organizational memories consist of more than 30 databases with more than 60,000 documents. Even though there are a lot of semantic relations between the

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 261-280 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477885

BPMJ 9,3

262

information stored in these databases, it is not possible to navigate between them or to identify knowledge structures. The capabilities for accessing knowledge objects are restricted to full text searching and navigating through context sensitive views and category structures. Full text indexes are not sufficient for searching information, and structures like document types or taxonomies are sometimes too constraining to qualify or categorize information (Biezunski and Newcomb, 2001). Furthermore, usage and scope of these techniques is limited to a single database. Analogous to this practical case, many organizations have a strong integration of groupware-based systems, both within their IT-infrastructure and within their working environment. To name only two examples, look at Deutsche Bank and PriceWaterhouseCoopers. Both of them use an extensive IT-infrastructure based on Lotus Notes/Domino. With an increasing integration of these groupware environments in internal business processes, organizational memories grow fast. These shared databases, often existing in groupware-based systems, allow transformation of individual knowledge into common knowledge. Looking at the technical perspective of knowledge management, groupware seems to be a suitable platform for managing organizational memories (Schliwka, 1998; Davenport and Prusak, 1998). The growing flood of available information requires powerful concepts and mechanisms to support users, who search for relevant information and knowledge objects. Mechanisms of navigation and linking, as well as functionalities for extensive searches and investigations, are needed to explore and use complex offers of knowledge. These functionalities are a necessary prerequisite for the core processes of knowledge identification and knowledge use (Probst et al., 1999). Therefore, effective search and navigation mechanisms which improve organizational use of existing individual and common knowledge objects contribute to the process of knowledge generation, i.e. the development or collection of new knowledge (Gu¨ldenberg, 1998). Particularly, visual navigation tools add an additional efficiency benefit to this process of knowledge acquisition. Topic maps – as defined in ISO/IEC 13250 – used on information sets create knowledge structures and form a structured semantic link network on a large set of information (Rath and Pepper, 1999). Thus, topic maps are a reasonable basis to realize the mentioned mechanisms and functionalities for the identification of relevant knowledge objects. In this contribution, we describe the fundamental idea of the research project K-Discovery[1]: using topic maps to identify distributed knowledge structures in groupware-based organizational memories. In addition to the existing techniques of search and navigation, like hierarchical navigation in categories of views or full text search, the user can be supported by the strong concept of associative navigation in, and visualization of, semantic networks provided by topic maps (Ahmed, 2000). Further, a groupware-based implementation of topic maps

benefits from several aspects of groupware paradigms and technologies, like distributed database architectures and replication, workflow management, or security and access control mechanisms. The purpose of this paper is to discuss technologies and concepts to support users in finding relevant knowledge in distributed organizational memories efficiently by combining groupware paradigms, topic maps, and visualization techniques. The described concepts provide a framework to address the limitations of hierarchical navigation and full text search in groupware environments. Therefore, they try to contribute to the research fields of computer-supported cooperative work (CSCW), knowledge management, and knowledge visualization. Complementarily, the prototypic implementations reveal practical feasibility and added value for users. This paper is structured as follows: the basic concepts of topic maps and the design of groupware-based applications are briefly described in section 2. In section 3, we present the conceptual approach and primary idea of the K-Discovery project: applying topic maps to groupware-based organizational memories to identify distributed knowledge structures. In section 4, we discuss concepts for visualization and navigation of groupware-based topic maps. First, requirements of an efficient navigation tool for structured data and information spaces are defined. Second, we suggest a concept of simplifying the relation structure of knowledge objects to enhance navigation efficiency of groupware-based topic maps. Third, we introduce and discuss selected navigational concepts for the exploration of organizational memories. Section 5 outlines concluding remarks, the current state of the project and future opportunities. 2. Theoretical background and basic terms 2.1 Introduction to topic maps With the ongoing increase of information in the intranets of organizations as well as on the Internet, a need for new information management technologies emerges. So far, several technologies have been developed to address different aspects of information management, e.g. computer networks, document management systems, workflow management systems or, recently, different kinds of portals. As pointed out above, with growing organizational memories on one side, and lack of powerful and effective mechanisms and functionalities for navigating, linking, searching and investigating on the other side, the need for enhanced access mechanisms to explore and use the complex offer of knowledge is getting more and more evident. To address these shortcomings the ISO standard ISO/IEC 13250 Topic Maps defines a model and architecture for the semantic structuring of link networks. By applying topic maps to large sets of heterogeneous information resources, reusable structured semantic link networks are created on top of those resources. Furthermore, the standard provides navigation paradigms to enable users to search knowledge structures (Rath and Pepper, 1999).

Visual navigation

263

BPMJ 9,3

264

Figure 1. Topics

The key concepts of topic maps are topics, occurrences of topics, and relationships between topics (topic associations). First, a topic is a construct that represents a real world subject, and, in this sense, a topic can be almost everything: a theme, a concept, a subject, a person, an entity, etc. (Figure 1). A concrete topic is an instance of a topic type. Therefore, a topic and a topic type form a class-instance relationship. At the same time, a topic type is also a topic. Topics have three kinds of characteristics: names, occurrences and roles in associations. Occurrences and roles in associations are described in detail later on. The base name of a topic is required. In addition, topics can have a display name and a sort name. These concepts apply to multilingual scenarios, or to the use in different geographical regions. An occurrence is a link to one or several real knowledge objects, like a report, a comment, a video or a picture (Figure 2). Generally, an occurrence is not part of a topic map. The link mechanism itself depends on the underlying system. For example, HyTime addressing, XPointer or document links in a groupwarebased system can be used as a link mechanism. To express different kinds of occurrences, the standard provides the concept of occurrence roles that are topics as well. Topic associations describe the relationships between topics (Figure 3). They are completely independent of the real knowledge object and represent the essential value-add of the topic map. This concept leads to some conclusions: a concrete topic map can be applied to different information repositories. Seen from the opposite side, different topic maps can be applied to one information repository and, therefore, they can provide different views, e.g. for the different needs of different users. Furthermore, topic maps are interchangeable and can be merged or connected. Generally, topic associations are not one-way relationships. They are symmetric as well as transitive and thus, they have no direction. The construct of association types can be used to group topic associations and the involved topics. An association role describes the role of a topic in a topic association. Again, both the association types and the association roles are topics as well.

Visual navigation

265

Figure 2. Occurrences

Figure 3. Topic associations

BPMJ 9,3

266

Additionally, the topic map standard provides the extended concepts of scope, public subject, and facets, which are not described here. For a comprehensive overview refer to Rath and Pepper (1999) and ISO/IEC 13250 (2000). A complementary approach, which addresses the same objectives and fields of application as topic maps, is called resource description framework (RDF). A detailed introduction as well as a comparison to topic maps can be found in Moore (2001). 2.2 The design of groupware-based applications For a better understanding, the basic components of a groupware-based application are now described shortly. All applications consist of one or more databases. These databases hold the data, logic, and design elements for the application (Figure 4). The basic design elements of a groupware-based application are pages, forms, fields, and views (Lotus Development Corporation, 2001). A page is a design element that displays static information. Pages can be used anywhere in an application where text, graphics, or an embedded control are presented to users. Forms, like pages, display information. Everything that can be done with a page can be done with a form. The difference between forms and pages is that forms can be used to collect information. A form provides the structure for creating and displaying documents, and documents are used to store data in a database. The information collected in a form is saved as a document. Further, a form is used by a document as a template to provide the structure for displaying dynamic data. A field is the part of an application that collects data.

Figure 4. Design of a groupwarebased application

Fields are created on forms in particular. Each field stores a single type of information, such as text, numbers, or dates. The information of all fields of a form is stored in an individual document. The contents of fields can then be displayed in documents and views. A view is a sorted or categorized list of documents. Views are the entry point to the data stored in a database. Every database must have at least one view; usually most databases have more than one view (Lotus Development Corporation, 2001). Apart from that, groupware-based applications consist of some more design elements that are of no interest in this context; therefore, they are not described here. For a comprehensive overview, refer to Lotus Development Corporation (2001). 3. Using topic maps in groupware-based environments By applying topic maps – as defined in ISO/IEC 13250 – to groupware-based organizational memories, it is possible to close a gap between knowledge and information (Rath and Pepper, 1999). So far, views and categories in groupware systems have enabled the creation of little knowledge by conveying documents into different contexts. But this concept is limited to a weak identification of relations between them. Topic maps provide much more appropriate and sophisticated concepts to overcome this limitation and to build a semantic link network from the documents of a database. The basic design elements of a groupware-based application, especially forms, can be used to identify the main subjects (topic types) and relationships between them (association types) (Figure 5). While text analyzing methods with the support of artificial intelligence concepts could provide more or less reasonable results in identifying the main topics of a document, we have chosen another approach in the K-Discovery project. To identify the relevant subjects of documents, the fields of forms must be considered. As mentioned before in the short case, in many groupwarebased systems the knowledge objects, particularly documents, contain many fields of various kinds and purposes: obvious information, like the parts of an address (e.g. city), or the category of a report (e.g. balance sheets 2000), or keywords and linking information (e.g. persons, locations, time, free or predefined keywords). These fields form a set of potential candidates for topic types. Moreover, by looking at the forms as a whole and at the views, a lot of basic associations can be found between topic types. For example, in the context of an address form, the associations “person lives in city” or “person works for company” can be identified. Similarly, a report form contains associations like “report is composed by person” or “report belongs to category”. When considering views, the content of fields is displayed in the columns of a view. This mapping between fields and columns creates another context and thus forms a base for additional associations, like “report is in work by person” in a view that displays the state of a document in a workflow.

Visual navigation

267

BPMJ 9,3

268

Figure 5. Concepts of topic maps and groupware

Obviously, the different forms, i.e. the different document types, can be considered as occurrence roles, while documents are occurrences in a concrete topic map. These considerations about fields, forms, and views of a groupware-based application, and their connections and relationships, result in a template for a topic map. By applying this template to groupware-based organizational memories, a concrete topic map can be built. In particular – following the definitions of topic types, association types, and occurrence roles made in the template – topics, topic associations, and occurrences are derived from all documents in groupware-based organizational memories (for architectural and implementation approaches refer to Smolnik and Nastansky (2002)). These elements form a topic map (Figure 6). The topic map provides a structured semantic link network and therefore allows easy and selective navigation to the required knowledge objects, independent of their origin (Rath and Pepper, 1999). In contrast to full text search and hierarchical navigation, searching in knowledge structures is enabled, i.e. search takes place on a meta level above the knowledge objects, and transparent access to these objects is provided. Further, knowledge objects can be found in new contexts, regardless of potential manually-assigned categories in groupware-based organizational

Visual navigation

269

Figure 6. “Infoglut” (top) and a structured semantic link network (bottom)

memories. Therefore, new contextual knowledge can be gathered (Klemke, 2000; Nonaka and Takeuchi, 1995). To sum up, applying topic maps to groupware-based organizational memories provides a huge potential and a lot of advantages for organizational knowledge management. Seen from the perspective of groupware, groupwarebased systems leverage the benefits of topic maps through their basic concepts and paradigms, like distributed database architectures and replication, workflow management, or security and access control mechanisms (Nastansky et al., 2000). A groupware-based environment enables the definition, generation and maintenance of a topic map in a consistent manner, on both a stationary and a mobile workstation. Thus, the distributed management of a topic map is independent of time and location. Differentiated access rights can be defined for a certain set of knowledge objects, single knowledge objects, and any part of a knowledge object, as well as for functionality and presentation elements and design elements like forms and views. These mechanisms enable the realistic adaptation of access structures against the background of complex and real organization models. As an example, some organizational members form an editorial team, which ensures that all published components of a topic map adhere to organizational standards. Another key concept of groupware-based systems, the workflow management, can be used to describe and support the

BPMJ 9,3

270

whole process chain of topic map publishing. A model of an organizational integration of topic maps and supporting workflows is defined in Smolnik and Nastansky (2002). The described concepts, paradigms, and functionalities of groupware-based systems facilitate the team-based and organization-wide use of topic maps. In combination, both the groupware-based systems and the concept of topic maps leverage their mutual potentials. Based on these conceptual approaches, the K-Discovery project defines a framework and architecture for a groupware-based knowledge management system using topic maps. This architecture integrates repositories as well as interfaces to these repositories, and tools that access the interfaces of the infrastructure model, the process model, and the topic map model of an organization. By applying a topic map to the set of completed workflows, it is possible to identify relationships between them as well as tacit and inherent process knowledge. The resulting knowledge can be used in several ways: for example, to group similar workflows and to create pre-defined workflows. Furthermore, a suitable framework for a new kind of knowledge management application – based on the underlying concepts and paradigms of groupware and topic maps – is provided. An example is a repository of workflows, or the support of the analysis of the process structure as well as the organizational structure. A concrete application for distributing workflow process knowledge in organizations, as well as for the semantic associative navigation in process networks, is described in detail in Smolnik et al. (2001) and Huth et al. (2001). In addition, the K-Discovery project introduces an implementation approach that provides a prototypic application and tool framework that implements some of the functionalities of the described architecture, and thus allows a generic and almost automatic generation of topic maps on groupware-based organizational memories by using various technologies like groupware technologies, relational database management systems, Java, XML. The K-Discovery architecture, as well as the implementation approach, are described in detail in Smolnik and Nastansky (2002).

4. Visual navigation of groupware-based topic maps After we have outlined the reasons why the K-Discovery project combines topic maps with groupware-based system environments, we are now going to discuss ways to efficiently use the resulting semantic structures. The original K-Discovery prototype only offered a textual user interface, which was less intuitive to use than a visual browser, because users had to be familiar with basic topic map specific terms and could not overlook larger parts of the structure. Therefore in this chapter, we will present a visual navigation approach to offer the user an intuitive tool to use the groupware-based topic maps in a more explorative way and add means of recognizing specific parts of

the structure by visually highlighting areas that are important to the user. These highlights are referred to as landmarks. 4.1 Requirements of an efficient navigation tool for structured data The use of landmarks (Ark et al., 1998) is a very important factor when considering a navigation space for users, as they support and enhance user orientation in complex hierarchical structures. Landmarks have a strong positive effect on the work efficiency of people that often navigate the structure and thus are familiar with it, as well as on people that navigate the structure for the first time. Additionally, landmarks make users feel acquainted with the structure, and, as a result, emend the user acceptance of the system and the user satisfaction. Therefore, a navigation tool should support landmarks such as color-coded objects, bold text, and particularly thumbnails, graphics, and icons. Additionally, the navigation of the structure should be as simple and intuitive as possible, and the tool should offer rich interaction opportunities to effectively navigate into adjacent and non-adjacent areas of the current area of interest. Especially in the field of explorative collaborative systems it is important that “users become completely absorbed in their task domain” (Shneiderman, 1998). “Focus+context” techniques can help users to become absorbed by the task domain navigation and knowledge browsing. Focus+context is the name of a class of visualization concepts that displays structured data in a way that makes it possible to present as much relevant knowledge objects as possible on the screen at the same time. The area that gets the current users attention is called focus, and the surrounding area is called context. The basic assumption is that users should be able to view focus and context areas at the same time to present an overview of the whole knowledge structure. That enables the user to easily refocus the display into areas that are close to the current focus and thus makes overall navigation more efficient. It can be assumed that the user needs more detailed information in the context area than in the focus area. The reason is that the visual acuity, and thus, the sensual faculty, rapidly decreases outside of the focus area (Glaser, 1994). Therefore, it is not necessary to offer knowledge objects, which are currently of minor importance. Stary (1994) even states that unnecessary knowledge objects burden users, because of the limited capacity of the human nervous system. Consequently, the display area per knowledge object in the context area is less than in the focus area, while the areas are not strictly divided. Moreover, the used area per knowledge object decreases with increasing distance to the focus, proportional to the sensual faculty of the user. As the available display area per object decreases, the amount of information per object has to be decreased as well. Techniques that can be used to decrease the amount of

Visual navigation

271

BPMJ 9,3

272

information are filtering, selective aggregation, highlighting, and distortion (Card et al., 1999). Detailed information and examples of early implementations of focus+context technologies can be found in Furnas (1981) and Spence and Apperley (1982). More current examples are presented in Rao and Card (1994), Lamping et al. (1995), and Pirolli et al. (2000, 2001). To summarize, the most important requirements of a navigation tool for structured visualizations of data are that the navigation should be efficient, the usage should be easily learnable, and it should support landmarks. Focus+context characteristics help to increase the efficiency of navigation. 4.2 Projecting semantic link network structures to trees In the visualization of knowledge structures, an important aspect is the visualization of the relations between knowledge objects. Often, a relation between two objects is indicated by drawing a line or an arc between them. In network structures, such as the WWW or a complex semantic network like a topic map, navigation of visualized network structures is difficult and not very intuitive. This is owing to the fact that all visible objects can have several connections between one another. First, objects can be located outside of the display area and users cannot see the object at the end of the line. Second, using focus+context techniques, objects can be located in the distorted area of the structure and thus connections between them cannot be displayed. Additionally, a clear layout, that guarantees a good oversight over the structure, is hardly impossible, as the data are not hierarchically structured. Figure 7 illustrates that even a simple network with 13 nodes makes it hard to identify which nodes are connected. Therefore, we suggest projecting the semantic link network structure to a hierarchically-organized model such as a tree. In a tree, one object has to be defined as the root object. This could be the best match out of a search using a groupware-based topic map or a defined root object in the company’s intranet

Figure 7. Network with 13 nodes

etc. We assume that a suitable root object can be found in a linked network. Further research has to be conducted to find how the first root can be selected by the user, or automatically offered by the navigation tool. With a root defined, all related objects can be laid out from the point of view of that specific object. The network structure becomes directed that way, as the path through the network starts at the defined root. In Figure 9, the object named “A” is chosen from Figure 8 and the network is strictly projected to a tree. Strict projection means that the children of a node represent all possible connections between the node and other nodes. Figure 9 only shows the first three levels of the projected tree. As connections of objects in an undirected network always represent two directions, each node has one child node that is syntactically the same as the parent node. As this relation is not relevant to a user that navigates within the structure, these relations can be removed from the projected tree (Figure 10). Another method to simplify the resulting projected tree is to filter loops. Paths in a network often contain loops, which let single objects appear several times in the projected tree. If users wish to filter these loops, the filtering should be automatically done by the navigation tool (Figure 11). Landmarks can be used to highlight objects that are syntactically the same. With the structure containing loops that are not filtered, the tree can grow up to an infinite level. Therefore, either the depth of the tree has to be limited to a

Visual navigation

273

Figure 8. Simple network with few relations between nodes

Figure 9. Strict tree projection of Figure 8 from node A, depth 3

BPMJ 9,3

274 Figure 10. Clean tree projection of Figure 8 from node A, depth 5

Figure 11. Full loop filtered tree projection of Figure 8 from node A

certain maximum level or it has to be filtered from loops. Using this concept, the user can easily navigate the knowledge structure using a hierarchy browser. Additionally, tree structures are very well suited for using focus+context techniques. But the projection of the network only visualizes that part of the network that displays the context of the currently defined root node. Therefore, it must be possible to select every node in the network as the root node of the tree. That way, the user can navigate deeper into the structure while keeping the maximum depth of the tree fixed, by selecting any node of interest as the new root node and continue the search for relevant knowledge objects from that point. Landmarks could be used to highlight the descendant nodes of the current root that have been root nodes once before. That way, users can see their entire search path in the network. 4.3 Navigational concepts for trees To create efficient, ergonomic navigational concepts, different teams in the research field of human-computer-interaction (HCI) introduced innovative navigation tool concepts that should increase efficiency of known browsing

concepts such as the Microsoft Windows Explorer. Two of the implementations will be shortly described in the following section and it will be discussed whether the approaches are suitable to be used as navigation tools in semantic link networks and projected semantic link networks. A more comprehensive overview of different visualization concepts can be found in Card et al. (1999) and Gloor (1997).

Visual navigation

275

4.4 Conetree A conetree visualizes hierarchical structures by linked cones in 3D space. All descendants of a node are arranged on the base circle of a cone, while the node itself is the apex of the cone. Each level of the structure has the same height within the conetree. To be able to visualize the whole hierarchy, the size proportions of the conetree are calculated, corresponding to the available display space. A conetree can be laid out horizontally or vertically. Figure 12 shows a simple vertical conetree and a zoom into a horizontal tree. To navigate within the tree, users can rotate the cones to examine the nodes displayed in the background in greater detail. Owing to the nature of conetrees, they are very well suited to visualize and navigate broad hierarchical structures, whereas they are not the appropriate choice if the hierarchies grow very deep. For more information about conetrees see Robertson et al. (1991). 4.5 Hyperbolictree A different approach to tree visualization is the hyperbolictree concept. A hyperbolictree initially displays a tree with its root node at the center. The hierarchical structure is then laid out on a plane in hyperbolic space and mapped on to a circular display region. The circumference of a circle on the hyperbolic plane grows exponentially with its radius, which means that exponentially more space is available with increasing distance from the root node. The tree uses distortion in the context area to reduce the information per

Figure 12. Examples of conetrees

BPMJ 9,3

276

knowledge object ratio. That way it is possible to display an overview of the entire structure on the available display area. A hyperbolictree is very well suited to visualize and efficiently navigate very deep hierarchical structures, whereas very broad structures lead to a situation where the distortion effect in levels close to the root is already too strong, which makes the structure very inconvenient to navigate. Figure 13 shows a hyperbolictree that uses graphical landmarks such as thumbnails and icons, as well as color coding textual information and node background. For more information about the hyperbolictree concept see Lamping et al. (1995) and Pirolli et al. (2000). Both presented concepts are suited to navigate and visualize a projected semantic link network and, as such, are suited to be used as navigational tools for topic maps in groupware-based organizational memories. The use of the conetree is limited to projections where the maximum depth of the tree is restricted to a low value, up to a level of about ten. The conetree would be the better choice for highly intense connected networks. More efficiently is the use of the hyperbolictree concept, because the maximum depth can be set to a very large number, which enables users to search a large part of the knowledge structure without having to refocus the tree to a new root. Additionally, the layout technique of the hyperbolictree offers greater flexibility in the use of graphical landmarks. 4.6 Visualizing of groupware-based topic maps with the hyperbolictree By applying the hyperbolictree concept to groupware-based topic maps using the projection of semantic link network structures, as described in section 4.2, it is possible to provide users with a simple, yet effective visual navigation tool.

Figure 13. Hyperbolictree including landmarks

Navigating a tree structure is much more intuitive and offers a better orientation within the structure than navigating a network. In the tree, nodes can represent topics, occurrences, and topic associations (Figure 14). Landmarks can be used to color code the specific type of topic map element that is represented by the node. In the beginning of a navigation process, the root node is either manually selected by users or is the result of an initial search in the topic map. Further research will be conducted to automatically select an appropriate initial root node for the projection process and as an entry point for user navigation. It can also be thought of as an advanced navigation interface for network structures, that lets users navigate the much more complex structure of the network and then offers the option to examine and explore the neighborhood of a specific node of interest, using the hyperbolictree to navigate in a projection from that node. During the navigation process in a projected tree, users can select new root nodes for the projection manually, while the search path, consisting of former root node objects, is highlighted using landmarks.

Visual navigation

277

5. Conclusions and future areas of research An architecture of a groupware-based knowledge management system based on the ISO standard ISO/IEC 13250 Topic Maps has been presented. The KDiscovery project combines groupware paradigms, topic maps, and visualization concepts, leading to substantial synergies for organizational knowledge management. These synergies leverage their mutual advantages and, therefore, provide benefits for the research fields. From our point of view,

Figure 14. Visualization of a topic map using the hyperbolictree concept

BPMJ 9,3

278

the K-Discovery project provides a considerable framework to identify, visualize, and navigate knowledge structures in groupware-based environments. Therefore, it supports users in finding required knowledge in organizational memories by using topic maps to establish enhanced mechanisms of navigation and linking as well as functionalities for extensive searches and investigations. In addition, the findings are of high practical relevance, as shown in the short case. In fact, there are companies that have shown interest in a solution based on the K-Discovery framework. Prototypes of the core modules have been implemented and can be used for trial purposes. The finalization of the system integration is planned for the very near future. In detail, implementations of an engine to automatically generate groupware-based topic maps and a topic map navigation tool were finished. The conception and implementation of a topic map modeler is about to begin. However, work on the prototypes will be continued to integrate some additional functionalities in the existing core modules like a distributed maintenance of topics and topic associations or personalization mechanisms. Additionally, a graphical visualization and navigation tool based on hyperbolic technologies has been implemented that can be used alternatively instead of the existing topic map navigator, which provides a tabular and textual user interface. This tool facilitates the intuitive and easy navigation to requested knowledge objects. Still to be solved is the definition of the root node in the graphical tree representation of topic maps. So far, topic maps on groupware-based organizational memories have been based on the design of the underlying applications, i.e. fields of forms form a template of a topic map. But knowledge structures contained in texts are not discovered with this approach. This limitation will be addressed in future research, i.e. the research focus will be on ontology aspects and, therefore, on the integration of text parser functionalities into the current prototypes. Eventually, further knowledge management applications on the groupwarebased topic map engine will be developed. Specifically, taxonomy issues in virtual organizations will be addressed. Note 1. Since 2000, K-Discovery is a research project in the context of a dissertation project at the Groupware Competence Center (GCC) of the University of Paderborn.

References Ahmed, K (2000), “Topic maps for repositories”, in Proceedings of the XML Europe 2000, Paris, France. Ark, W., Dryer, D.C., Selker, T. and Zhai, S. (1998), “Landmarks to aid navigation in a graphical user interface”, in Proceedings of Workshop on Personalized and Social Navigation in Information Space, Sweden.

Biezunski, M. and Newcomb, S. R. (2001), “Topic maps and XTM – a manager’s overview” in Proceedings of the Knowledge Technologies 2001 (KT2001), Austin, Texas, USA, 5-7 March 2001.

Visual navigation

Bornschein-Grass, C. (1995), “Einsatzformen und Betriebswirtschaftliche Gestaltungspotentiale von Groupware als Grundlage von (Computer Supported Cooperative Work)”, Deutscher Universita¨ts-Verlag, Wiesbaden. Card, S.K., Mackinlay, J.D. and Shneiderman, B. (1999), Readings in Information Visualization: Using Vision to Think, Morgan Kaufmann Publishers, San Francisco, CA. Davenport, T.H. and Prusak, L. (1998), Working Knowledge – How Organizations Manage What They Know, Harvard Business School Press, Boston, MA. Furnas, G.W. (1981), “The fisheye view: a new look at structured files”, in Card, S.K., Stuart, K., Mackinlay, J.D. and Shneiderman, B. (Eds), Readings in Information Visualization: Using Vision to Think, Morgan Kaufmann Publishers, San Francisco, CA, pp. 312-30. Glaser, W.R. (1994), “Menschliche informationsverarbeitung”, in Eberleh, E., Oberquelle, H. and Oppermann, R. (Eds), Einfu¨hrung in die Software-Ergonomie, Walter de Gruyter, New York, NY, pp. 7-51. Gloor, P.A. (1997), Elements of Hypermedia Design: Techniques for Navigation and Visualization in Cyberspace, Birkha¨user, Boston, MA. ¨ Gu ldenberg, S. (1998), “Wissensmanagement und Wissenscontrolling in lernenden Organisationen: ein systemtheoretischer Ansatz”, DUV: Wirtschaftswissenschaft, Edition O¨sterreichisches Controller Institut, Zugl.: Wien, Wirtschaftsuniv., Diss., 1996 2. durchges. Auflage, DUV, Deutscher Universita¨ts-Verlag GmbH, Wiesbaden. Huth, C., Smolnik, S. and Nastansky, L. (2001), “Applying topic maps to ad hoc workflows for semantic associative navigation in process networks”, in Proceedings of the 7th International Workshop on Groupware (CRIWG’2001), Darmstadt, IEEE Computer Society Press, Los Alamitos CA, Washington, Brussels, Tokio, September. International Organization for Standardization (2000), Information Technology – SGML Applications – Topic Maps, ISO/IEC 13250, ISO, Geneva. Klemke, R. (2000), “Context framework – an open approach to enhance organisational memory systems with context modelling techniques” in Proceedings of the Third International Conference on Practical Aspects of Knowledge Management (PAKM2000), Basel, Switzerland, 30-31 October. Lamping, J., Rao, R. and Pirolli, P. (1995), “A focus+context technique based on hyperbolic geometry for visualizing large hierarchies”, in Proceedings of 1995 Conference on Human Factors in Computing Systems (CHI’95), Denver, CO, May 7-11. Lotus Development Corporation (1995), “Groupware – Communication, Collaboration, Coordination” Cambridge, MA, November. Lotus Development Corporation (2001), “Domino Designer Help [Release 5.06a]” Database, Cambridge, Massachusetts, USA. Moore, G. (2001), “RDF and topic maps – an exercise in convergence”, in Proceedings of the XML Europe 2001, Berlin. Nastansky, L., Bruse, T., Haberstock, P., Huth, C. and Smolnik, S. (2000), “Bu¨roinformationssysteme”, in Fischer, J., Herold, W., Dangelmaier, W., Nastansky, L. and Suhl, L. (Eds), Bausteine der Wirtschaftsinformatik, Erich Schmidt, Berlin. Nonaka, I. and Takeuchi, H. (1995), The Knowledge-Creating Company – How Japanese Companies Create the Dynamics of Innovation, Oxford University Press, New York, NY.

279

BPMJ 9,3

280

Pirolli, P., Card, S. K. and van der Wege, M. (2000), “The effect of information scent on searching information – visualizations of large tree structures”, in Proceedings of the Advanced Visual Interfaces 2000 Conference (AVI2000), May 23-26, Palermo, Italy. Pirolli, P., Card, S. K. and van der Wege, M. (2001), “Visual information foraging in a focus+context visualization”, in Proceedings of the CHI 2001 Conference on Human Factors in Computing Systems (CHI ‘2001), Seattle, WA, 31 March-5 April. Probst, G., Raub, S. and Romhardt, K. (1999), Wissen Managen: Wie Unternehmen ihre Wertvollste Ressource Optimal Nutzen, 3. Aufl, Gabler, Wiesbaden. Rao, R. and Card, S.K. (1994), “The table lens: merging graphical and symbolic representations in an interactive focus+context visualization for tabular information”, in Card, S.K., Mackinlay, J.D. and Shneiderman, B. (Eds), Readings in Information Visualization: Using Vision to Think, Morgan Kaufmann Publishers, San Francisco, CA, pp. 343-9. Rath, H. H. and Pepper, S. (1999), “Topic maps: introduction and allegro”, in Proceedings of the Markup Technologies 99, Philadelphia. Robertson, G.G., Mackinlay, J.D. and Card, S.K. (1991), “Cone trees: animated 3D visualizations of hierarchical information”, in Proceedings of the CHI 1991 Conference on Human Factors in Computing Systems (CHI ‘91), New Orleans, LA. Schliwka, A. (1998), Teamorientiertes, Medienu¨bergreifendes Know-how-Management in Verteilten Umgebungen, Shaker, Aachen. Shneiderman, B. (1998), Designing the User Interface: Strategies for Effective Human-ComputerInteraction, 3rd ed., Addison-Wesley, Reading, MA. Smolnik, S. and Nastansky, L. (2002), “K-Discovery: using topic maps to identify distributed knowledge structures in groupware-based organizational memories”, accepted for: 35th Annual Hawaii International Conference on System Sciences, (CD-ROM), January, Computer Society Press. Smolnik, S., Huth, C. and Nastansky, L. (2001), “Distribution of workflow process knowledge in organizations”, in Proceedings of the 2nd Oldenburger Forum Wissensmanagement, Oldenburg, June. Spence, R. and Apperley, M. (1982), “Data base navigation: an office environment for the professional”, in Card, S.K., Mackinlay, J.D. and Shneiderman, B. (Eds), Readings in Information Visualization: Using Vision to Think, Morgan Kaufmann Publishers, San Francisco, CA, pp. 333-40. Stary, C. (1994), Interaktive Systeme – Software-Entwicklung und Software-Ergonomie, Vieweg, Braunschweig/Wiesbaden. Further reading Inxight Software, Inc. (2001), “Inxight Star Tree”, available at http://startree.inxight.com/, Santa Clara, CA. Pavone AG (2001), “Enterprise Office Help [Release 4.6d, Build 320]” Database, Paderborn. Xerox (2001), available at: www.parc.xerox.com/istl/projects/uir/projects/Information Visualization.html (accessed April 2001).

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

A framework for designing a workflow-based knowledge map

Workflow-based knowledge map

281

Intae Kang, Yongtae Park and Yeongho Kim Department of Industrial Engineering, Seoul National University, Seoul, Korea Keywords Knowledge mapping, Workflow software, Integration, Prototyping Abstract A complete form of knowledge management system comprises both process management and contents management. Process management is concerned with handling activities to generate and utilize knowledge, whereas contents management deals with knowledge contents themselves. Workflow, considered as a core component of process management, is to define and administer business processes automatically. A knowledge map is a representation tool to visualize knowledge sources and relationships among knowledge artifacts. Noting the crucial needs to integrate process management and contents management and recognizing that previous research has paid little attention to this issue, we propose a framework for developing a workflow-based knowledge map. The proposed process-perspective knowledge map takes the structure of processes and tasks defined in workflow into account. We then materialize the framework by developing a prototype and applying it to the car seat design process of the automobile industry. The integration represents an exploratory effort to combine process management and contents management.

1. Introduction Although theoretical underpinning is yet fragile, knowledge management (KM) has attracted increasing attention from academicians and practitioners alike (Harvard Business School, 1998; Kakabadse et al., 2001). The spectrum of KM is so ample, encompassing both organizational aspects and technical factors. Among others, however, knowledge management systems (KMS) serves as an indispensable tool for KM in that it facilitates knowledge generation, dissemination, and utilization (Liebowitz, 1999). Broadly, KMS is composed of a process management module to administer knowledge activities to generate and utilize knowledge and a contents management module which deals with knowledge contents, input and output of knowledge activities. Furthermore, on the one hand, workflow is considered as a core component of process management since it defines and administers business processes automatically. On the other hand, the knowledge map accounts for the cornerstone of contents management, since it functions as the representation tool to visualize knowledge sources and identify relationships among knowledge artifacts. Thus far, a paucity of studies has dealt with the knowledge map. Although the utility has long been recognized, research output of the knowledge map has largely remained conceptual and metaphysical. In particular, there has been no attempt to integrate workflow and the knowledge map or, in other words, to

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 281-294 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477894

BPMJ 9,3

282

propose a workflow-based knowledge map. Voluminous research on workflow has been carried out independently of the knowledge map and past research on the knowledge map has been conducted separately from workflow. However, the majority of knowledge that needs to be managed is generated and utilized in-house, along the workflow of organization. Therefore, it is meaningful and useful to build the process-perspective knowledge map based on the structure of processes and tasks defined in the workflow. To this end, we will propose a framework for designing a knowledge map that takes the workflow-related factors into account, which is called a workflow-based knowledge map. In the framework, the set of necessary artifacts of knowledge can be collected and structured into a knowledge map based on the relationships among knowledge users, tasks, and knowledge described in the workflow. We will also present a prototype of a Web-based knowledge map designed and implemented in accordance with the proposed framework. In doing that, the car seat design process of the automobile industry is employed to demonstrate the detailed procedure. 2. Background 2.1 Research on workflow Workflow is a tool to define and administer business processes automatically. It, in turn, becomes the core engine of the workflow management system (WFMS) that electronically operates and controls the business process. The usefulness of WFMS is highlighted as the amount of knowledge becomes intractably large, the business units are geographically decentralized but more closely networked, and the importance of collaboration among individual workers is emphasized. The majority of past studies on workflow mainly have focused on information technologies to implement real-time collaboration or process automation. Therefore, the primary research objectives are owing to the improvement of scalability and stability of WFMS or definition and representation of processes and tasks. To illustrate a few, Kumar and Zhao (1999) proposed a framework to implement dynamic routing and operational control mechanisms composed of workflow control tables, sequence constraints, and event-based workflow management rules to improve flexibility, reusability, scalability of process modeling. Workflow management coalition (WfMC, 2001) has established the standards of reference model to design workflow and specifications of data and documents to realize interoperability between workflow systems. Casati et al. (1996) proposed automatic derivation techniques of active rules that form conceptual workflow specification and provided a collection of caseindependent rules of workflow enactment to give operational semantics. This study concentrated on the architecture and the database schema to describe a workflow of an organization.

2.2 Research on knowledge map Workflow-based The terms knowledge map or knowledge cartography are defined as a visual knowledge map architecture of knowledge domain that enables us to examine the knowledge on a global scale and from different perspectives (Eppler, 2001). Although it is a relatively new label, the knowledge map is considered as a core component of KM. It is the basic tool for knowledge users to retrieve necessary knowledge and 283 to analyze the relationship among knowledge. It also helps knowledge managers grasp the status of knowledge and plan knowledge development strategies. So far, the majority of research on the knowledge map has been done from the point of knowledge managers and/or system developers, rather than of knowledge users. Therefore, past studies focused on filtering of knowledge, representation of knowledge structure and implementation of KM strategies at the corporate level. Specifically, Gomez et al. (2000) suggested a knowledge map that can be used at the stage of holistic testing in KMS implementation. They defined the knowledge map as “a type of mental diagram by means of which complex ideas can be easily and quickly set out in a logical order” and proposed a framework to present a knowledge map through relationships of artifacts of knowledge and related attributes. Gaines and Shaw (1995) proposed a concept mapping tool that facilitates visualizing the relationships among actor, activity, objective and material in dispersed learning communities. Godbout (1999) presented a model to filter knowledge based on such crucial factors as actor relevance factor, technical relevance factor, authority relevance factor, fidelity factor, and scientific acceptability factor. Eppler (2001) classified knowledge maps into five types like knowledge source map, knowledge asset map, knowledge structure map, knowledge application map, and knowledge development map. 2.3 Integration of workflow and knowledge map As pointed out before and reviewed above, previous work on workflow and the knowledge map has been made separately from one another and the integration of the two has been the missing link in KMS design. Furthermore, the majority of past research on the knowledge map has been developed for knowledge managers, rather than for knowledge users. In this paper, we suggest a framework to build a workflow-based knowledge map that attempts to integrate workflow and knowledge map. As depicted in Figure 1, since workflow is the backbone of process management and the knowledge map is the building block of contents management, the integration represents an exploratory effort to combine process management and contents management. 3. Framework of workflow-based knowledge map In the corporate setting, knowledge is created and re-used during the execution of business processes, and thus our system supports KM, going through the business process and dealing with the specific knowledge that is relevant for a

BPMJ 9,3

284

Figure 1. Composition of knowledge management

Figure 2. Framework for workflow-based knowledge map

particular task on the process. The overall framework of a workflow-based knowledge map is presented in Figure 2. The figure shows that the construction of a knowledge map consists of two steps: knowledge collection and knowledge structuring. First, in the step of knowledge collection, knowledge in the KR is filtered based on the knowledge users and related tasks. While building a knowledge map, it is critical to provide the knowledge relevant to the knowledge user’s specific task. To achieve such user and task relevancy in knowledge provision, we need to have information on user and task attributes. This information can be obtained from a process definition if the knowledge activities are performed in a WFMS. Our knowledge collection takes the relationship of user-task and task-knowledge into account, which will be elaborated in Section 4.1. Second, the filtered knowledge is structured and presented using vertical relationship and horizontal relationship of knowledge. We use the term vertical relationship to mean the hierarchical organization of knowledge artifacts, that is, a knowledge artifact includes several more detailed ones, each of which can

be similarly decomposed further. The horizontal relationship indicates the Workflow-based input-output relationship of knowledge. A knowledge item may need to cite a knowledge map certain set of knowledge artifacts that have been produced previously. This implies that there is a sequence in knowledge production. Since in our approach knowledge is associated with tasks in a process, the knowledge production sequence can also be identified from the workflow definition. This issue is 285 detailed in Section 4.2. 4. Detailed procedure of framework 4.1 Knowledge collection The amount of knowledge in a corporate is usually too large for an individual knowledge user to manage with ease. Therefore, it is often impossible for knowledge users to capture and control the whole set of knowledge. In fact, while carrying out a specific task, the user does not have to know all the knowledge and, in most cases, requires only a part of the knowledge contained in the corporate KR. This means that it is important to provide the user with the knowledge relevant to the specific task in charge. To improve the performance of knowledge retrieval and the effectiveness of knowledge utilization, we propose a method of knowledge filtering based on its relationship with prospective users and related tasks specified in workflow. In our approach, the knowledge filtering is carried out in two steps. First, user-task relationships are identified from task specifications in workflow and user profiles in the human resource database. This information is used to confine the subsequent knowledge search within the scope of user-relevant knowledge. Second, our system captures task-related knowledge. A task under consideration needs knowledge generated in other tasks that are associated with it, in a certain type. The type of task association is defined in a workflow definition. Based on this association as well as the above user-task relationships, our system retrieves and classifies task-related knowledge artifacts from the KR. Figure 3 shows the conceptual scheme of the two steps of knowledge collection. For user-task relationship, we use user’s access rights to a task. User’s rights means the authority that a user can execute and monitor a certain task. Generally, the access rights are established depending on the department that a

Figure 3. Steps of knowledge collection

BPMJ 9,3

286

user works for, and on the role that the user, or department, is expected to play. The information on the user, department, and role is available from the user profiles in the human resource database, and the relationship between the information and tasks is materialized in workflow definitions. Using the usertask relationship, we are able to sort out the knowledge that is authorized and directly relevant to a user. 4.2 Task-related knowledge Often, a certain knowledge activity is related with many other activities. This means that a knowledge worker needs to refer the results of other tasks that are somehow related with the worker’s current task. Such knowledge is called taskrelated knowledge in this research. The task-related knowledge can be captured from the relationship between the current task and its related tasks. Since the relationship is also defined in a workflow definition, we can collect the task-related knowledge based on the definition. Task-related knowledge is classified into the following four categories: input knowledge, output knowledge, applied knowledge, and parallel knowledge. This classification can be identified from the task relationship in the workflow definition. Figure 4 exhibits the categories of task-related knowledge and their relations, and each category is further described as follows. . Input knowledge: knowledge that is generated in the tasks that precede the current task. A knowledge worker of the current task builds up his/her own knowledge on top of the input knowledge. . Output knowledge: knowledge produced in the current task. Such knowledge is used directly in the current task because it was created while performing the same task in the past. When the task itself is deployed further into a more detailed sub-process, those that built in this sub-process also form the output knowledge (this aspect will be considered later in structuring of knowledge). . Applied knowledge: knowledge created in the tasks following the current task. Referring to this type of knowledge, a knowledge worker of the current task may be able to predict the effects of the current task results.

Figure 4. Task-related knowledge

.

Parallel knowledge: knowledge generated in the tasks that are not directly Workflow-based related with the current task, but executed in parallel to it. It is not knowledge map unusual that knowledge workers performing those tasks need to collaborate with one another and to exchange the parallel knowledge to coordinate their task results.

Our approach to collect relevant knowledge using the relationships of task and knowledge provides a basis for structuring and presenting a knowledge map. Since the knowledge artifacts collected by our approach are different depending on users and tasks, it allows us to construct a personalized knowledge map. This personalized knowledge map is expected to improve the performance of knowledge retrieval and reduce the efforts required for KM. The issue of structuring and presenting a knowledge map is addressed in the next section. 4.3 Structuring and presentation of knowledge map The knowledge collected in the previous section needs to be presented to users in a convenient form. The way of organizing and visualizing knowledge exerts a considerable impact on the efficiency of knowledge retrieval and utilization. We propose the structure of a knowledge map that corresponds to the structure of workflow processes. In order to do this, we adopt the nested process model suggested by Kim et al. (2000). (1) Nested process model. As the scale of corporate becomes larger and as the resulting workflow processes gets more complicated, a one-dimensional and flat process model is not sufficient to analyze and represent such a complex model. Kim et al. (2000) suggested a nested process model in their study on WFMS. The nested process model allows the user to break the composite business process into several sub-processes and provides the structure for hierarchically arranging them. Therefore, it makes it easy for the user to grasp the relationship between processes and tasks. Owing to the advantages of the nested process model, it is also used in the knowledge map structuring. In our map, knowledge artifacts are interrelated vertically and horizontally, which are comparable to the relationships of sub-processes and tasks that have produced the knowledge artifacts. Such relationships can be identified in the nested process model definitions. A nested process model is composed of primitive tasks and nesting tasks. A primitive task can not be broken down into smaller elements, and a nesting task is deployed into a sub-process. A sub-process provides more detailed specifications for the corresponding nesting task. In the model, a task can become both a parent of a lower-level process and a child of another process at a higher-level. For example, in Figure 5, the root task, T0, is at the uppermost level. This represents the whole process as detailed in process P1. In this process, tasks TU and TUs are nesting tasks, so that each of them is again deployed to processes P2 and P3.

287

BPMJ 9,3

288

Figure 5. Nested process model

In structuring the knowledge map, we use the relationship between nesting tasks and nested processes to capture vertical relationships of knowledge artifacts generated in those tasks and processes. We also grasp the horizontal relationships of knowledge artifacts with the sequence of tasks and processes in nested process model. (2) Structuring the knowledge map. Collected knowledge is organized and visualized in a knowledge map using the nested process definitions described in the previous section. A knowledge map in general shows how various types of knowledge are interrelated with one another. In our map, knowledge artifacts are interrelated vertically and horizontally, which are comparable to the relationships of sub-processes and tasks that have produced the knowledge artifacts. For example, consider the knowledge structure in Figure 6, where the knowledge is organized from the perspective of task T. First, the vertical relationship is presented in three tiers: nesting knowledge, par knowledge and nested knowledge. The nesting knowledge is the knowledge generated at the upper level process; the par knowledge that at the process containing the current task; and the nested knowledge at the lower level process. In our example, task T is contained in process P, so that it becomes the par level process. P is abstracted as a task T0 in the upper level process P0. Task T is further deployed in process P1. Using this vertical relationship of processes, we are able to organize the knowledge artifacts in a hierarchical form. That is, an

Workflow-based knowledge map

289

Figure 6. Structure of knowledge map

artifact of nesting knowledge may include or sum up those of par knowledge. As such, an artifact of par knowledge may also be detailed or elaborated into more detailed nested knowledge. The horizontal relationship is the order of task execution, which can also be obtained from a process definition. The relationship considers only tasks in the par process. Depending on the precedence relations of tasks, there are three types of horizontal relationships: preceding, parallel, and succeeding tasks. In our example, the knowledge created in task T becomes output knowledge for task T and T0, applied knowledge for task T1p, and T0p, and input knowledge to task T1s and T0s, and parallel knowledge to task T1r. Such horizontal relationships can be found in every level of process. The importance of a knowledge map is guiding users to find out required knowledge artifacts easily and quickly. With the above three-tier structure of knowledge map, a user can navigate through the knowledge space in the context of business processes. A user can capture not only the hierarchical structure, but also the direct relationships of knowledge, such as the inputoutput of knowledge artifacts. In addition, but more importantly, our approach allows us to create different knowledge maps depending on users and tasks, even though it uses the same scheme. That is, another knowledge user carrying out a different task has a personalized knowledge map in the same structure. 5. Workflow-based knowledge management system We have designed and developed a prototype system for the proposed workflowbased knowledge map. The system is called workflow-based knowledge management system (W-KMS). In this paper, an operation scenario of the system is explained, and captured images of user interfaces are demonstrated. 5.1 Operation scenario The major modules of W-KMS include user interface (UI), knowledge repository (KR), workflow management (WM), and users profile management

BPMJ 9,3

290

Figure 7. An operation scenario of W-KMS

(UPM). The UI displays a personalized knowledge map where users interact with W-KMS. While navigating the map, a user can browse and retrieve relevant knowledge from KR. The KR stores the knowledge generated in a corporate. With WM, a user can define a business process and execute it. Finally, the UPM is to administer users, departments and their roles. Figure 7 illustrates a typical operation scenario represented in a sequence diagram of unified modeling language (D’souza, 1999; Eriksson and Penker, 2000). Figure 7 shows the interactions among W-KMS modules involved in the scenario. The interactions proceed in the following sequence. (1) A knowledge user logs into the system, and then the user identification (ID) and password are sent to the UPM. (2) The UPM verifies the ID and password, and gets the user’s department and role. (3) This user information is sent to WM. (4) WM retrieves a list of tasks that the user executes and monitors, and sends the task-related information to the UI. (5) The user selects a task from the list of tasks on the UI. (6) The UI sends the selected task ID to WM. (7) WM identifies the tasks that are related with the selected task. The identified tasks are sent to KR. (8) WM returns the relationships of the tasks identified in (7) to the UI. (9) The KR collects the knowledge artifacts that have been produced by the tasks identified in (7), which are transferred to the UI.

(10) The UI creates a knowledge map for the knowledge artifacts, and Workflow-based presents it to the user. knowledge map (11) The user chooses a knowledge artifact on the UI. (12) The UI sends the ID of the selected knowledge artifact to the KR. (13) The KR retrieves the selected knowledge artifact and transmits it to the 291 UI, so that the user can use it. 5.2 Prototype system Our framework of a workflow-based knowledge map has been applied to a realworld case. We have investigated a process of designing car seats for a local company in the automobile industry. The seat design process in the automobile industry is very complex in that it provides constraints and at the same time is constrained by many other tasks. The huge amount of knowledge that the designer has to refer to also causes difficulties. There has been a great need for KMS that can provide a personalized and well-structured knowledge map. Figure 8 shows some of the knowledge activities involved in the car seat design process. The figure shows the seat design activity and its related tasks. It also shows the upper level process and the detailed process of seat design. Our prototype system provides the designer with an environment where the user can search required knowledge considering the task relationships. In the knowledge map, the knowledge artifacts are structured with a nested process definition. Knowledge in vertical relationships can be identified in the upper (body, interior, and integration) or lower (shape, material, and color design) level processes,

Figure 8. Example of knowledge retrieval using a workflow-based knowledge map

BPMJ 9,3

292

whereas that in horizontal relationships within the process containing the seat design task (interior space, controller, and illumination design). W-KMS is a Web-based system implemented on the Internet information service (IIS) and active server page (ASP), and thus a user can utilize the system with any Web-browser. Figure 9 shows a UI for a project manager in the seat design department. The user can browse the status of the entire project and the list of tasks to execute and monitor. When one of the tasks is clicked, the system constructs and displays a knowledge map for the task. The resulting map is presented in Figure 10. Using the UI, a user can freely navigate the knowledge space of seat design. Notice that the map has upper, par, and lower levels, and each level encloses its associated process representation. The right-hand side of the UI displays the specific knowledge created by a knowledge activity. For example, in Figure 10, the user currently selects the activity of body design, and thus those knowledge artifacts regarding the activity are provided for the user. 6. Conclusions KMS, as a complete form, necessitates both process management module and contents management module. In that regard, the integration of workflow and knowledge map is crucial, since the former is the backbone of process management and the latter is the building block of contents management. In this research, we proposed a framework to integrate knowledge map and workflow, ultimately to develop a workflow-based knowledge map. The framework was then materialized by developing a prototype of a workflowbased knowledge map. The workflow-based knowledge map filters required knowledge artifacts using workflow definitions, and creates a vertical and horizontal structure for the collected knowledge based on nested process definitions. The integration can be considered an extension of conventional

Figure 9. User interface of task selection

Workflow-based knowledge map

293

Figure 10. User interface of knowledge map and knowledge artifacts

knowledge mapping approaches, to provide a basis for integrating process management and contents management. Yet, the current research is merely exploratory and therefore subject to some limitations. In particular, we believe that the knowledge map proposed in the current study and other conventional maps are not mutually exclusive, but complementary. In that regard, we may suggest the following future research themes. First, we considered only the process-perspective knowledge map. However, organization-perspective knowledge maps, found in other traditional knowledge mapping approaches, are still useful. Thus, the performance of knowledge retrieval and the effectiveness of knowledge utilization may be enhanced if the two types of knowledge maps can be linked together. Second, the linkage with contents-perspective maps deserves attention. The contents-perspective map intends to classify knowledge artifacts according to the subject matters contained in the knowledge sources. It assists knowledge users in identifying the relationships among knowledge items based on the contents of knowledge itself, independent of related tasks. It also facilitates filtering knowledge by the fidelity and reliability of knowledge. Such supportive methods as data mining in general, and text mining in particular, may be employed to this end. Finally, a more systematic and automatic tool needs to be proposed in developing a knowledge map since the manual approach becomes hopeless as the knowledge map becomes intractably large and complex.

BPMJ 9,3

294

References Casati, F., Ceri. S., Pernici, B. and Pozzi, G. (1996), “Deriving active rules for workflow enactment”, Proceedings of 7th International Conference on Database and Expert Systems Applications, Lecture Notes in Computer Science, Springer-Verlag, New York, NY, pp. 94-110. D’souza, D. (1999), Objects, Components and Frameworks with UML, Addison Wesley, Reading, MA. Eppler, M. (2001), “Making knowledge visible through intranet knowledge maps: concepts, elements, cases”, Proceedings of 34th Hawaii International Conference on System Sciences, IEEE, Piscataway, NJ, pp. 1530-9. Eriksson, H. and Penker, M. (2000), Business Modeling with UML – Business Patterns at Work, John Wiley & Sons, New York, NY. Gaines, B. and Shaw, M. (1995), “Collaboration through concept maps” Proceedings of CSCL 95 Computer Supported Cooperative Learning, Lawrence Erlbaum, Hillsdale, NJ, pp. 135-8. Godbout, A. (1999), “Filtering knowledge: changing information into knowledge assets”, Journal of Systemic Knowledge Management, available at: www.scoap.com/ki/articles/godbout/ godbout03.htm Gomez, A., Moreno, A., Pazos, J. and Sierra-Alonso, A. (2000), “Knowledge maps: an essential technique for conceptualization”, Data and Knowledge Engineering, Vol. 33 No. 2, pp. 169-90. Harvard Business School (1998), Harvard Business Review on Knowledge Management, Harvard Business School Publishing, Boston, MA. Kakabadse, N., Kouzmin, A. and Kakabadse, A. (2001), ““From tacit knowledge to knowledge management: leveraging invisible assets””, Knowledge and Process Management, Vol. 8 No. 3, pp. 137-54. Kim, Y., Kang, S. and Kim, D. (2000), “WW-FLOW: Web-based workflow management with runtime encapsulation”, IEEE Internet Computing, Vol. 4 No. 3, pp. 55-64. Kumar, A. and Zhao, J. (1999), “Dynamic routing and operational controls in workflow management systems”, Management Science, Vol. 45 No. 2, pp. 253-72. Liebowitz, J. (1999), Knowledge Management Handbook, CRC Press, New York, NY. WfMC (2001), “The workflow coalition specification – Interface 1: process definition interchange process model”, Workflow Management Coalition, available at: www.wfms.org

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

Do we practise what we preach? Are knowledge management systems in practice truly reflective of knowledge management systems in theory?

Knowledge management systems 295

Nilmini Wickramasinghe James J. Nance College of Business Administration, Cleveland State University, Cleveland, Ohio, USA Keywords Knowledge management, Knowledge management systems, Knowledge workers, Organization, Memory, Competitive advantage Abstract Knowledge management systems are predominant in both theory and practice. However, are the same systems discussed in theory actualized in practice? By comparing and contrasting knowledge management systems in theory and practice, this paper demonstrates that they are indeed dissimilar. In theory, they have both subjective and objective components. In practice, only the objective component appears to be actualized; hence, these systems in practice are essentially organizational memory systems at best and not knowledge management systems at all. By unravelling the mystique of knowledge management systems, this paper exposes a fundamental anomaly. Further, an apparent void currently in practice is highlighted; namely, the lack of the subjective component of knowledge management systems in practice. They are being heralded as key systems that are vital for organizations to survive and thrive in the intense competitive environment of the information age. Surely then, a system that in practice supports not only the objective component, but also the subjective component of knowledge management, would indeed be a truly powerful system.

1. Introduction We are not only in a new millennium, but also a new era. A variety of terms such as the post-industrial era (Huber, 1990), the information age (Shapiro and Verian, 1999), the third wave (Hope and Hope, 1997) or the knowledge society (Drucker, 1999) are being used to describe this epoch. However, whichever term one subscribes to, most are agreed that one of the key defining and unifying themes of this period is knowledge management. Knowledge management is a key approach to solving current problems such as competitiveness and the need to innovate, which is faced by businesses today. The premise for the need for knowledge management is based on a paradigm shift in the business environment where knowledge is central to organizational performance (Drucker, 1993). This macro-level paradigm shift This project was funded by a staff development grant from the Faculty of Economics and Commerce, The University of Melbourne. The author is indebted to the generous support from the faculty, as well as the support from the members of the organizations who participated in this study. A shorter version of this paper has been accepted at the 35th Hawaii International Conference on System Science, January 2002, and will appear in the proceedings for that conference.

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 295-316 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477902

BPMJ 9,3

296

also has significant implications on the micro-level processes of assimilation and implementation of knowledge management concepts and techniques (Swan et al., 1999) i.e. the knowledge management systems (KMS) that are in place. This study focuses on these KMS. Some of the pioneers and early adopters of KMS have been the large consulting companies; launching into these systems as early as the late 1980s. By analyzing three of these companies in detail, some of the mystique surrounding KMS is unravelled. In so doing a fundamental anomaly is exposed; namely, that KMS in practice and KMS in theory are indeed dissimilar. In fact, KMS, in practice, are arguably more like organizational memory systems than OMS. 2. Knowledge management 2.1 Definition of knowledge management Knowledge management deals with the process of creating value from an organization’s intangible assets (Wigg, 1993). It is an amalgamation of concepts borrowed from the artificial intelligence/knowledge-based systems, software engineering, business process re-engineering (BPR), human resources management, and organizational behaviour (Liebowitz, 1999). In essence then, knowledge management not only involves the production of information, but also the capture of data at the source, the transmission and analysis of this data, as well as the communication of information based on, or derived from, the data, to those who can act on it (Davenport and Prusak, 1998). 2.2 The need for knowledge management Sustainable competitive advantage is dependent on building and exploiting core competencies. In order to sustain competitive advantage, resources which are idiosyncratic (and thus scarce) and difficult to transfer or replicate are required (Grant, 1991). A knowledge-based view of the firm identifies knowledge as the organizational asset that enables sustainable competitive advantage, especially in hyper competitive environments (Alavi, 1999; Davenport and Prusak, 1998; Zack, 1999). This is attributed to the fact that barriers exist regarding the transfer and replication of knowledge (Alavi, 1999), thus making knowledge and knowledge management of strategic significance (Kanter, 1999). Since knowledge management addresses the generation, representation, storage, transfer and transformation of knowledge (Hedlund, 1994), the knowledge architecture is designed to capture knowledge and thereby enable the knowledge management processes to take place. Underlying the knowledge architecture is the recognition of the binary nature of knowledge; namely its objective and subjective components. Knowledge can exist as an object, in essentially two forms – explicit or factual knowledge – and tacit or “know how” (Polyani, 1958, 1966). It is well established that while both types of knowledge are important, tacit knowledge is more difficult to identify and thus

manage (Nonaka, 1994, 1991). Further, objective knowledge can be located at various levels, e.g. the individual, group or organization (Hedlund, 1994). Of equal importance, though perhaps less well defined, knowledge also has a subjective component and can be viewed as an ongoing phenomenon, being shaped by social practices of communities (Boland and Tenkasi, 1995). The objective elements of knowledge can be thought of as primarily having an impact on process. Underpinning such a perspective is a Lockean/Leibnitzian standpoint (Malhotra, 2000). While the subjective elements typically impact on innovation by supporting divergent meanings consistent with Hegelian/Kantian modes of inquiry, both effective and efficient processes, as well as the function of supporting and fostering innovation, are key concerns of knowledge management in theory. The knowledge architecture thus recognises these two different, yet key aspects of knowledge and provides the blueprints for an all-encompassing knowledge management system. Clearly then, the knowledge architecture is defining a KMS that supports both objective and subjective attributes of knowledge. The pivotal function underlined by the knowledge architecture is the flow of knowledge. The flow of knowledge is fundamentally enabled (or not) by the KMS (see Figure 1).

Knowledge management systems 297

Figure 1. The knowledge architecture

BPMJ 9,3

298

2.3 Knowledge management systems Knowledge management in its broadest application refers to how a firm acquires, stores and applies its own intellectual capital. From a theoretical standpoint, KMS refer to the information systems adopted and designed, which efficiently and effectively leverage the collective experience and knowledge of employees to support information processing needs, as well as enabling and facilitating sense-making activities of knowledge workers; i.e. systems that actualize the knowledge architecture. In practice, knowledge management has been fuelled by the rapid development of technologies; in fact the emergence of Web-based technologies has enabled and fuelled the widespread awareness and adoption of KMS in much the same way as it has had an impact on e-commerce. KMS in practice are touted as being able to support and enable the knowledge management process within the respective organizations. By examining the KMS in practice, it is possible not only to shed light on the knowledge management processes adopted, but also facilitate an understanding of the role of knowledge management in the organization. More importantly, it is possible to uncover any anomalies between theory and practice. What follows are case findings from three consulting companies – approximately ten years after they respectively and independently adopted a knowledge management perspective and implemented their first KMS. After an overall industry analysis, a synopsis of each case is discussed in turn, before a summary of key findings is presented. The data were gathered using standard techniques for conducting qualitative multiple case study research (Yin, 1994; Kvale, 1996; Boyatzis, 1998). The Appendix details the specific steps taken. 3. KMS in practice The management consulting industry is far from homogeneous and it can be thought of as having several different segments of players. Three major segments are as follows: (1) Generalist/strategy, which consist of companies such as McKinsey and Co., Booz-Allen and Hamilton and The Boston Consulting Group. (2) The big five CPA firms, which consist of companies such as Ernst & Young[1], Price Waterhouse & Coopers, Deloitte & Touche and KPMG. (3) Information technology (IT), consisting of large IT companies such as IBM Consulting Group or American Management Systems[2]. The three case studies discussed in this paper are drawn from within these segments of consulting companies[3]. Three key industry trends having significant impact, particularly on the CPA consulting companies, since the late 1980s have been growth, services integration and globalization. Historically, these firms positioned themselves as generalists, but since the late 1980s early 1990s clients have been interested

in seeking advice from consulting firms with a wide array of corporate expertise. Clients realised that such expertise and knowledge cannot reside with a few consultants, but rather in a collective asset of the firm that results from a wide range of projects in different industries and geographic regions. Hence, not only were clients requiring quality consultants, but also deep industry and/or function-specific expertise. Contemporaneously with changing client needs and expectations, these companies were experiencing mergers and globalization pressures. The combination of these distinct, yet not necessarily unrelated needs, was that these firms had to adapt. Specifically, strategies were required to address these new business needs and systems were needed to support the new business processes. Knowledge management was identified to be the strategy to adopt in order to provide these firms with a new vision for managing their internal knowledge resources. KMS were identified as the systems to implement in order to support and enable the required business processes that would ensure the realisation of this new strategic direction. The investments by these firms in knowledge management have been significant indeed. Some initial costs are as much as 12 million with about 10 per cent of revenues annually required for building and managing the knowledge capital. Such investments demonstrate a belief on the part of these firms that knowledge management is key to enable these organizations to maintain their competitive advantage. Let us now turn to the three specific case examples. 3.1 Case 1 Knowledge management was identified to be a key component for this consulting company as early as 1992. By 1997 this organization had made significant progress in instituting the infrastructure and processes required to make knowledge management the backbone of its strategic vision. This firm’s KMS was considered pioneering among its competitors. Consisting of a knowledge web that is essentially a re-collection of content and associated infrastructure, the system is complemented and supported by three internal knowledge-oriented centres. These centres are all located in the USA and are responsible for managing the development, capture, storage and dissemination of the firm’s knowledge globally. When the company decided to follow a knowledge management focussed strategy it soon realised that there was no precedent for knowledge sharing within the firm, nor was there consistency in the technological infrastructure to allow for electronic sharing between its consultants. Thus, an integration of the various practices into a unified consulting practice was required and established. The design and implementation of a firm-wide KMS was dependent on an appropriate technological infrastructure as well as a supporting culture and organizational processes. The following quotation from

Knowledge management systems 299

BPMJ 9,3

300

an interview with a consultant in the knowledge management area of this firm clearly articulates the vision for the KMS within this organization. The vision was to create a system, which depends on both centralized and decentralized resources, in which the national office would design the system and provide the necessary infrastructure/process/technical support while the field would contribute the content. The system would be designed with the remote worker in mind who required access to as many resources as possible. It was believed that the system would improve efficiency and quality of service thus enabling sustainable competitive advantage and would strengthen relationships with target accounts as well as enable segments that previously could not afford consulting services and opportunity.

A sharing culture is critical for the success of knowledge management within any organization. This firm realised that it needed to develop and foster such a sharing culture if its knowledge management endeavours were to succeed. A variety of carrot-and-stick methods such as financial rewards and bonuses, evaluations and guiding principals, were adopted. The cultural change has been slow, which was expected; however, the firm is confident that its techniques will continue to be instrumental in facilitating the required change. Support for knowledge management is truly top down. In addition, this firm has developed a two-tiered approach to its knowledge management, i.e. it has both an internal and external component to its knowledge management focus. The internal component refers to how it as an organization adopts and deploys knowledge management within the firm, while the external component relates to how it advises and sells knowledge management to its clients. It believes that by supporting these two components it has now become a truly knowledge-based business. 3.2 Case 2 In the early 1990s this firm’s strategy could be summarised as wanting to increase its market share and improve productivity, to enable it to become number one among the then Big 6, now Big 5, companies. This strategy involved several dimensions including the following: market leadership and innovation, recruiting and retention, and increasing net income per partner. Market leadership referred to targeting and focussing on specific services, products and market segments, with the greatest returns; prioritising the firm’s resources; and exploiting areas of competitive differentiation to drive market expansion in those segments. Market innovation involved becoming a thought leader among the professional services firms, as well as in the industries and markets it served. It was generally believed within this organization that in the 1990s and beyond customers would demand high quality, customer focused solutions and services at highly competitive prices. Finally, an important aspect of the growth strategy was connected with increasing net income per partner. This was envisaged to occur via expansion of market share and via cost reductions, elimination of duplication of effort, and the streamlining of internal processes.

In 1989 the concept of knowledge management first appeared in this firm. A proposal for a KMS was put forward to facilitate the acquisition and sharing of expertise and experience among the firm’s professionals. However, in the late 1980s, unlike some of its competitors, this firm was not ready to embrace the concept of knowledge management. The partners were not convinced that the large investment necessary for a successful KMS was justifiable at this time. The concept re-emerged around 1991. The underlying concepts of enhancing, leveraging and distributing the firm’s knowledge assets were the same; however, in the time period between 1989-1991 the technologies had greatly changed. This enabled this firm to utilise newer technologies and its capabilities within its planning of KMS. The system introduced had two major components: (1) The conference model, which was designed to support ongoing discussion and exchange of ideas among the firm’s professionals. (2) The library model, which was to be a collection of knowledge about the firm, its clients and their businesses and industries – in essence a repository of lessons learnt. By 1995 this KMS was globally available. The KMS initially used first class groupware, but was very shortly updated to exploit Web-based technologies. Within the industry this firm is considered to have been a relatively late entrant into the knowledge management abyss. The delay in entering knowledge management, relative to some of its competitors, while having some disadvantages, did enable this firm to move more quickly and easily to Web-based architecture and tools, thereby exploiting and incorporating some of the latest technologies within its KMS. In addition, it was easier to justify the incorporation of knowledge management in its strategy, as the concept was now more widely established and accepted throughout the industry. As one partner noted in the following interview extract. Knowledge management tools are critical to the professional services firms. What do we sell? We sell advice. . .The most valuable thing we offer is the collective, institutional knowledge of our firm to solve business problems. . .Our relatively late entrance has enabled us to leapfrog some of the others and go straight into newer technologies.

3.3 Case 3 This firm, very early on in the knowledge management euphoria, realised that: Knowledge capital is our most valuable asset and it drives our organization. It’s what we sell and what we must continue to protect and perfect. Our people should diligently find new ways to share and reuse information and deploy it around the world [quoted from internal documents].

The firm’s knowledge sharing system began in 1991 with the introduction of Lotus Notes and is today a vast electronic knowledge repository, housing more

Knowledge management systems 301

BPMJ 9,3

302

than 100,000 documents used by its personnel throughout the world. Knowledge management has been part of a deliberate strategy to bring to bear the cumulative knowledge and experience of the firm onto client engagements, to enable each client to receive the best thinking of the entire firm, and not just that of the particular engagement team. The task of providing the clients with the best thinking is becoming rather complex for this firm, owing to its sheer size. Its knowledge management effort has generated a very large repository of documents, and a growing body of contributors and users to be managed. One of the critical challenges now is how to bring the best and most relevant information, insight, expertise and approach to consultants working on client engagements, and in particular, how to increase the application of relevant knowledge throughout the firm. This firm is organized by the industries it serves and by the competencies necessary to provide services. Each industry group then has several segments; for example, within the financial services industry their exist segments such as insurance, banking and investments. Consultants are typically designated a geographic region; however, the identification with an industry group is very strong and tends to lead to the formation of communities. In addition, less formal communities of practices also exist. The existence and establishment of these communities has served to foster a sharing culture across this firm that was indeed an asset when it incorporated knowledge management. The current KMS in this firm consists of over 3,000 databases that support several needs, including: articulating core methodologies, stating the industry vision, facilitating major business processes and best practices, providing discussion databases on various topics to consultants in the field, storing comments and lessons learnt, as well as being repositories of, or pointers to, external information of various types, including technology research and customized news alerts on business topics. Clearly, this represents a large and complex store of information and knowledge. This firm has made significant investment in Lotus Notes. One of the limitations, particularly with earlier versions of Lotus Notes, is connected with its searching capabilities; while Lotus Notes does possess some search tools, these are not sufficient for a system with the size and complexity of case 3. The KMS has grown exponentially for this organization and a vast repository of information and knowledge has accumulated. Overall, the use of the system has not been as successful as anticipated. Incentives and promotions are tied to incorporating and supporting the knowledge management perspective and there is general evidence that a strong sharing culture exists; however, the inefficiencies connected with search techniques appear to be impacting on the enthusiastic use of this knowledge management system.

3.4 Summary of case study findings From the data collected on the KMS adopted by the respective consulting companies under investigation, many comparisons and contrasts can be made. These are detailed in Table I in terms of the following categories: environment, strategy, technology, culture, structure, knowledge management perspectives, key questions regarding the KMS and current issues facing all these firms. The following serves to highlight some of the significant issues. While each system is reflective, to a large extent, of its organization’s conceptualization of knowledge management, as well as its stage and scope of development of its KMS, by comparing these respective systems in practice, many useful generalizations can be made which enable a better understanding of the direction of KMS to date. 3.5 Environment By the late 1980s the consulting industry had become very competitive and most firms had experienced several mergers and/or acquisitions prior to launching into knowledge management. In fact, the global focus that had developed for all these firms, making the respective organizations not only much larger, but also more complex, was a key driver for pushing these companies into knowledge management. Another challenge at this time was connected to the recruitment and retention of staff. All the consulting companies experienced low staff retention. Hence, there was a strong need to integrate the firms’ practices and preserve the knowledge that was lost when consultants left the firm. 3.6 Strategy All three companies articulated knowledge management in their strategy as a means for enabling sustainable competitive advantage. It appears that the launch into knowledge management has been a strategic necessity. Clearly, without knowledge management these firms would not be enjoying the current market structure and positions they have today. Now, in order to enjoy continuing success from their respective investments in KMS, these firms must address some of the existing concerns and deficiencies of the current systems. These include fostering a sharing culture, ensuring relevant, appropriate material is stored and accessed from the system and searching times are kept as low as possible. Knowledge management continues to be an integral component of these firms’ strategies. The critical issue faced by these three firms is to capture the full power of knowledge management and ensure that their respective KMS are supporting their knowledge management strategies. 3.7 Technology The technologies adopted to make up the respective KMSs were very similar. Of the three firms, case 2, which entered knowledge management later, had less of an investment in groupware such as Lotus Notes and consequently found the

Knowledge management systems 303

Rapid development in IT providing appropriate infrastructure to support KMS Some components of KMS Webbased, Lotus Notes, client/server technology, multimedia features, databases Variety of offerings of KMS tailored to various customer segments Emphasis on consistency of technology infrastructure

Originally a wide area network inter connecting a number of 64K parallel processors with desktop workstations Updated to support two models of knowledge sharing: conference and library Initially used FirstClass – cross-platform intranet groupware product, client/server technology, databases, multimedia features Quickly embraced Web technologies

(continued)

Started with e-mail only Significant investment in Lotus Notes Lotus Notes, gradual shift to incorporate Web-based technologies, databases, client server technology, multimedia features Other features include navigator features, electronic yellow pages, front page and document finder

1991 started knowledge management focus Interweaving industry capabilities and competence areas to drive business solutions to clients Wanted to encourage innovation “let every flower bloom” idea

Technology

Future plans developed in early 1990s Knowledge management was thought to be critical to professional service firms Knowledge management first discussed in 1989, however, thought to be difficult to justify and not appropriate at the time Began global knowledge management initiatives by 1996

Knowledge management was identified in 1992 as a key component of its strategic plan Five mega processes involved in the strategic plan of which knowledge management was one Wanted to adopt a very proactive approach to adoption and implementation of knowledge management

Case 3

Strategy

Table I. Case summary 1995 management consulting industry was estimated to be 40 billion dollar industry world wide and 50 per cent of this was in the USA Cases 1, 2 and 3 part of Big 6 Three industry trends included growth, integration of services, and globalization Recruitment and retention of people was proving to be challenging Market innovation was becoming essential for consulting companies Clients were expecting value added products and services at competitive prices Mergers and acquisitions that had occurred meant there was a need to consolidated knowledge, techniques, standards and approaches across the globally dispersed offices of these firms

Case 2

304

Environment

Case 1

BPMJ 9,3

Considered a pioneer in its ability to leverage its own intellectual capital While culture is collegial – historical performance evaluations had discouraged information sharing Lacks a strong sharing culture to support knowledge management Has three separate key centres in the USA to support distinct aspects all relating to knowledge, including; methodologies and technology tools, research and analysis as well as knowledge-based administration and finally research initiatives and executive roundtables Emphasis on team-based learning Organised around lines of business and geographic areas Knowledge management: describes how a firm acquires, stores, and applies its own intellectual capital Initiated incentives for KM culture Developed internal and external knowledge management focus Continual emphasis on innovative ways to enhance current knowledge management initiatives Support for knowledge management clearly comes from the top Knowledge management part of strategic vision

Culture

Structure

Knowledge management perspective

Case 1

Case 3

Always had a strong sharing culture Had strengths in its design, development and implementation of computer systems – made adoption of KMS easy Had a history of developing “methods” and “techniques”, hence much support for KM methods and many existing methods and techniques incorporate KM perspectives Re-structured along lines of business a Organized by industries it serves and change from a geographic/functional competencies necessary to provide structure services Established knowledge centres and new Set up role and responsibilities for KM, expanded organizational roles e.g. 200 people full time KM managers Focusing on their human resource asset Emphasis on cross functional teams such Fusion between internal processes, systems as communicates of practice and and technologies was critical competency groups The three dimensional structure of geographic areas, lines of business and functions enabled the matrix search strategy for their KMS Does not have a clear definition of Knowledge management considered its knowledge management that permeates most valuable asset driving the throughout he organization organization Wanted to create a knowledge management Knowledge management here involves the environment that tied together legacy promoting of ongoing business sources systems and intranet, WWW, data through the systematic acquisition, warehousing, document managing, synthesis, sharing and use of information, networking applications and network insights and experiences management Knowledge management part of business Knowledge management was part of vision strategy statement and support for knowledge In developing KM key responsibilities management is definitely from the top were identified to include ensuring leading edge currency of our knowledge capital, keep the KMS demand driven rather than supply driven (continued)

Diverse culture – owing to many partnership mergers Of the three cases most individualistic – traditionally performance was based/rewarded on individual performance Very minimal sense of a sharing culture

Case 2

Knowledge management systems 305

Table I.

How can we bring the best most relevant information, insights, expertise and approach to consultants working on client engagements? Improve usability of the system? Determine and quantify the value of knowledge management activities? Improve quality and relevance of the knowledge in the KMS?

How much should these firms continue to invest in knowledge management Can knowledge management be a sustainable competitive advantage? Does the corporate culture support knowledge management? Is the current KMS now too big and overwhelming? How should they leverage existing expertise in building knowledge management systems? How should they achieve worldwide co-ordination of their KMS? Looking ahead – what next is the next step – where do they go from here?

How fast and effective will the knowledge management evolution be? Three critical needs include: large scale culture change; a mechanism for knowledge and intellectual asset assessment; and robust technology and systems implementation – are these needs being satisfied?

Case 3

Current issues

Table I. How could effectiveness of KMS be measured? Did the current system provide the basis for a sustainable competitive advantage? Was KMS consistent with the strategic direction for the next five to ten years?

Case 2

306

Key questions about KMS

Case 1

BPMJ 9,3

transition to Web-based technologies faster and easier. Cases 1 and 3 have more investment in Lotus Notes, which also has a drawback with respect to searching. Ease of searching and finding relevant and appropriate material is particularly an issue in case 3. Nevertheless, all three firms are now rapidly incorporating and expanding on their Web-based technologies. Case 1 has segmented its knowledge management approach depending on type of client (small, big). Thus, it has more separate technology components to support the various segments of its KMS, such as an online computer system that clients can access for self-help, as well as the large-scale KMS used by consultant teams on client engagement. This segmenting appears to be facilitating the knowledge management process for this firm. 3.8 Culture As is well documented in the literature, organizational culture plays a critical role in the acceptance and adoption of systems, the modification of their use once installed and overall success of the system in the facilitating and conducting of tasks (Robey and Azevedo, 1994; Coombs et al., 1992; Orlikowsky, 1992). The literature pertaining to knowledge management highlights the importance of a sharing culture in order to support and foster a knowledge management focus (Alavi, 1999; Davenport and Prusak, 1998; Zack, 1999). Case 3 has always fostered a team culture and this was an advantage in adopting knowledge management, even so, still increasing and sustaining use of its KMS is a key issue. Thus, we can see that a sharing culture is a necessary but not sufficient condition for the satisfactory adoption of a KMS. Case 2 appears to have had the most problems with trying to foster and develop a knowledge-sharing culture and one that would support its total knowledge management approach. It is quite possible that since this case was the last to implement its KMS, it has not enjoyed the window of time required to fully develop and foster a sharing knowledge management climate. Perhaps a more sharing culture will be evidenced after the incentives to encourage sharing behaviour have been well established. To date, case 1 appears to be moving in the right direction regarding developing a sharing culture. This is mainly attributed to its correct mix of carrot-and-stick strategies. There is plenty of scope to improve the sharing culture and central to this is that the current approach must be maintained. 3.9 Structure Three centres located in different geographic locations in the USA supported key areas of knowledge management in case 1. In addition, globally across the organization there is a strong emphasis on team-based learning. Case 2 re-structured along a three-dimensional matrix structure involving its lines of business, functional areas and geographic areas, believing that this is the appropriate way to construct its knowledge bases in an efficient and effective

Knowledge management systems 307

BPMJ 9,3

308

manner. Case 3 developed its KMS around its industries and competencies structures that were already in existence. Obviously the respective KMS are structured to fit the nuances and unique needs of the respective firms; however, it does appear that a connection between structure and searching exists. In particular, it would seem that the more complex the structuring, the easier it is to search the KMS. This was particularly apparent on comparing search times with case 2 (a complex three-dimensional structure) and case 3 (a simpler basic structure). 3.10 View of knowledge management Two of the three cases had clear definitions of knowledge management that were articulated throughout the respective organization. In fact case 1 has gone further and is now moving to what it calls being a knowledge-based business, as opposed to just adopting knowledge management strategies in 1992. This firm has two clear and distinct sections for knowledge management within its organization, an internal section, which is involved with managing its own knowledge, and an external component, which is involved with providing consulting expertise to its clients regarding knowledge management. Case 3 also has a knowledge management for clients component, where it focuses on developing and implementing KMS for its clients, but this is not in the same magnitude as case 1. Case 2 has a definition of knowledge management; however, other areas of the organization did not freely articulate it. Once again this is attributed to the fact that this firm was later in adopting knowledge management and it is anticipated that in time the diffusion of knowledge management throughout this firm will occur. 3.11 Key issues for the future Measuring the effectiveness and contribution of knowledge management is a key concern for all these organizations. Another area of concern that is prevalent is how to ensure quality of content, and that the most appropriate relevant information is always accessed. These issues are clearly critical both for the overall success of knowledge management in these respective firms, as well as to ensure that these firms can confidently deliver the best advice to their clients. The solutions though, are truly challenging and not obvious. 3.12 Knowledge management system’s support of the knowledge architecture The respective KMS definitely supported the objective component of knowledge, i.e. they function as repositories, storing both tacit and explicit knowledge and even go so far as to support the knowledge spiral (Nonaka, 1991) of turning explicit knowledge to tacit and vice versa. However, these systems do not support the subjective component of knowledge well, if at all. In fact, all these firms largely ignore the subjective component of knowledge. Given that only the objective component of knowledge is truly supported by these KMS, these systems essentially function as an organizational memory

system, as opposed to an actualization of a KMS, as described by theory. This point is elaborated on in section 4.1. 4. Discussion From the case data (section 3.0) several major areas are identified as being important to understanding the mystique behind KMS. These will now be discussed in turn. 4.1 Role of KMS in practice The KMS in the three consulting companies clearly and unequivocally support the objective component of knowledge management. However, it was not apparent that the subjective component was supported by these systems at all; i.e. there was no support for the sense-making aspects of knowledge management based on Hegelian/Kantian models (Malhotra, 2000). It is argued that this major deficiency of KMS in practice means that the systems in practice create and sustain an organizational memory, since they enable the storing and retrieving of information and knowledge, as well as providing directories of who knows what, and where they might be located, and even enable connections to these people. A KMS in practice should be an actualization of KMS from theory and support all elements of the knowledge architecture. Both organizational memory and knowledge management are concepts that dominate much of the recent literature. Furthermore, they have essentially the same main goal of trying to help organizations to cope with the following three changes in the business world: . an increasing complexity, dynamics, fragmentation and decentralization of knowledge or knowledge development; . an increasing complexity of organizational structures and the permanent need to change these structures; and . an increasing amount of non-traditional data to be managed (Maier and Lehner, 2000). However, organizational memory and knowledge management differ in that they suggest different methods and strategies to deal with these complexities and changes. Organizational memory is concerned with the storing and subsequent accessing and replenishing of an organizations “know-how”, which is recorded in documents or in its people (Maier and Lehner, 2000). Organizational memory systems, then, enable this storing of knowledge as well as facilitating access to it and the people who possess it. However, organizational memory systems do not support sense-making knowledge creation activities. Clearly, there is an intersection or overlap between organizational memory and knowledge management and thus it is to be expected that organizational

Knowledge management systems 309

BPMJ 9,3

310

memory systems and knowledge management systems should share some similarities. Knowledge management is concerned with the management of the organization’s knowledge and does focus on the stages of storing, retrieving and distributing knowledge. However, a key component of knowledge management not addressed in the construct of organizational memory is the subjective aspect. Table II serves to highlight many of the key dimensions and identify which are unique to KMS and OMS respectively, and which are shared between KMS and OMS. We can summarise the needs driving the three case studies to adopt knowledge management as follows: . to reduce complexities in the environment caused by globalization; . to reduce complexities as a result of mergers; . to address loss of knowledge owing to poor staff retention; and . to ensure that all consultants could bring to any client engagement the full power of the organization. Thus, both the concepts of knowledge management and organizational memory as discussed in theory are relevant to the situations faced by these respective consulting companies. However, if we focus on the knowledge management systems adopted and implemented by the respective consulting firms, we can see that they actualize an organizational memory system, rather than a knowledge management system. The knowledge management system in each of the cases enables gathering and storing of collective knowledge (both tacit and explicit), collective memories, even collective intelligence, as well as providing a resource for locating and contacting anyone, anywhere within the organization activities consistent with an organizational memory system. Further, the knowledge management system in each of the consulting companies does not enable discourse and sense-making to take place as in a Hegelian/Kantian model (Malhotra, 2000). This is indeed a significant anomaly and has implications for both theory and practice. 4.2 Importance of culture structure dimensions Irrespective of whether the KMS in practice in the respective case studies was an organizational memory system rather than a true KMS, the moderating impacts of culture and structure cannot be understated. In all three cases culture was clearly a significant barrier that had to be overcome, in order for the specific knowledge management approach to be embraced. To date none of the cases have been totally successful in fostering a sharing culture. Structure too has been a factor in moderating the impact of the respective knowledge management systems. Each case handled the structure component differently – one reorganized totally, one did not change its structure and the other case study adopted a half-and-half solution. The impact of structure clearly affects many aspects of the knowledge management, including

Subjective and objective

Impacts on increasing complexity, dynamics fragmentation and centralization of knowledge or knowledge development Technologies utilised Highly dependent on Internet-based technologies

Unique

Shared

Shared

Tacit and explicit knowledge

Unique

OMS

Highly dependent on Internet-based technologies

Addresses both past and future since it is Addresses only past since it focuses on focusing on problem solving memory – essentially a collective corporate memory Supports both tacit and explicit Supports the storage of only explicit knowledge knowledge – any tacit knowledge that is stored is first transformed into explicit knowledge via the knowledge spiral, before it can be stored Should in theory support both objective Should only support objective aspects and subjective aspects since it is forming a corporate repository or collective memory By enabling the sharing of knowledge By enabling the storing of knowledge and throughout the organization thus maintaining a corporate memory

Time

Unique

KMS

Dimensions

Unique attributes

Knowledge management systems 311

Table II. KMS vs OMS

BPMJ 9,3

312

searching and getting the correct information, both relevant and up-to-date, to the correct people. 4.3 Impact on process and innovation As has already been mentioned, the respective knowledge management system, while being most successful in impacting on process did not appear to have great impact on innovation, only on making processes more effective and efficient. This finding is consistent with much of the recent literature in knowledge management which states that knowledge management has forgotten the people side (Swan et al., 1999). The findings from this study strongly suggest that the knowledge management systems in practice are deficient with respect to one of the key components of knowledge – the subjective component. Without the subjective component the knowledge management systems are not able to achieve their full potential or be truly successful. Furthermore, without supporting the subjective component it is unlikely that KMS will be able to support the people factor. 5. Conclusion This study set out to compare KMS in practice with KMS from a substantive, theoretical standpoint. Three cases were chosen from large consulting firms as examples of successful, well-established KMS. The data collected on these respective systems showed that a sharing culture is necessary. Furthermore, the data also suggest that the structuring of KMS impacts on key activities such as searching and retrieving relevant data. Current issues of concern in practice include how to justify or demonstrate quantitatively the benefits of these KMS and also how to store better, more relevant knowledge, and enable it to be accessed in a timely fashion. On comparing the KMS in practice to theory, a key unexpected finding surfaced. From a theoretical standpoint arguments were presented to show that a KMS should support the knowledge architecture and the two faces of knowledge; namely, knowledge as both a subject and an object. However, the KMS did not support the subjective aspect of knowledge, and by not doing so, they actually function as an organizational memory system, not a KMS. This fundamental anomaly is indeed a significant finding and has implications for both practice and theory. Such a lack of the subjective aspect of KMS in practice should form the focus of future research. There appear to be several possible strategies to address this apparent void with KMS in practice. One such strategy might be to incorporate greater artificial intelligence capabilities and perhaps build an intra and interorganizational expert network as a key component of the KMS. This strategy should enable the filtering, interpretation and utilization of various elements of corporate knowledge for problem solving. Another strategy might be to change the organizational culture and encourage knowledge sharing practices among

communities of knowledge workers and thereby utilise the KMS to facilitate and enable such behaviour. This latter approach is the route Kaiser Permanente has decided to adopt in using MARS (the automated medical record system designed and developed by this organization), the KMS for this organization (Wickramasinghe and Mills, 2001). Clearly, this is an important area that requires further confirmatory empirical studies to determine which strategy or strategies is most suitable. It is likely that from such research no one strategy would always be appropriate; however, given that we are in the information age and knowledge management is a critical aspect of managing, such research is both important and beneficial. Furthermore, organizations that implement KMS, which do indeed support the subjective aspects of knowledge management, are the organizations that are likely to enjoy sustainable competitive advantage. Notes 1. Before merger with Cap Gemini. 2. Source: 1996 Consulting News published by Kennedy Information. 3. Refer to Appendix for details on the research methodology adopted. References Alavi, M (1999), “Managing organisational knowledge”, working paper. Boland, R. and Tenkasi, R. (1995), “Perspective making perspective taking”, Organisation Science, Vol. 6, pp. 350-72. Borghoff, U. and Pareschi, R. (1998), Information Technology for Knowledge Management, Springer-Verlag, Berlin. Boyatzis, R. (1998), Transforming Qualitative Information Thematic Analysis and Code Development, Sage Publications, Thousand Oaks, CA. Coombs, R., Knight, D. and Willmott, H. (1992), “Culture, control and competition: towards a conceptual framework for the study of information technology in organisations”, Organisation Studies, Vol. 13 No. 1, pp. 51-72. Davenport, T. and Prusak, L. (1998), Working Knowledge, Harvard Business School Press, Boston, MA. Drucker, P. (1993), Post-Capitalist Society, Harper Collins, New York, NY. Drucker, P. (1999), “Beyond the information revolution”, The Atlantic Monthly, October, pp. 47-57. Grant, R. (1991), “The resource-based theory of competitive advantage: implications for strategy formulation”, California Management Review, Vol. 33 No. 3, pp. 114-35. Hedlund, G. (1994), “A model of knowledge management and the N-form corporation”, Strategic Management Journal, Vol. 15, pp. 73-90. Hope, J. and Hope, T. (1997), Competing in the Third Wave, Harvard Business School Press, Boston, MA. Huber, G. (1990), “A theory of the effects of advanced information technologies on organisational design, intelligence, and decision making”, Academy of Management Review, Vol. 15 No. 1, pp. 47-71.

Knowledge management systems 313

BPMJ 9,3

314

Kanter, J. (1999), “Knowledge management practically speaking”, Information Systems Management, Fall, pp. 7-15. Kvale, S. (1996), Interviews An Introduction to Qualitative Research Interviewing, Sage, Thousand Oaks, CA. Liebowittz, J. (1999), Knowledge Management Handbook, CRC Press, London. Maier, R. and Lehner, F. (2000), “Perspectives on knowledge management systems theoretical framework and design of an empirical study” in Proceedings of 8 th European Conference on Information Systems (ECIS). Malhotra, Y. (2000), “Knowledge management and new organisational forms”, in Malhotra, Y. (Ed.), Knowledge Management and Virtual Organisations, Idea Group Publishing, Hershey, PA. Miles, M. and Huberman, A. (1994), Qualitative Data Analysis: An Expanded Sourcebook, Sage, London. Nonaka, I. (1991), “The knowledge creating company”, in Harvard Business Review on Knowledge Management, Harvard Business School Press, Boston, MA. Nonaka, I. (1994), “A dynamic theory of organisational knowledge creation”, Organisation Science, Vol. 5, pp. 14-37. Orlikowsky, W. (1992), “The duality of technology: rethinking the concept of technology in organisations”, Organisational Science, Vol. 2 No. 3, pp. 398-427. Polyani, M. (1958), Personal Knowledge: Towards a Post-Critical Philosophy, The University Press Chicago, Chicago, IL. Polyani, M. (1966), The Tacit Dimension, Routledge & Kegan Paul, London. Robey, D. and Azevedo, A. (1994), “Cultural analysis of the organisational consequences of information technology”, Accounting Management and Information Technology, Vol. 4 No. 1, pp. 22-38. Shapiro, C. and Verian, H. (1999), Information Rules, Harvard Business School Press, Boston, MA. Swan, J., Scarbrough, H. and Preston, J. (1999), “Knowledge management – the next fad to forget people?” Proceedings of the 7th European Conference in Information Systems. Wickramasinghe, N. and Mills, G. (2001), “MARS: the electronic medical record system the core of the Kaiser galaxy”, International Journal Healthcare Technology Management, Vol. 3 No. 5/6, pp. 406-23. Wigg, K. (1993), Knowledge Management Foundations, Schema Press, Arlington, VA. Yin, R. (1994), Case Study Research: Design and Methods, 2nd ed., Sage Publications, Newbury Park, CA. Zack, M. (1999), Knowledge and Strategy, Butterworth Heinemann, Boston, MA.

Further reading Alavi, M. and Leidner, D. (1999), “Knowledge management systems: issues, challenges and benefits”, Communications of the Association for Information Systems, Vol. 1 No. 5. Beckman, T. (1999), “The current state of knowledge management”, in Liebowitz, J. (Ed.), Knowledge Management Handbook, CRC Press, New York, NY. Evans, P. and Wurster, T. (1999), Blown to Bits, Harvard Business School Press, Boston, MA. Shtub, A. (1999), Enterprise Resource Planning, Kluwer Academic Publishing, Boston, MA.

Wickramasinghe, N. (1999), “Knowledge management systems: the hidden and the manifest”, Grant proposal submitted to The Faculty of Economics & Commerce, The University of Melbourne, Australia. Wilcox, L. (1997), “Knowledge-based systems as an integrating process”, in Liebowitz, J. and Wilcox, L. (Eds), Knowledge Management and its Integrative Elements, CRC Press, New York, NY.

Knowledge management systems 315

Appendix. Thematic analysis details The qualitative analysis performed in this study comprised analyzing data gathered from interviews, observations and documents, conducting thematic analysis on this data and the construction of site summaries. These various stages are now discussed in turn. Interviews: The average length of an interview was 90 minutes. Copious hand written notes were taken (on average five to six pages for face-to-face, three pages for phone) during both the phone and face-to-face interviews. It was not possible to tape interviews, as the organizations did not wish to have taped interviews conducted. In addition, follow-up interviews were conducted, as required, to validate accuracy of information obtained. This was an iterative process; initial interview data were collected and evaluated for their appropriateness, relevance and richness of information before returning to conduct further interviews to generate more detailed information and fill in gaps. Finally, in order to maximise reliability, Yin (1994) recommends the use of an interview protocol. Therefore, the following interview protocol was developed and then followed when conducting the interviews of the key actors. The interview protocol is a guide that is developed a priori and specifies the issues and underlying themes for which information is being sought to explain (Kvale, 1996). The use of the interview protocol ensures that a systematic approach is applied to each case and all interviews. From the interview protocol, specific semi-structured questions were developed. The interview protocol consists of the following six parts: (1) addresses general background issues for the organizations; (2) addresses the actual decision to adopt knowledge management; (3) addresses the specific KMS in place; (4) addresses the use of the KMS and any problems or complications associated with such use; (5) identify future challenges; and (6) establish whether knowledge management systems are likely to be a fad or the future for this organization. Interview protocol The following six stages were used to gather information: (1) Identify the chosen consulting company and describe, including a brief history, its structure, as well as its goals and objectives. This serves to facilitate a complete understanding of the consulting company. In addition, it enables a sense of the culture within this organization to be understood and how this consulting company fits into the industry as a whole. (2) Understand why the company launched into knowledge management. This involves understanding what were the key drivers and whether the move was supported and/or initiated from the top. (3) Understand the KMS in place and its role. This involves understanding the components of the technologies that make up the KMS in the respective organizations. How does the

BPMJ 9,3

316

system work and what are the objectives of the system, and does it meet these objectives? It is also important to discover whether or not the system supports the objective information processing aspects of knowledge management, and if so, how, and if the system supports the sense-making aspects of knowledge management, and if so, how. (4) Elucidate problems/complications with the use of the KMS. In order to do this it is important to establish how the system is used and how effective it is in performing desired tasks. The anticipated outcome from these inquiries is that it will be possible to obtain data to show whether the different organizations that have KMS show similar problems and whether they are better or less effective at handling specific tasks. (5) Identify future challenges. It is expected that each of the organizations will have key components that they believe must now be addressed regarding their KMS. (6) Establish whether the KMS is seen as the future or a fad. This would be decided by the investment of the company in KMS and knowledge management as well as the level of top management/partner support – thus more support from management and high investment would indicate that KMS is viewed by the organization as the future, not a fad. After gathering information using the above interview protocol coupled with interview questions, data gathered from site visits, observations and documents, it was possible to conduct the thematic analysis. Observations: Observations of the systems were made at each of the sites; i.e. case 1, case 2 and case 3. Thematic codings: All interviews were coded. Pattern coding techniques were used (Boyatzis, 1998; Kvale, 1996; Miles and Huberman, 1994). In particular, it was necessary to go through three iterations to refine and distil the relevant information pertaining to these themes. The first iteration categorized phrases in the interview notes as they related to the knowledge management systems in place. The second iteration coded within broad themes, subthemes that were evident such as culture and structure. The final iteration reduced these themes to a final set of subthemes that were mutually exclusive, but collectively exhaustive. Thematic coding was performed on a case by case basis. At the conclusion of coding all cases another iteration was conducted to cross-check cases for occurrences of themes. Where possible, supporting evidence from documents gathered and observations made were combined to support findings. Site summaries: Site summaries were formed for each site. Once again these included the three case studies. These summaries were created by performing data reduction strategies, described by Boyatzis (1998), Kvale (1996) and Miles and Huberman (1994), on the interview data, and combining this information with documents obtained that described the various organizations, as well as observations made. The focus for the site summaries centred on the knowledge management systems in place.

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

Knowledge conversion with information technology of Korean companies

Knowledge conversion

317

Hyun-Soo Lee and Yung-Ho Suh Department of Business Administration, Kyunghee University, Seoul, Korea Keywords Knowledge management, Knowledge-based organizations, Information technology Abstract This research aims at finding out how the efforts of Korean enterprises to convert data and information into knowledge are related to the use of information technology. It also aims to come up with plans to efficiently introduce external knowledge by applying information technology.This research was conducted focusing on the cases of enterprises that are carrying out knowledge management by applying information technology and the result shows that Korean enterprises place high value on the application of information technology in introducing knowledge management. However, the use of information technology is not common to all enterprises in four kinds of knowledge conversion. For an organization to efficiently acquire and apply valuable knowledge, the boundary of an enterprise should be extended, and for this, the enterprise should be changed into an extended knowledge enterprise, which is founded on the trust and compensation of all the interested parties. In an extended knowledge enterprise, its first consideration is to acquire knowledge that can efficiently manage and control the interests of customers, suppliers, public institutions, and competitors that exist in the surroundings and to acquire the core knowledge that can make those people and institutions participate. In addition, the organization should possess knowledge base and information communication technology that can smoothly convert data and information into knowledge, by efficiently connecting them.

Introduction It is self-evident that knowledge will be the source of competitiveness in the twenty-first century, and the government and the academic world have recognized knowledge-based management of enterprise as a means to enhance national competitiveness. However, countless enterprises have failed in knowledge management (KM). Enterprises that planned to introduce KM had difficulty conducting proper knowledge, as they faced the barrier of the application, acquisition, and extension of knowledge and the source of knowledge, which was acquired from the introduction method. However, a lot of effort and the development of information technology (IT) are gradually solving the problems mentioned above and are extending the source of knowledge acquisition to a broader extent, focusing on their application. Enterprises that are introducing KM are making efforts to obtain, integrate, share and apply knowledge more efficiently, based on the capability of themselves. The problem for the enterprises has been in acquiring and integrating internal knowledge. This is because the enterprises thought of internal

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 317-336 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477911

BPMJ 9,3

318

knowledge as a stepping-stone to completing knowledge-based management internally, and then extending it to the exterior. However, this kind of sequence does not apply to all the enterprises, it is dependent on the characteristics, size, and the organization culture, because to some enterprises external knowledge is more important than internal knowledge. Existing knowledge has a limit in its acquisition and application, knowledge can lose its value unless the source of new knowledge flows in. An organization has to convert the knowledge of individuals into that of an organization, and it has to create new knowledge by sharing external knowledge. For this, present IT and new IT should be introduced and applied. IT converts certain information into knowledge that is essential to management. Also, companies are able to set up a proper business strategy for competitive advantage through a knowledge-based system, using best practice information that contains necessary information about competitors (O’Leary and Selfridge, 2000). This strategy became an indispensable element in today’s global world. This research is aimed at exploring the IT Korean companies utilize for KM and knowledge conversion (KC) and providing some suggestions for the best use of IT by working out how Korean enterprises’ efforts for KC are related to the use of IT. For this purpose, we review the literature and interview KM team managers in some Korean companies that implement knowledge-based management. Based on surveys and the literature reviews, we present an expanded KC model based on Nonaka’s four processes of KC. The rest of the paper is organized as follows: Section 2 introduces usage of IT in KC. In Section 3, we survey utilization of knowledge management systems (KMS) in Korean companies. Section 4 presents external knowledge in KM. Section 5 proposes extended knowledge enterprise. The paper concludes with a summary of the outcomes, limitation, contribution of this research and future research. The usage of IT in terms of four types of KC Knowledge is not something that we simply share and collect. In order to utilize knowledge, it is important to properly convert knowledge into something necessary to the business environment a certain company is situated in. This is because each business should consider what kind of knowledge assets would be essential to them among four knowledge processes (socialization, externalization, combination and internalization) when they set up a business process or business model. KC refers to a set of actions related to the transformation of knowledge from one form to another. It consists of two types of knowledge: tacit knowledge and explicit knowledge. KC involves a transformation process of these two types of

knowledge into the same or different kind of knowledge. Tacit knowledge is highly individual, hard to formulate into a certain form, and hard to share and communicate with other people (Nonaka and Konno, 1998). Therefore, tacit knowledge can be shared only through a high level of interactive communication (Zack, 1999). Tacit knowledge has its roots in individual experience, thinking and values. Explicit knowledge is expressed through letters or numbers and can be shared in the form of a manual, specification, formulation and data: it is formally and systematically transferable among individuals (Nonaka and Konno, 1998). Explicit knowledge is easy to standardize and deliver; therefore, it increasingly plays a bigger role in a company or organization. Except for the ones mentioned above, knowledge can be divided into many different kinds, according to its types and ranges. But this paper is mainly focused on tacit knowledge and explicit knowledge. This research will focus on explaining how to utilize IT suitable to Nonaka’s four processes of KC through the case study of Korean companies. IT for socialization Socialization refers to a process where certain individuals or groups share their own experience to create and deliver their tacit knowledge such as spiritual models and skills. It is a process of transforming one tacit knowledge into another (Nonaka, 1998). A good example of socialization can be found in a craftsman’s passing down of his skill on to others, or a relationship between the skilled and non-skilled in the apprenticeship system of the past. Today, socialization or passing down of knowledge occurs in the form of on-the-job training (OJT), that is implemented in the modern military or companies. However, time and cost problems can arise when companies deliver tacit knowledge to their employees, because the delivery of tacit knowledge is realized through a face-to-face instruction or personal experience. The reason is because tacit knowledge is a personalized knowledge based on individual experience, attitudes, values and behaviors, and therefore, is not easy to identify, express and formulate (Gore and Gore, 1999). But, despite time and cost problems, many companies place a great deal of emphasis on tacit knowledge, since they know it is an essential key to creating new knowledge. In an effort to effectively exchange tacit knowledge through its groupware “Arisam”, Samsung SDS Co. opened a new cyber space called “forum” where an individual or group can exchange their views and knowledge. This groupware includes “cyber consulting” through which a person can acquire professional knowledge to create new knowledge. At the time of writing, “forum” is open only to employees but is expected to be open to outside users from the year 2001 so that competitors and business partners can participate.

Knowledge conversion

319

BPMJ 9,3

320

LG-Electronic Data Systems thinks that the effective exchange of tacit knowledge is very important and set up a KM system on a Web-based intranet. The company has a different system for tacit knowledge and explicit knowledge. For tacit knowledge, it opened up an integrated Q&A session in “smart community” where a person or organization can ask a question and advise one another on various matters. To further promote the exchange of tacit knowledge, it also set up “k-village” for experts and community groups, and “knowledge mall” to help exchange knowledge acquired at work, various ideas and know-how. As we can see from the above examples, the usage of IT made it easy for Korean companies to pass down and share knowledge among employees and, thus, transform personal knowledge into knowledge that becomes a part of corporate assets. As a result, companies are able to effectively create and convert knowledge by removing time and space restrictions that existed in past practices such as a group meeting, brainstorming and management by walking around. Acquiring knowledge from customers is also important, but direct knowledge acquisition from customers is not feasible in reality. Therefore, IT such as the Internet homepage and e-mail is very effective in gathering and utilizing customers’ needs. In a comparison between the usage of the IT of Microsoft and that of Netscape, the former wasted time and money by delivering a beta-test program through physical media such as CD-ROMS and other media, while the latter could save a lot of time and money by delivering their beta-test program through free means. Currently, the two corporations are using the Internet to get knowledge from customers and other IT and software companies are also using the Internet to get knowledge from customers. In other words, since customers want and choose a new and easy method, old and outdated methods are highly likely to be abandoned. In this sense, today’s companies should work hard to develop a user-oriented and customer-oriented medium, using multimedia and graphic technology. The purpose of using IT in socialization is to construct a virtual space that can replace face-to-face contact. Therefore, it is important to develop products or realize customization by developing and sharing customers’ images and concepts through multimedia-based presentation systems, video-conference systems and design support systems. In this regard, knowledge-creating companies should set up a proper system that can draw customers’ participation in creating knowledge in development, design and production processes. It is also essential for companies to establish a communication system based on telecommunications networks, for knowledge sharing among employees.

IT for externalization Externalization refers to a process where tacit knowledge that has gone through a socialization process is transformed into a specific concept (explicit knowledge). Through an externalization process, tacit knowledge becomes specified, and metaphors, analogies, concepts, hypotheses and models, take shape (Nonaka, 1998). Externalization includes creating concepts during a product development process, crystallizing CEOs’ thoughts and experts’ know-how into a language, and expressing and materializing covert needs of customers. It always involves the creation of easy-to-share and easy-to-deliver explicit knowledge within an organization by using a set of languages and visual messages (Nonaka, 1998). Not only foreign, but also Korean companies regard the conversion of tacit knowledge into explicit knowledge as their first priority in KM, to make tacit knowledge a part of organizational knowledge through transformation, and share it among employees. Many companies are constructing KM systems with this purpose in mind and trying to attract more participation from their employees by drawing up more detailed methods and setting up a new compensation system like the knowledge mileage system. They are making the utmost effort to attract as much support as possible from employees throughout the whole organization. Anderson Consulting is one of the companies that show the most active knowledge sharing. Its corporate culture is based on the principle of knowledge sharing. In principle, Anderson Consulting exchanges knowledge on a face-toface basis, but it has another system called “k-xchange” to enable experts around the world to exchange their knowledge more effectively. The firm allows an individual or group to have access, through “ANET”(a network for Lotus Notes users), to its knowledge database where explicit knowledge (transformed from tacit knowledge) is stored. Its employees, or teams, register their tacit knowledge acquired through work on the knowledge database, making it possible for all organization members to share and use the database. Meanwhile, Korea Telecom is setting up a portal Internet site to help all employees retrieve and use necessary information or knowledge by using a Web-based single platform. The corporation is also setting up a new KM system with the intention of connecting it with the legacy system. Core issues involving Korea Telecom’s construction of KM system is that, first, it is difficult to maximize the usage of information and knowledge with the legacy system and its conventional way of processing information. Hence, Korea Telecom is focused on establishing a new system that can manage the life cycle of knowledge from its creation to extinction through an externalization process. Second, the corporation is focused on constructing a new system, where different systems can be connected and operated together. The legacy information system was developed unit by unit and, therefore, had different

Knowledge conversion

321

BPMJ 9,3

322

structures where it was hard to retrieve or use information in an integrated way. Third, the new system should include a process where knowledge is shared and controlled on an each department basis. Such a process has not been put into practice owing to the absence of a proper system where information and knowledge can be systematically accumulated and utilized. The focus of Korea Telecom’s KC plan lies on establishing a KM system based on externalization. Unlike other companies, Korea Telecom is trying to establish a KM system on a step-by-step basis. The first step is establishing a pilot system for the core part of marketing. The second step is expanding the marketing area and developing a management system (e.g. personnel and finance management systems). The third step is developing a system for technology and management support. Daelim Information & Telecommunications Co. is working on KC through “Logos”, a groupware that the corporation created themselves. Daelim’s groupware “Logos” was created with three major principles in mind: “from explicit knowledge to another explicit knowledge”, “tacit knowledge sharing”, “transforming tacit knowledge into explicit knowledge”. Especially for the principle of transforming tacit knowledge into explicit knowledge, Daelim divided knowledge into two kinds: “on-line” and “batch.” It is also promoting systematic development of knowledge into explicit knowledge through HELP and Q&A sessions on online bulletin boards. The KM system that Daelim put into place is EDMS and Logos. EDMS mainly controls explicit knowledge, while Logos manages the conversion of tacit knowledge into explicit knowledge. Both Korea Telecom and Daelim Co. adopted a knowledge mileage system, as most companies did, in order to encourage employees’ participation in KM. On the contrary, Anderson Consulting has no such compensation provided in exchange for employees’ knowledge sharing and registering. This difference stems from different corporate cultures. Anderson Consulting considers it natural to create and share knowledge among personnel. As the above example shows, arrangement of the concept using a knowledge-based system is a requisite for externalization. Besides, the usage of CAD or CAM for the design development can be added to the process. There are also other methods that can be added such as documentation of successful cases and business know-how, knowledge delivery through e-mail, a groupware and utilization of database through a corporate intranet. Through the process of transforming tacit knowledge into explicit knowledge, employees can naturally realize the necessity of KC. It is important to convert individual thinking processes into corporate information and knowledge so that it can be integrated into a part of corporate assets and, for this, an employee-friendly system should be established.

IT for combination Combination refers to a process where explicit knowledge is converted into more complicated sets of explicit knowledge through the systemization of concepts and conversion of knowledge (Nonaka and Konno, 1998). It is a process where an individual or group can create new knowledge by combining its explicit knowledge with other explicit knowledge. Individuals exchange and combine information through such media as documents, meeting, telephone and computer networks. They create new knowledge by reconstructing the existing information through classifying, adding and combining tacit knowledge. Knowledge created through the combination process can be dubbed as the creature of traditional management information system (Riggins and Rhee, 1999). The combination process includes various works such as crystallizing related strategies and concepts and making management data and product specifications. Western enterprises are ahead of others in this field. When comparing them with Japanese companies, Western companies tend to create concepts by synthesizing all the explicit knowledge they have gathered, while Japanese companies create concepts from tacit knowledge by externalizing onsite information. This comparison shows different business characteristics in terms of information usage between Western and Japanese companies: the latter tend to think much of on-site information while the former do not (Nonaka and Kono, 1998). Hyundai Department Store Co. replaced paper and documents with digital files in processing and recording information. By doing so, the organization and its employees have their work experience stored on digital files for the corporate knowledge assets. The corporation classified knowledge into four types: product knowledge, customer knowledge, service knowledge and task knowledge. In order to control these four types of knowledge, they set up a new KM system, a groupware called “office-ware” that handles the conversion of existing knowledge into explicit knowledge. Knowledge can be registered, retrieved and acquired through “office-ware”. It also includes a special session for experts to answer questions from employees. Korea United Pharm. Inc., as one of small-to medium-sized companies, is making an effort to have KM in place by setting up a “cyber knowledge management center” through consultation with Anderson Consulting. The company collects and shares knowledge through a Web-based intranet. As one of the small-to medium-sized companies, it is difficult for the company to acquire and accumulate high-quality knowledge, owing to many unstabilizing factors such as difficulty with securing high-quality human resources and high employee turnover.

Knowledge conversion

323

BPMJ 9,3

324

To overcome these weaknesses, the company is implementing KM focusing on the combination of explicit knowledge. Its “cyber knowledge management center” consists of “intell-base”, “knowledge-base”, “management-insight”, and “discussion”. “Intell-base” provides all kinds of information about competition such as related circumstances, market, and competitors. Knowledge-base is a storehouse for corporate knowledge assets where achievement assets, finance assets, intellectual assets, human resource assets and infrastructure assets are classified into appropriate categories to systematically store and utilize corporate knowledge. “Management-insight” stores secret documents within the organization that is open only to its CEO. “Discussion” is a place for exchanging tacit knowledge. But the company depends highly on “melting pot meeting”(where employees can freely discuss certain issues in an environment isolated from regular work) for the exchange of tacit knowledge. Therefore, the rate of using cyber space to exchange tacit knowledge is low. SK Computer & Communications Co. puts more emphasis on IT in implementing KM because it considers that IT plays the role of an enabler for KM. The company’s KM system is a combined form of its database and groupware intended for sharing, accumulating and diffusing knowledge. The company utilizes its KM system focusing on the management of explicit knowledge (combination) and conversion of tacit knowledge into explicit knowledge (externalization). In addition, the company set up a communication plaza that mainly deals with the exchange of tacit knowledge. For IT in this area, a simulation technology for concepts creation and products development process – and a media technology for visual understanding – help companies enhance their competitiveness. Product development costs are huge, companies spend 70 to 90 percent of the development costs in the design process. Therefore, the development costs can be significantly reduced if predictable data are provided in advance, and a simulation technology using computers or workstations, experts system (ES) and AI are being used. Corporate competitiveness is dependent on how effectively a company takes advantage of a state-of-the-art technology in designing a product in the development process. IT for internalization Internalization refers to a process where new knowledge is created through the conversion of explicit knowledge into tacit knowledge within an organization (Nonaka and Konno, 1998). It is closely connected with “learning through practice” (Nonaka, 1998). Internalization takes place when explicit knowledge is exposed to a new concept or method that is better than the existing ones (Riggins and Rhee, 1999). Internalization involves a process where different kinds of explicit knowledge such as know-how, manuals and documents, that went through

socialization, externalization and combination is understood and experienced by each individual through role-playing or simulation. The way through which successful business examples are diffused within a company is relevant to this process. Experimental and reflective trial or experience is significant in this process (Nonaka, 1998). The examples of this process are business games, cyber university and cyber learning courses run by some companies. Not only Korean companies, but also foreign companies are showing much interest in employee training. Among Korean companies, Samsung Multi-Campus, Daewoo Information System e-Learning and LG Academy facilitate employees’ internalization process through the Internet and on-site education. Companies that lack their own education facilities entrust employee training to rival companies’ educational institutions (e.g. Daelim Information & Telecommunications Co. entrusts its employee training to Samsung and Daewoo, its rival companies). Educational link-up is possible not only between companies, but also between a company and a regular educational institution. For example, Korea Telecom is providing its employees with video-based education through satellites, through the link-up with Korea Advanced Institute of Science and Technology (KAIST). This shows that IT-based cooperation within a company, between companies and between a company and an educational institution is possible in training and educating employees. This also shows that the idea of e-commerce is being incorporated into the education sector. In other examples of using IT, manufacturing industries effectively operate reverse engineering using a three-dimensional screen, enabled by CAD and CAM. In the service sector, companies record and store customer complaints or questions on database so that they can retrieve necessary information later when developing new products, and accurately understand customers’ needs. In order to immediately respond to customers’ questions, ES or AI would be helpful. Figure 1 re-constitutes the usage of IT in Nonaka’s four types of KC, which companies can utilize when they implement KC. The utilization of KM systems by Korean enterprises Ernst & Young’s (1997) research shows that 431 American and European organizations think that the order of carrying out related projects for the introduction of KM systems should be as follows: creating an intranet (47 percent), data warehousing and creating knowledge repositories (33 percent), implementing decision-support tools (33 percent), implementing groupware to support collaboration (33 percent), creating a network of a knowledgeable workers (24 percent) and mapping sources of internal expertise experts (18 percent). In the same research, barriers to KM were as follows: changing people’s behaviors (54 percent), measuring the value and performance of knowledge

Knowledge conversion

325

BPMJ 9,3

326

Figure 1. IT for KC

assets (43 percent), determining what knowledge should be managed (40 percent) and justifying the use of scarce resources for knowledge initiatives (34 percent). Barriers to knowledge conveyance were as follows: culture (54 percent), top management’s self-assurance (32 percent), lack of shared understanding of strategy of business model (30 percent), organizational structure (28 percent), lack of ownership of the problems (28 percent), non-standardized process (27 percent) and information/communication technology restraints (22 percent). These results show that human resource and organization culture are a precondition for successive KM. Ruggles (1998) determined the degree of importance among process and technology as 50/25/25 respectively, stressing the importance of human resource-related factors. His point is not about ignoring IT altogether. He just points out that excessive focus on IT-based KM can be a burden to a company. The research team for this paper visited and interviewed several companies that adopted KM systems to create new knowledge through KC, in order to work out what their motivations were for adopting the system and how they run and use the system. The research team selected 13 organizations that adopted KM systems and contacted them through an e-mail to ask for an interview. Out of the 13 organizations that were contacted, only nine allowed an interview to be conducted. Interviewees were the heads of the teams within the companies who participated in implementing KM systems, or had been in charge of it. Table I shows which type of KC Korean companies consider important when establishing a KM system: seven out of nine companies placed greatest importance on combination (2.78 points on average). The next important process was externalization (2.56 points on average), the third one was socialization (1.56 points on average) and the last one was internalization (1.33 points on average). .

This is because the internalization process mostly occurs within the boundaries of an individual’s experience and thinking, rather than within an organization. Some companies placed equal amounts of importance on the four types of KC, but mostly companies placed different stresses on different types. This is because each company has a different corporate size and situation. For example, exchanging tacit knowledge through a face-to-face meeting might be much more effective for small-to medium-sized companies. And, it is natural that a combination process would be more important for a company that already retains a great deal of explicit knowledge. The most important thing in KC is that companies should place different importance on different knowledge depending on their corporate characteristics, to properly introduce and utilize a KM system. Table II shows each company’s KM system in detail, including purpose, who led the introduction of the system method used for the construction of the system, communication technology and characteristics. As for the purpose of introducing the system, most companies wanted to secure competitive advantage through customer satisfaction and strengthened competitiveness. Eight companies showed that a CEO led the introduction of the system, while only one company showed that a middle-level manager led the introduction. As for the construction methods, companies that already had a groupware system in place renovated it, while companies that did not have a system, or like the existing one, set up a new Web-based system. One particular characteristic concerning the introduction of KM is that it is divided into two methods: one is based on IT while the other is based on culture. This is the first among four characteristics that a CEO should take into account when implementing KM. The issue is about what should be the first priority. If companies simply utilize technology and process without considering human factors, they would fail to implement KM (Lee, 2000). Companies that introduced KM in the introduction stage of KM, were focused on IT. But companies that introduced KM in the growth or maturity stage of KM became more focused on culture. The reason the former companies mostly failed to implement KM properly was because they concentrated too much on IT, ignoring employee education

Socialization Externalization Combination Internalization

A

B

C

D

E

F

G

H

I

Average

K K W W

K W K K

K K W K

K K K K

K W K £

£ K K K

W K K K

K W K W

W W K K

1.56 2.56 2.78 1.33

Note: Relative importance within a company: very important (three points), a little important (two points), less important (one point), not important (no points)

Knowledge conversion

327

Table I. KC in terms of IT

DC Legacy S/E All GW Technologybased Internal/ external

DC New S/C All Intranet Culturebased Internal/ external

Large CS CEO

SI

Phased New C All Intranet Culturebased Internal

Manufacturing Large CA CEO

C

DC New C All Intranet Technologybased Internal

Manufacturing S/M-sized CA CEO

Communications Public CS Gov. Phased New E All Intranet Technologybased Internal

E

D

DC Legacy C All GW Technologybased Internal

Large KA CEO

SI

F

DC Legacy C All GW Culturebased Internal/ external

Multinational KA CEO

Consulting

G

DC Legacy E All GW Technologybased Internal

Large CVA CEO

SI

H

Large CS Middlemanager DC Legacy C Full-time GW Culturebased Internal/ external

Distribution

I

Notes: CS ¼ Customer satisfaction, CA ¼ Competitive advantage, KA ¼ Knowledge assets, CVA ¼ Creating value added, C ¼ Combination, E ¼ Externalization, S ¼ Socialization, DC ¼ Direct cutover, GW ¼ Groupware

Source

Implementation System Focus Target Communications Characteristics

Large CS CEO

Size/type Purpose Intro-level

Table II. Comparing nine companies in the establishment of KMS

SI

B

328

Industry

A

BPMJ 9,3

and promotion. Given this fact, it is better for the latter companies to work hard on establishing an organization culture for approximately a year before actually implementing the system. The importance of IT might seem relatively reduced because of cultural importance. However, forming an organization culture is just another precondition for successive business, while IT is always at the center of spreading and sharing an organization’s culture. Other preconditions for successive business are employee education and promotion. For this, most companies use an internal broadcast or e-mail. These means are not effective, but forceful. Hence, utilizing appropriate IT would be necessary to form an effective organization culture. As the above cases show, the second issue is that externalization and internalization processes cannot generate a synergy effect in KC if the processes are only focused on individuals. The use of networks is needed to solve this problem. A groupware would be effective within an organization while an intranet and extranet would be more effective for the integration of the within and without of an organization. The third issue is that companies should adopt different technologies and apply them to different parts, depending on their size and characteristics. Hansen et al. (1999) argued that method and scale of investment in IT should be different depending on organization strategies (e.g. personalization or codification). Excessive investment in IT would be a waste of money. On the other hand, companies would not be able to effectively respond to the changes in the competitive environment if the system was not capable. In summary, IT-utilizing strategies should change according to company sizes, characteristics and financial situation. Finally, companies should prioritize and introduce the most necessary IT first. If a company uses a high-level of IT, despite its low-level of KM, or vice versa, IT would be meaningless and ineffective in both cases, and a waste of money. Therefore, it is recommended that companies should assess their own level of KM, and check out necessary conditions for the introduction of the most adequate IT beforehand. It is desirable that they should work out what kind of IT is possible at each stage and predict the results in advance. The utilization of external knowledge in KM The relationship between competitiveness of an enterprise, and knowledge and education is getting close, and the competitive environment of an enterprise is getting fierce. In addition, the evaluation of an enterprise is based on knowledge. The management of knowledge is becoming an important factor in an enterprise, for which existing knowledge should be acquired and managed, and new sources of knowledge should be developed for a long-term achievement

Knowledge conversion

329

BPMJ 9,3

330

Figure 2. The model for extended knowledge enterprise in terms of KC

(Sawhney and Prandelli, 2000). In particular, special efforts should be made on the acquisition of external knowledge. Since internal knowledge is not sufficient for the development of an enterprise and proper competitiveness can be acquired only if external knowledge combines with internal knowledge, the acquisition of external knowledge is necessary. External knowledge should be acquired from cooperative enterprises, partners, and the government, and clients, who have a direct relationship with the profits of enterprises. Research on the subject of knowledge retention and those in a systematic scale take up most of the research conducted until now. However, not all research deals with the acquisition, sharing, extension, and creation of knowledge inside the organization. Researchers think that the inflow of external knowledge in a systematic, rather than private scale is needed, and extended knowledge enterprise, which embraces all the subjects, is also needed. An extended knowledge enterprise is a virtual form of knowledge firm that has one phase, upgraded from existing knowledge community. This means the knowledge of the stakeholders concerned is shared through KM systems, thus forming a cartel centered on subsidiary firms involving the suppliers, competitors, and customers. The extended knowledge enterprise model for KC in Figure 2 shows that the KC model of the stakeholders concerned has been integrated, based on Nonaka’s KC model. This shows that the acquisition, sharing, application, and creation of knowledge should be conducted in the whole organization, not in just one part of an organization.

General opinion of external knowledge is confined to the knowledge of cooperative enterprise, rival companies, partners, and clients. However, it is desirable to secure as much valuable knowledge as possible, rather than limiting the subjects, which are the sources for an enterprise to acquire knowledge. Many might think that the source of knowledge comes from the inner part of an enterprise, but according to the research done by Truran (1998), 64 percent of employees in charge of marketing in 200 companies answered that it comes from the outer part of an enterprise. This shows that an enterprise depends more on internal knowledge than on external knowledge. This also shows that enterprises have recognized the limitation, uncertainty, and insufficiency of internal knowledge, thus introducing external knowledge. Introduction of external knowledge is already getting a theoretical basis in respect of KM. Lim et al. (1999) conducted a research on the interchange of knowledge with external environments including providers, clients, and organizations using IT, and the management and measurement of such knowledge. Another similar research was conducted by Duffy (2000) and this research suggests an extended enterprise for the flow and sharing of knowledge, and both researches done by Lim and Duffy present the direction to efficiently acquire and apply valuable knowledge. Extended knowledge enterprise and IT Companies researched for this paper have one thing in common: they are all focused on KC within an organization. But the amount of knowledge that can be acquired within an organization is limited. In order for companies to acquire high-quality knowledge in an effective way, they should expand their knowledge boundaries. The boundaries should not be confined to the link-ups among organizations (especially, link-ups among affiliated companies in a conglomerate or in a large business group). Companies should set up an expanded border that includes all stakeholders. The door for an input should be open while the door for an output should be closed from outside. This border should form a comprehensive community covering customers, suppliers, competitors, the government and academic institutions. Many scholars have emphasized the necessity of a knowledge community to share knowledge with one another (Powell, 1998; Brown and Duguid, 1998; McDermott, 1999; Sawhney and Prandelli, 2000). However, this community has its own limitation in sharing knowledge: it does not have a formalized or structured link. As a result, the community has a weak connection among members, which makes KC and sharing hard. Several factors that obstruct KC and sharing are members’ unwillingness to expose their core competence and knowledge to others, a lack of trust, and rival relationships among members.

Knowledge conversion

331

BPMJ 9,3

332

That is, knowledge sharing between organizations or between organizations and customers is recognized as the same transfer of value as that of a product or service that involves a reasonable compensation. The compensation can be provided not just by material means, but also by customer satisfaction, by providing quality products or services. There is a great deal of difference on this matter among stakeholders because each stakeholder has different needs and expectations. On the part of suppliers, a continuous relationship with customers can be a compensation for providing and sharing their knowledge. On the part of academic institutions (including universities and research institutes), the compensation for their knowledge sharing would be the acquisition of practical knowledge from industries and sustainable cooperative relationships with companies. A knowledge community alone would not be enough because it has a weak connection among its members and, therefore, its effectiveness is not certain. Therefore, it should form an extended knowledge enterprise, initiated by the main company for KC. An extended knowledge enterprise is similar to the Internet-induced virtual enterprise in some ways. But it is different from virtual community collaboration, in that the latter has a different main operating body. In virtual community collaboration, a new organization that runs in the form of a consortium is created. But virtual community collaboration is not effective, in that all decisions are made by all participating companies with different interests, and, therefore, a dispute over who will have a right to KC and management can arise after the shut-down of the Web site. But there is only one company at the center of an extended knowledge enterprise, which plays a leading role and is responsible for KM. An organization member of this enterprise provides its knowledge to share it with other members, thus causing no problems such as intellectual property right. That is because each organization has the ownership of its knowledge, even after it is provided for sharing. The most essential IT for an extended knowledge enterprise is a network. Currently, most companies use such network technologies as a groupware and Web-based intranet. This implies that companies are still focused on KC within the organization. But knowledge sharing and conversion within an organization is not enough to deal with a rapidly changing competition environment (Amidon, 1997). Most IT that companies adopt is, basically, based on an established knowledge base that focuses on a KM system, and a network technology within an organization that focuses on the linkage of all departments. Therefore, they can easily enlarge these legacy systems into an extended knowledge enterprise by constructing a network environment linked with the outside. A network for an extended knowledge enterprise can be constructed with the same concept as the one for an information system. The construction

process can vary depending on relationships among participating members, technological capabilities and asset size. When relationships among members are a major concern, the appropriate network – horizontal or vertical – would have to be decided upon. Even if a network for an extended knowledge enterprise is constructed based on the same idea as the one for the information system, the two systems have differences in the amount and direction of knowledge transfer, as well as in the re-processing and sharing of knowledge. A network for information systems tends to be one-way, where only one-time use is possible. On the other hand, a network for an extended knowledge enterprise is twoway: it can re-gather and re-process the knowledge provided for customers and then deliver it within and without the enterprise. This, in the broad term, means that circulation of knowledge through which little knowledge is expanded and developed while shifting itself to the outside of an organization (together with the internal organization). Most companies are using a groupware or intranet for their knowledge sharing network technology. In order for these networks to be set up with the internal point of view to extend further into the outside, companies should adopt a new network technology or enlarge the existing networks. Companies that established a groupware as their knowledge-sharing network should construct a new Web-based open network. But companies that have a knowledge-sharing network in place through an intranet can easily extend its network, which means more effectiveness. The construction of a Web-based network generally consists of a different level of platforms such as the Internet, intranet and extranet. When adapting an existing groupware to the Internet, companies should set up a “group-net”, a combined form of a groupware and the Internet, which is compatible with the Internet. A group-net is a Web-based and group-focused network, where its users can move between the Internet and groupware, depending on its openness. Therefore, it is as effective as an intranet and extranet in terms of control and management. In order to use a groupware on the intranet platform, users should download client software for that groupware. Another difference between a groupware and a group-net is that the former operates independently of user interface for the Internet while the latter operates maintaining it, providing users with the same environment as the Internet. A group-net’s advantages are that it enables users to freely exchange knowledge and information with the outside through the Internet-enabled environment while maintaining the existing network system, and that it forms a single system under the platform of the existing groupware, eliminating redundancy and providing integrity. A model for an extended knowledge enterprise in terms of KC, as Figure 2 shows, has four distinctive borders of KC, while a model (Figure 3) for an

Knowledge conversion

333

BPMJ 9,3

334

Figure 3. An extended knowledge enterprise using IT

extended knowledge enterprise using IT has a dotted line among four types of KC, which means bringing down a barrier to KC. IT plays a more effective role when applying to an extended organization rather than within an organization. Conclusions More and more companies are implementing KM in Korea. As they did with other kinds of new management strategies, companies are about to reorganize their structure and introduce appropriate IT that is suitable for the new structure. In the business world, on a CIO’s advice and persuasion, a CEO may introduce the best technology and strategy available at that time and experience both success and failure with it. This paper is intended to present the most appropriate IT by reviewing Nonaka’s four types of KC, to provide an opportunity for benchmarking through case studies of Korean companies, and to provide a solution for an effective usage of IT in creating, accumulating and sharing knowledge. The recent controversy about IT is whether its usage is truly effective to a company. KM experts admit that utilizing IT can be an infrastructure for KM. They suggest companies can effectively share, convert, accumulate and create knowledge by using IT. In summary, Korean companies evaluated the usage of IT in introducing KM positively. All nine companies studied, regardless of their types and sizes, have already established a KM system through IT and are implementing the KM system.

All interviewees stressed the necessity of IT and considered that IT is an effective tool to establish a KM system. But this empirical study finds that not every business needs IT for KC. Small- to medium-sized companies consider face-to-face KC more adequate to socialization because of the small number of employees and some chronic problems such as high replacement rates and difficulty with securing highquality human resources. As discussed earlier, companies can establish a proper KM system only if they take into consideration such factors as corporate cultures and characteristics, individuals and organizations, and technological characteristics. In particular, it is self-evident that the role of the stakeholders is getting more important since an organization’s knowledge is limited and their knowledge should serve as core knowledge that helps an organization survive in a competitive environment. An organization should extend its borders to acquire and use this knowledge, and, for this, it should expand itself into an extended knowledge enterprise that is based on trust and compensation for the stakeholders. The first priority of an extended knowledge enterprise is to obtain the knowledge that can be used to manage the interests of customers, suppliers, public sector and competitors (that exists outside of the central organization) and core knowledge that promotes their participation. In addition, this study finds that network-based IT can make the knowledge base that facilitates KC be effectively shared among the above groups/stakeholders. This study is significant in that it has established a concept on a new knowledge firm model for efficient acquisition and management of external knowledge. For managers, this model can be a guide to acquiring and applying external knowledge through IT in relation with outside. This study is subject to a number of limitations. First, our new model is not confirmed by empirical test. Second, the result may include some biases since the sample was nine organizations – all of which were Korean. Therefore, the result of our study may have to be carefully interpreted. In the next stage of this research, first, research on the main cause for successful extended knowledge enterprise is necessary. Second, comparative research with knowledge community and virtual corporation is needed. Third, demonstrative analysis to certify the research is required. References Amidon, D. M. (1997), “Customer innovation: a function of knowledge”, available at: www.entovation.com. Brown, J.S. and Duguid, P. (1998), “Organizing knowledge”, California Management Review, Vol. 40 No. 3, pp. 90-111. Duffy, J. (2000), “Knowledge management: what every information professional should know”, The Information Management Journal, Vol. 34 No. 3, pp. 10-16.

Knowledge conversion

335

BPMJ 9,3

336

Ernst & Young Center for Business Innovation and Business Intelligence (1997), “Executive perspectives on knowledge in the organization”, available at: www.ey.com Gore, C. and Gore, E. (1999), “Knowledge management: the way forward”, Total Quality Management, Vol. 10 No. 4/5, pp. S554-60. Hansen, M.T., Nohria, N. and Tierney, T. (1999), “What’s your strategy for managing knowledge?”, Harvard Business Review, Vol. 77 No. 2, pp. 106-16. Lee, J. (2000), “Knowledge management: the intellectual revolution”, IIE Solutions,, Vol. 32 No. 10, pp. 34-7. Lim, K.K., Ahmed, P.K. and Zairi, M. (1999), “Managing for quality through knowledge management”, Total Quality Management, Vol. 10 No. 4/5, pp. 615-21. McDermott, R. (1999), “Why information technology inspired but cannot deliver knowledge management”, California Management Review, Vol. 41 No. 4, pp. 103-17. Nonaka, I. (1998), “Intellectualizing capability”, in Na, S.U. (Ed.), Knowledge Management, Meilkyungje Shimunsha Publishing, Seoul, South Korea. Nonaka, I. and Konno, N. (1998), “The concept of ‘ba’: building a foundation for knowledge creation”, California Management Review, Vol. 40 No. 3, pp. 40-54. O’Leary, D.E. and Selfridge, P. (2000), “Knowledge management for best practice”, Communications of the ACM, Vol. 43 No. 11, pp. 12-24. Powell, W.W. (1998), “Learning from collaboration: knowledge and networks in the biotechnology and pharmaceutical industries”, California Management Review, Vol. 40 No. 3, pp. 228-40. Riggins, F.J. and Rhee, H. (1999), “Developing the learning network using extranets”, International Journal of Electronic Commerce,, Vol. 4 No. 1, pp. 65-83. Ruggles, R. (1998), “The state of the notion: knowledge management in practice”, California Management Review, Vol. 40 No. 3, pp. 80-9. Sawhney, M. and Prandelli, E. (2000), “Communities of creation: managing distributed innovation in turbulent markets”, California Management Review, Vol. 42 No. 4, pp. 24-54. Truran, W. (1998), “Pathways for knowledge: how companies learn through people”, Engineering Management Journal, Vol. 10 No. 4, pp. 15-20. Zack, M.H. (1999), “Managing codified knowledge”, Sloan Management Review, Vol. 40 No. 4, pp. 45-58. Further reading Nambisan, S., Agarwal, R. and Tanniru, M. (1999), “Organizational mechanisms for enhancing user innovation in information technology”, MIS Quarterly,, Vol. 23 No. 3, pp. 365-95. Nonaka, I. and Takeuchi, H. (1995), The Knowledge-Creating Company, Oxford University Press, Oxford.

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

Interaction management

Interaction management

The next (and necessary) step beyond knowledge management Stephen W. Smoliar

337

FX Palo Alto Laboratory, Palo Alto, California, USA Keywords Knowledge management, Information exchange, Crisis management Abstract Purveyors of knowledge management software have a disconcerting tendency to promote the myth that all problems may be solved by more powerful tools for the exchange of information in the workplace. This fallacy is based on the faulty assumption that knowledge management is about the management of knowledge (as if knowledge were a commodity that could be managed), as opposed to the management of people whose work depends critically on what they know. The origins of knowledge management are far more firmly rooted in the psychological legacy of organizational communication than they are in the technological legacy of information management systems. However, even organizational communication is an inadequate foundation, since various schools of thought in social theory, particularly the structuration theory of Anthony Giddens, inform us that interaction (in the workplace or in any other social setting) is not strictly limited to communication. Knowledge management thus requires moving beyond simplistic models of information exchange to more challenging problems of leveraging social interaction to the advantage of the enterprise. This paper focuses on the claim of structuration theory that the dimension of communication should be supplemented with additional dimensions of power and sanction. This perspective is then examined in light of a case study of crisis management practices, and the case study provides a basis for addressing implications for technological support.

Introduction Motivation The “knowledge movement” has now been with us for about two decades, at least if we trace its origins to Ikujiro Nonaka’s research on “organizational information creation” in the 1980s[1]. Has it brought us new insights regarding innovative practices in the workplace, or has it devolved into little more than a business opportunity for software vendors claiming to support the exchange of This paper is a result of many long and stimulating conversations. The author was originally encouraged to pursue these ideas by Masao Kato in response to material discussed at the 1999 International Conference on Computer Communication, whose theme was “Digital Convergence for Creative Divergence”. He received invaluable feedback and assistance from Sara Bly as he first began to articulate his ideas for Dr Kato. Subsequent cultivation of these thoughts benefited from valuable suggestions from Noam Cook, Brigitte Jordan, and Ralph Sprague. More recently he has enjoyed the input from stimulating conversations with his teammates in the Social Computing Project at FX Palo Alto Laboratory, Les Nelson and particularly Project Leader Elizabeth Churchill. He has also received useful feedback on the social computing concept from his colleagues at Fuji Xerox Singapore. Finally, the author must thank John Seely Brown for always prodding him to think in new directions. He very much enjoyed a lecture in which Dr Brown proposed to rewrite the Cartesian motto as “We participate, therefore we are”; and in many ways this essay is a sermon on that attempt to revise Descartes!

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 337-353 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477920

BPMJ 9,3

338

information? The only answer we can give with conviction is that the injection of “knowledge-based terminology” (knowledge creation, knowledge management, tacit knowledge, etc.) has led to more confusion than enlightenment. Phrases like “knowledge creation” and “knowledge management” entail the faulty assumption that knowledge is a commodity that not only can be produced, but also whose production can be managed. However, even the most rigorous attempts at “scientific management” – terminology this time due to Taylor (1911) – recognize that it is people, rather than commodities, that are managed. Thus, the more substantive message to be found in the writings of Drucker (1964), Nonaka and Takeuchi (1995), Davenport and Prusak (1998), and many others who have published under the banner of knowledge management, is concerned with the management of people whose work depends significantly on what they know. This paper is an attempt to review critically the concept of knowledge management in terms of two fundamental questions: (1) What does it really mean to manage people whose work depends significantly on what they know? (2) How may technology facilitate this management task? The first question will lead to abandoning one terminology for another, proposing that “interaction management” is a more suitable descriptor than “knowledge management”. Background on knowledge The confusion engendered by knowledge-based terminology can probably be traced back to the epistemological traditions of Western philosophy, particularly where that tradition draws on concrete representations, such as the formulae of symbolic logic, which almost can be viewed as commodities. While Nonaka and Takeuchi (1995) have claimed that the Japanese intellectual tradition poses an important departure from Western epistemology, that claim is not really substantiated by any survey of that tradition, such as that of Collins (1998). What is really at stake in any review of philosophical foundations is the opposition of objectivity and subjectivity. Both positions have a long history in Western philosophy (dating back at least to Plato and Aristotle); but the Japanese proponents of Kyoto School philosophy (which emerged in the early twentieth century and has been cited by Nonaka and Takeuchi (1995) as a major inspiration) attempted a synthesis (possibly under the influence of Georg W. F. Hegel) of the objectivity of German metaphysics with the subjectivity of Zen. Nonaka and Takeuchi (1995) have also claimed that this synthesis is reflected in Japanese management style; but the connection seems unlikely, given a general lack of interest in Kyoto School philosophy in contemporary Japan – possibly because of its association with the Japanese invasion of China in 1937 (Collins, 1998).

If we wish to leverage the history of philosophy, we would probably do better to turn to the social theorist Mannheim (1985), who was writing in Germany at the same time that Kyoto School philosophy was emerging. Mannheim (1985) derived from an examination of the history of philosophy an evolutionary trend of how man has wrestled with understanding the nature of knowledge. What is most important about Mannheim’s approach is that he presents epistemology as only the first stage in this evolutionary progression, subsequently followed first by psychology and ultimately by sociology. Let us briefly examine what distinguishes each of these points of view. The earliest Greek philosophers quickly discovered that different people not only had different views of the world but also different foundational assumptions concerning the very objects that constituted the world. When Theaetetus informs Socrates that he was told “that true belief with the addition of an account (lo´goz) was knowledge” (Plato, 1963), Socrates can easily uncover the uncertainties surrounding what we mean by “belief,” “true,” and justifiable “account.” As Mannheim (1985) observed: Epistemology sought to eliminate this uncertainty by taking its point of departure not from a dogmatically taught theory of existence, nor from a world-order which was validated by a higher type of knowledge, but from an analysis of the knowing subject.

This seed of “the knowing subject” would ultimately flower into the Cartesian motto, so significant that, for a change, it deserves to be set in its proper context (Descartes, 1985): Lastly, considering that the very thoughts we have while awake may also occur while we sleep without any of them being at the that [sic] time true, I resolved to pretend that all the things that had ever entered my mind were no more true than the illusions of my dreams. But immediately I noticed that while I was trying thus to think everything false, it was necessary that I, who was thinking this, was something. And observing that this truth “I am thinking, therefore I exist” was so firm and sure that all the most extravagant suppositions of the sceptics were incapable of shaking it, I decided that I could accept it without scruple as the first principle of the philosophy I was seeking.

However, if we are going to talk about a subject as a person, as opposed to an embodiment of logical reasoning processes, then we have to depart from the abstractions of epistemology and acknowledge that every individual also has a psychology (Mannheim, 1985): It became evident that much more could be said about the manner in which the structure of the subject influenced his world-view when one made use of animal psychology, child psychology, the psychology of language, the psychology of primitive peoples, and the psychology of intellectual history than when one set about it with a purely speculative analysis of the achievements of a transcendent subject.

Thus, the sense of not only belief but also truth and justification may be derived from the behavior of the “knowing subject”, rather than relationships between the properties of the subject and the properties of the world in which that subject exists.

Interaction management

339

BPMJ 9,3

340

Nevertheless, knowledge must involve more than the behavior of the subject, because no individual lives in isolation from all other individuals. Individuals are always members of social groups; and what Mannheim calls “the context of group life” (Mannheim, 1985) is as important to establishing what constitutes belief, truth, and justification as is the psychology of the individual. The recognition of this context forms the basis for what Mannheim came to call “the sociology of knowledge”, thus providing the lens through which knowledge may be seen as a product of interaction and identity a product of participation. Indeed, the very act of participation figures significantly in Mannheim’s own characterization of thought (Mannheim, 1985): Strictly speaking it is incorrect to say that the single individual thinks. Rather it is more correct to insist that he participates in thinking further what other men have thought before him.

This “context of group life” thus poses a challenge to the usual premises of knowledge management. The practice of knowledge management tends to concentrate on the exchange of information through different approaches to communication[2], while the sociology of knowledge assumes that the kind of interaction from which knowledge is derived is not strictly a matter of communication. Indeed, if we are to try to understand interaction, we quickly encounter a rich variety of points of view as to what constitutes its nature. Let us now consider how previous research into the nature of interaction offers alternatives to current thinking about knowledge management. Previous research on interaction A system-centric position In his 1990 presidential address to the American Association for Artificial Intelligence, long before knowledge management was emerging as a new opportunity to invent software solutions, Daniel Bobrow challenged his audience to turn their attention beyond “the principles of construction of . . . intelligent, but deaf, blind, and paraplegic agents” and to focus instead “on systems consisting of active agents with multiple goals, communicating among themselves, and interacting with the world” (Bobrow, 1991). The body of his talk then addressed issues involving what he called “three dimensions of interaction: communication, coordination and integration”. Bobrow’s dimensions provide one useful perspective on why interaction involves more than communication. Bobrow began by recognizing that there is more to communication than worrying about insuring that the signal bits from a source can be passed through a channel to a destination without being corrupted by noise (Bobrow, 1991): For communication to exist between two agents, there must be some common ground of mutual understanding. Where does this come from, and how does it develop? What techniques are used by people and systems to build and extend this base for communication? Communication between a particular pair of agents might not always be easy, or even possible. In such cases, communication can be facilitated by interposing a mediating agent.

While Bobrow grounded communication in mutual understanding, he approached coordination more as a matter of resource allocation (Bobrow, 1991): With multiple agents with multiple active goals, progress requires agents to share resources and work toward some common goals. Various organizational structures, for example, based on markets and business hierarchies have been used in the resource-allocation process. But resources are not the only thing that must be shared. For independent agents to work together, they must be able to predict other’s behavior, but not necessarily in great detail. Joint commitments to future action are a useful way of organizing this information.

Interaction management

341

Finally, Bobrow (1991) viewed integration in terms of not only relationships among systems but also relationships between systems and work practices: Systems do not exist in isolation. Systems must solve problems that have some payoff outside the system itself. For agents that we build to be useful, they must fit in with the current work practices of both people and other computer systems.

Bobrow’s challenge, while laudable, particularly for its time, still focused its attention on the technology of systems, rather than the sociology of the users. Before we can aspire to his goal of building better systems that interact with people, we need to refine our understanding of how people interact with one another. Let us now complement Bobrow’s dimensions of interaction with an alternative approach, grounded in the structuration theory of social theorist Giddens (1984). Giddens’ dimensions of structure as a basis for interaction Like Bobrow, Giddens viewed interaction as a three-dimensional “space”. He called the three dimensions of his space signification, domination, and legitimation; and he drew the two charts, illustrated in Figure 1, to summarize them. Let us discuss the nature of each of these dimensions in a bit more detail.

Figure 1. Giddens’ dimensions of interaction

BPMJ 9,3

342

Figure 2. The individuality of interpretation

Signification. Signification may be described most simply as understanding what things mean. The term is generally associated with semiotics (Barthes, 1973), where the “things” are usually taken to be symbolic expressions. In the business world “things” tend to be manifested in documents; but their manifestation in talk often plays a critical role when the negotiation of understanding is at stake. The problematic issue is that understanding cannot be grounded in the structures of the symbolic expressions themselves. Rather it is based on each individual’s scheme for interpreting those symbolic expressions; and, as is illustrated in Figure 2, there is never any guarantee that any two individuals will interpret the same symbolic expressions in the same way (Mannheim, 1985). Signification is thus not a problem of abstract logic but rather one of the individual and social psychologies of “knowing subjects”. Domination. If signification is about understanding what things mean, then domination is about understanding who has authority; and, if differences in interpretative understanding of what things mean has become a norm in business operations, then managing authority has also become an increasingly problematic issue. Knowledge management has raised our awareness that value resides in the personal knowledge of individuals (Davenport and Prusak, 1998); but, when those individuals appreciate the value of their own personal knowledge, they have little tolerance for domination. Thus, we see an increasing need for more cooperative approaches to organization, such as communities of practice (Wenger, 1998); but, while communities of practice provide a conducive environment for negotiating those differences in interpretation that lead to problems of signification, they are not always decisive enough to be both rapid and effective. Furthermore, the very notion of a community is one that raises the distinction between “self” and “other” to the level of groups; so, while cooperation within a community is rarely a problem,

issues of domination involving the relationship of the community to the rest of the organization can still be very difficult. Legitimation. Legitimation complements domination’s understanding of who has authority with the understanding of what is acceptable. The scope of legitimation is usually both practices and artifacts, and the challenge of legitimation is the regulation of normative behavior without suppressing individual identity. Any practice of legitimation must thus recognize privacy; and, if social capital is to have any valid currency, particularly where trust is concerned (Putnam, 1995), any normative behavior must include the encouragement of tolerance. Legitimation establishes what Giddens (1984) calls “ontological security”, which is basically a sense of knowing what to expect, without which the risk of taking any action might become crippling. Objective The context of this social perspective reveals that any problem of managing people, particularly people whose work depends significantly on what they know, is basically a problem of managing how those people interact, hence the argument that the term “interaction management” better suits the needs of today’s workplace than does the term “knowledge management”. The “knowledge movement” was useful in its efforts to revive our consciousness of the significance of organizational communication, particularly when the business world was in danger of ignoring that significance amidst a flood of results in experimental psychology whose impact was not always evident (Redding, 1972). However, as knowledge management is gradually assimilated into the workplace, we see that its perspective tends to be limited by simplistic models of information exchange, at a time when it is necessary to address the more challenging problems of leveraging social interaction to the advantage of the enterprise. Confronting those problems must be served by recognizing, as Giddens has done, that a work environment is a much richer collection of social interactions, all of which fall under the responsibility of effective management. Thus, rather than concentrating on roles that technology can play in the flow of information, we should direct our attention to the role of technology in Mannheim’s richer “context of group life”, particularly life in the workplace. Expected contribution Invoking Giddens’ structuration theory to model the nature of interaction in the workplace is not, in itself, a novel idea. Yates and Orlikowski (1992) took a structurational approach in their effort to identify genres of organizational communication. Another approach to organizational communication is that of Poole (1985), who studied the impact of “organizational climate” on communication. He applied structuration theory to these “climatic” conditions along three dimensions:

Interaction management

343

BPMJ 9,3

. . .

344

structural properties of the organization; apparatuses that directly produce and reproduce climates; and members’ knowledge and skills.

Both of these studies, however, deal only with the signification dimension of Giddens’ framework. This paper is based on a case study that illustrates the manifestation and interaction of all three dimensions. It thus constitutes an attempt to honor Giddens’ own admonition to avoid dealing with these dimensions in isolation (Giddens, 1979): The communication of meaning in interaction does not take place separately from the operation of relations of power, or outside the context of normative sanctions. All social practices involve these three elements.

It is important for the reader to recognize, however, that this is not a case study of our own design. Rather, it is based on a close reading of Andrew Grove’s documentation of his experiences at Intel in his book Only the Paranoid Survive: How to Exploit the Crisis Points That Challenge Every Company and Career (Grove, 1996). The key theme of this book is that a major problem at Intel was the need to recognize “strategic inflection points”. Grove argues that the only way to do this “is through the process of clarification that comes from broad and intensive debate” (Grove, 1996). In other words, the organization as a whole will not recognize a strategic inflection point unless it supports sufficiently rich processes of interaction, through which that “process of clarification” can take place. This paper hopes to demonstrate that a better understanding of those processes of interaction can lead to a new generation of software that facilitates such interactions and that this software will constitute the basis for a new approach to “interaction management”. Structure of the paper The following (and central) section thus presents the case study that illustrates the three dimensions of Giddens’ structuration theory put into practice through Grove’s approach to the management of “broad and intensive debate”. Giddens’ perspective is discussed in terms of the value it adds to the way in which knowledge management might contribute to such situations. This is then followed by a section that examines how new technologies can provide support for this new “interaction management” perspective. Finally, a concluding section summarizes the results, cites specific contributions, addresses possible limitations, and identifies opportunities for future research. Case study: Giddens’ dimensions in practice Signification Signification is clearly important, since clarification comes from members of the organization negotiating differences in understanding what things mean.

This is similar to what Bobrow (1991) calls “mutual understanding” in his characterization of communication; but there is a broad gulf between a multiagent artificial intelligence system and the social world of Intel! Unlike any artificial intelligence system, Intel is a vast organization whose members are endowed with a rich palette of shared experiences. Unfortunately, there is rarely any guarantee that any two members of that organization will interpret a shared experience in the same way. This is why “broad and intensive debate” is a recurring leitmotif in Grove’s text and why he believes in applying it across all the different kinds of experiences that are shared. The significance of the role of debate stems from the fact that, as Figure 2 attempts to illustrate, signification is a highly personal process and that the debate provides the platform for negotiating differing interpretations of shared experiences. It is also important to observe that Grove (1996) does not see consensus as the ultimate objective of debate. As we have seen, signification is always going to vary from individual to individual; and Mannheim (1985) emphasized repeatedly that, while various approaches to argumentation and rhetoric can often have a profound impact on changing a point of view, neither will ever totally eliminate variation in signification. For Grove (1996), signification is actually a space of interpretations that may span the entire organization. What is important for any individual is to clarify his or her own sense of where he or she is situated in that space: It is important to realize what the purpose of these debates is and what it isn’t. Don’t think for a moment that at the end of such debates all participants will arrive at a unanimous point of view. That’s naive. However, through the process of presenting their own opinions, the participants will refine their own arguments and facts so that they are in much clearer focus. Gradually all parties can cut through the murkiness that surrounds their arguments, clearly understand the issues and one another’s point of view. Debates are like the process through which a photographer sharpens the contrast when developing a print. The clearer images that result permit management to make a more informed – and more likely correct – call.

In this respect Grove’s language actually reflects Mannheim’s (1985) vision of the “unfolding” of knowledge through group interaction: In actuality it is far from correct to assume that an individual of more or less fixed absolute capacities confronts the world and in striving for the truth constructs a world-view out of the data of his experience. Nor can we believe that he then compares his world-view with that of other individuals who have gained theirs in a similarly independent fashion, and in a sort of discussion the true world-view is brought to light and accepted by the others. In contrast to this, it is much more correct to say that knowledge is from the very beginning a co-operative process of group life, in which everyone unfolds his knowledge within the framework of a common fate, a common activity, and the overcoming of common difficulties (in which, however, each has a different share).

Such language stands in sharp contrast to the knowledge management perspective on signification. Consider, for example, some “mottos” of knowledge management extracted from a recent vision piece from the corporate world (Smith and Farquhar, 2000):

Interaction management

345

BPMJ 9,3

The organization knows what it knows and uses it and knows what it needs to know and learns it. For any project, for any customer, the project team delivers the knowledge of the overall organization.

346

The organization delivers the right information to the right people, at the right time – with the tools they need to use it!

This is not a vision of “broad and intensive debate” as part of life in the work environment. It is a vision of answers being delivered accurately, reliably, and rapidly. Any problems of signification are simply dealt with through how the answers are delivered. Mannheim’s vision may thus be seen as a sharp critique of at least this particular vision of knowledge management. Domination It generally tends to be assumed that strong leadership is most necessary in times of crisis, but this raises questions as to what role such leadership can play in a setting that encourages the free expression of different interpretations of shared events. Grove’s approach is to deal with crisis situations in terms of the relationship between domination and cooperation and to manage this relationship through an infrastructure based on responsibility, rather than authority. Thus, for example, contributing to debate over signification is presented very much as a matter of responsibility (Grove, 1996): Don’t justify holding back by saying that you don’t know the answers; at times like this, nobody does. Give your most considered opinion and give it clearly and forcefully; your criterion for involvement should be that you’re heard and understood. Clearly, all sides cannot prevail in the debate but all opinions have value in shaping the right answer.

Similarly, acting on the results of a debate is a responsibility to be shared across the different levels of management, rather than being a buck that stops at a sufficiently high level of authority (Grove, 1996): The more complex the issues are, the more levels of management should be involved because people from different levels of management bring completely different points of view and expertise to the table, as well as different genetic makeups.

It goes without saying that this is a perspective that is not even acknowledged, let alone honored, in most visions of knowledge management. Legitimation From this point of view, debate is then resolved by supporting and managing community opinion, rather than by authoritarian decision. In other words legitimation of a communitarian sense of what is acceptable is more important than how the organization chart determines who has authority. As Grove (1996) observes, such legitimation is no easy matter: This kind of debate is daunting because it takes a lot of time and a lot of intellectual ergs. It also takes a lot of guts; it takes courage to enter into a debate you may lose, in which

weaknesses in your knowledge may be exposed and in which you may draw the disapproval of your coworkers for taking an unpopular viewpoint. Nevertheless, this comes with the territory and when it comes to identifying a strategic inflection point, unfortunately, there are no shortcuts.

Grove (1983) also makes it a point to emphasize that community opinion is not the same thing as consensus, reflecting an observation he made in an earlier book, High Output Management: An organization does not live by its members agreeing with one another at all times on everything. It lives instead by people committing to support the decisions and the moves of the business.

Finally, legitimation should involve more than just the members of the organization itself, but must also take into account those individuals who, in the course of their own business activities, interact with the organization (Grove, 1996): The debate should involve people outside the company, customers and partners who not only have different areas of expertise but also have different interests.

Such an emphasis on opinion challenges the very ontological foundations of knowledge management. As we have seen, those foundations have a clear role for the concept of belief; and epistemology, as a discipline, has devoted considerable attention to dealing with that concept. However, opinion is not the same as belief, which is one reason that Wenger (1998) emphasizes that “shared beliefs . . . are not what shared practice is about.”[3]. This is not to say that knowledge management systems are not capable of dealing with opinion (at least when that opinion is documented); but they do tend to treat the documentation of opinion like any other form of documentation, which can be a dangerous category error when the contents of documents have a major impact on the activities of an enterprise (which is just about always). In summary, we see that Grove’s own words reflect the same issues concerned with negotiating differences of understanding through interaction that form the basis of Giddens’ theoretical model and that, at least as far as Grove’s experiences at Intel are concerned, Giddens’ dimensions are as interdependent as Giddens claims them to be. Technology support for interaction management If we are to address questions of how technology can contribute to social interaction, we must begin by recognizing that interaction is not strictly limited to communication, nor should it be viewed through the technical lens of exchanging information. The different perspectives on interaction that we have examined provide vivid examples of how inadequate and confining a “technocentric box” can be. Furthermore, the three dimensions of Giddens’ structuration theory demonstrate in very explicit terms that interaction cannot be limited to communication.

Interaction management

347

BPMJ 9,3

Nevertheless, technology can provide means by which we can extricate ourselves from our “techno-centric box”. Furthermore, when we try to identify what these means can be, we discover that they can be characterized, by once again turning to Giddens’ dimensions of interaction. Let us now examine each of the dimensions in terms of how technologies can be harnessed to facilitate it.

348

Signification As was illustrated by the experiences at Intel, signification is rarely a simple matter, particularly in mission-critical situations. Nevertheless, it is easy to succumb to the illusion that technology can simplify complex problems of signification. Those susceptible to this illusion would do well to recall what Dickson (1981) dubbed “Mencken’s meta-law”; “For every human problem, there is a neat, plain solution – and it is always wrong”. As has already been observed, there is a tendency, encouraged by prodigious advances in search engine technology, to assume that the Internet is always there to provide the right answer, in as simple and compact a manner as possible, to the right person at the right time. However, rather than trying to build the Internet as an all-knowing agent that is always there for us with the right answers, we should think of it as the primary vehicle with which we learn how to deal with our most challenging problems of signification, pursuing John Seely Brown’s vision that the Internet can “become an incredibly powerful medium to unleash a culture of learning” (Corcoran, 2000). Once we recognize that understanding cannot be achieved through a sufficiently powerful search engine, where can we turn to honor Brown’s vision in practice? Davenport (1997) believes we should turn to a better understanding of how librarians (or, as he prefers to call them, “information staff”) help their clients achieve understanding. Davenport’s (1997) model is based on the goals and tasks of information staff members; and there are rich opportunities to identify how technology may facilitate those tasks and the achievement of the goals (Smoliar and Sprague, 2002). Domination The legacy of artificial intelligence has tended to cultivate the assumption that technology is best applied in the service of signification; but it can also serve the management of domination, particularly if, as was observed in the Intel example, such management has more to do with responsibility than with authority. For example, work practices may be mediated by conversational agents as a move away from authority figures. The capabilities of such artificial agents have been investigated by Churchill et al. (2000a). In addition, various forms of existing collaboration technologies, such as chat rooms and shared work areas, may be invoked to support the self-organization of communities and networks. Churchill and another set of colleagues have investigated how documents may serve as “boundary objects” (Star and Griesemer, 1989) around which such self-organization may take place through a technology that she calls “anchored conversations”, converting the sort of

spatial artifact Microsoft Office implemented for comments into a chat space area, thus enabling conversations to be conducted over specific (multiple) locales within the document (Churchill et al., 2000b). Another area of technology that needs to be reviewed in light of the issue of domination is workflow support. Grinter’s (2000) study of the use of workflow systems indicates that, in the general context of work practices, criteria for success are severely limiting: . “the developers [of the workflow system] understood and accepted the model of work”; . the workflow system “provided understandable and useful representations”; . “the ‘right’ work was automated”; and . “the corporation was supportive”. One way to interpret the experiences at Intel is that, if the criteria for the success of workflow systems are to be less limiting, those systems need to be reconceived towards the negotiation of capabilities and responsibilities. What is required is an account of what individuals bring to the entire work environment, touching on mission-critical issues such as the following: . Who has skills in which core technologies? . Who has the best intuitive understanding of the nature of the product? . Who has the best intuitive understanding of the types of markets in which the product may be involved? There is frequently a tendency to question the second of these items, based on the assumption that any understanding of a product should be based on a formal engineering specification; but, as Cook and Yanow (1993) have observed, such an intuitive understanding can be necessary for successful production. Legitimation Community opinion played a strong part in the role of legitimation in the Intel example, so opportunities for technology should reside in the support and management of community opinion, without disturbing priorities of privacy and tolerance. This can involve a variety of different approaches: (1) One consists in providing means for collecting opinion with minimal burden on the user community. Such collection may be implicit in user actions, such as tracking the frequency of visiting Web sites. However, in the interest of privacy, the anonymity of users should be respected when it is requested (or, indeed, treated as a default condition). (2) Once opinion has been collected, it should be visualized in ways that are likely to be meaningful to the community at large. Visualization is rarely a straightforward matter (Card et al., 1999). Indeed, if the user community is

Interaction management

349

BPMJ 9,3

350

sufficiently large and varied, it may be necessary to provide multiple forms of visualization for the benefit of different sectors of the organization. (3) In such cases it may also be necessary to invoke means of filtering opinion based on community membership, drawing on techniques such as those of collaborative filtering (Arnheim, 1996). However, it is important to bear in mind that any technology concerned with community opinion will only succeed if it is reinforced with appropriate investments in social capital (Putnam, 1995). Members of an organization are not going to use a technology that they do not trust, nor will they use it if they do not trust the people who are providing and encouraging use of that technology. The technology will only succeed if its usage is perceived as a behavioral norm. This means that it must be perceived as equally fair to all users; but it must also be perceived as furthering the co-destiny of the entire community. Conclusion Summary of results We may now return to the questions with which this paper began: (1) What does it really mean to manage people whose work depends significantly on what they know? What it really means is that the work these people do depends heavily on both their capacity and opportunities for social interaction. It does not mean that such people must “convert” their “tacit knowledge” to “explicit knowledge” and then communicate the resulting explicit knowledge to other work groups. Rather, it means that the work environment must exploit the full range of interactions that constitute the basis of Giddens’ structuration theory. (2) How may technology facilitate this management task? Technology can deal with more than problems of signification. It can deal with domination by providing channels for interaction that circumvent impediments that tend to be associated with problems of authority. It can deal with domination by managing the accumulation and distribution of community opinion. The resulting approach to management is sufficiently different from the connotations of knowledge management that it deserves a name of its own: interaction management. Contribution The primary contribution of this paper is the justification of interaction management as a new and necessary discipline, reinforced by evidence of the significant role that interaction plays at work in a case study. Limitations The primary limitation of dealing with interaction management as a new discipline is essentially the same limitation that continues to trouble

knowledge management: a lack of sufficient quantitative evidence of the true value of the discipline. Unfortunately, the more “social” a situation gets, the harder it is to collect such evidence. Thus, for example, in discussing management issues associated with service productivity (an area that has received considerable attention from promoters of knowledge management), Quinn (1992) observed that “standard (ROI or PV) financial evaluations were often inappropriate or useless”. Evaluating the impact of interaction management is likely to be even more problematic, given the increase of subjective variables that are involved. Workplace attitude is also likely to be a major source of limitations. While there tends to be little argument over the limitations of scientific management (Taylor, 1911), there remains a lingering desire for a “rule book” that both managers and workers can follow. The tendency to think in terms of such rules is the “dark side” of Giddens’ ontological security. However, just as it has already been observed that people are not going to use a technology that they do not trust, they will be equally suspicious of any new approach to management. When Mao Tze-Tung introduced his “policy of letting 100 flowers blossom” in his 27 February 1957 speech in Beijing, who would have anticipated the Cultural Revolution of 1966 that would mow down those flowers so brutally and systematically? This is why a commitment to interaction management is only going to work if it is reinforced with an equally committed investment in social capital (Putnam, 1995). Opportunities for future research We have already seen that all three of Giddens’ dimensions of interaction may be supported by innovations in information technology. Taken as a whole, we may wish to call those innovations examples of “social computing”. This would basically be a vision of software technology that supports the realization of a seamless network of people, devices, and information (Smoliar, 2002). However, in light of the limitations discussed above, any implementation of social computing must take into account the integration of social values into document usage. In addition, the concept of the seamless network opens the door to a wide spectrum of research opportunities covering the entire grid of interactions that can take place along the different pairings of people, devices, and information. Notes 1. Nonaka’s book on organizational information creation was published only in Japanese but is cited in The Knowledge-Creating Company (Nonaka and Takeuchi, 1995)]. Fritz Machlup was investigating the economic significance of knowledge in this same time frame. (By way of historical perspective, Friedrich Hayek had published his view of “the price system as . . . a mechanism for communicating information” (Hayek, 1945).” Machlup (1980-1984) had

Interaction management

351

BPMJ 9,3

352

planned a massive eight-volume work to document his findings. Sadly, he died after only three of these volumes had been published. 2. The four “modes of knowledge conversion” that constitute Nonaka and Takeuchi’s (1995) SECI model (socialization, externalization, combination, internalization) all basically reduce to different perspectives on how communication can take place. 3. As a matter of fact, Wenger observes that what Grove calls “broad and intensive debate” is basically the concept of discourse introduced by Foucault (1976), which Wenger (1998) describes as “a social, interactive resource for constructing statements about the world and coordinating engagement in practice.” References Arnheim, L. A. (1996), Summary of Proceedings in Collaborative Filtering Workshop, March 16, Berkeley, CA, available at: www.sims.berkeley.edu/resources/collab/collab-report.html Barthes, R. (1973), Elements of Semiology, Lavers, A. and Smith, C. (Trans.), Hill and Wang, New York, NY. Bobrow, D.G. (1991), “Dimensions of interaction”, AI Magazine, Fall, pp. 64-80. Card, S.K., Mackinlay, J.D. and Shneiderman, B. (1999), Readings in Information Visualization: Using Vision to Think, Morgan Kaufmann, San Francisco, CA. Churchill, E.F., Cook, L., Hodgson, P., Prevost, S. and Sullivan, J. (2000a), “May I help you? Designing embodied conversational agent allies”, in Cassell, J. et al. (Eds), Embodied Conversational Agents, MIT Press, Cambridge, MA. Churchill, E.F., Trevor, J., Bly, S., Nelson, L. and Cubranic, D. (2000b), “Anchored conversations: chatting in the context of a document”, in CHI 2000 Conference Proceedings, ACM Press, New York, NY, pp. 454-61. Collins, R. (1998), The Sociology of Philosophies: A Global Theory of Intellectual Change, Harvard University Press, Cambridge, MA. Cook, S.D.N. and Yanow, D. (1993), “Culture and organizational learning”, Journal of Management Inquiry, Vol. 2 No. 4, pp. 373-90. Corcoran, E. (2000), “Brown’s law”, Forbes, March, available at: www.forbes.com/forbes/2000/ 0306/6506052a.html;$sessionid$IS0ABPQAAH4R3QFIAGWCFFQ. Davenport, T.H. (1997), Information Ecology: Mastering the Information and Knowledge Environment, Oxford University Press, New York, NY. Davenport, T.H. and Prusak, L. (1998), Working Knowledge: How Organizations Manage What They Know, Harvard Business School Press, Boston, MA. Descartes, R. (1985), “Discourse on the method”, in Stoothoff, R. (Trans.), The Philosophical Writings of Descartes, Cambridge University Press, Cambridge, Vol. I,. Dickson, P. (1981), The Official Rules, Tiews, K. (Illustrator), Dell, New York, NY. Drucker, P.F. (1964), Managing for Results, Harper, New York, NY. Foucault, M. (1976), The Archaeology of Knowledge, A. M. Sheridan Smith (Translator), Harper, New York, NY. Giddens, A. (1979), Agency, Structure, in Central Problems in Social Theory: Action, Structure and Contradiction in Social Analysis, University of California Press, Berkeley, CA. Giddens, A. (1984), The Constitution of Society: Outline of the Theory of Structuration, University of California Press, Berkeley, CA. Grinter, R.E. (2000), “Workflow systems: occasions for success and failure”, Computer Supported Cooperative Work, Vol. 9 No. 2, pp. 189-214.

Grove, A.S. (1983), High Output Management, Random House, New York, NY. Grove, A.S. (1996), Only the Paranoid Survive: How to Exploit the Crisis Points that Challenge Every Company and Career, Currency Doubleday, New York, NY. Hayek, F.A. (1945), “The use of knowledge in society”, The American Economic Review, Vol. 35 No. 4, pp. 519-30. Machlup, F. (1980-1984), Knowledge: Its Creation, Distribution, and Economic Significance, Princeton University Press, Princeton, NJ. Mannheim, K. (1985), Ideology and Utopia: An Introduction to the Sociology of Knowledge, Wirth, L. and Shils, E. (Translators), Harcourt Brace & Company, San Diego, CA. Nonaka, I. and Takeuchi, H. (1995), The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford University Press, New York, NY. Plato (1963), The Collected Dialogues of Plato Including the Letters, Cornford, F.M (Translator), Princeton University Press, Princeton, NJ. Poole, M.S. (1985), “Communication and organizational climates: review, critique, and a new perspective”, in McPhee, R.D. and Tompkins, P.K. (Eds), Organizational Communication: Traditional Themes and New Directions, Sage Publications, Beverly Hills, CA. Putnam, R.D. (1995), “Tuning in, tuning out: the strange disappearance of social capital in America”, Political Science & Politics, December, pp. 664-83. Quinn, J.B. (1992), Intelligent Enterprise: A Knowledge and Service Based Paradigm for Industry, Free Press, New York, NY. Redding, W.C. (1972), Communication Within the Organization: An Interpretive Review of Theory and Research, Industrial Communication Council, New York, NY. Smith, R.G. and Farquhar, A. (2000), “The road ahead for knowledge management: an AI perspective”, AI Magazine, Winter, pp. 17-40. Smoliar, S.W. (2002), “The value of social computing in implementing EDMS/ERS”, unpublished lecture delivered at DocuShowXtraordinaire, Fuji Xerox Singapore, March, p. 22. Smoliar, S.W. and Sprague, R. (2002), “Communication and understanding in decision support”, in Proceedings: International Conference on Decision Making and Decision Support in the Internet Age (DSI-Age). Star, S.L. and Griesemer, J.R. (1989), “Institutional ecology, ‘translations’ and boundary objects: amateurs and professionals in Berkeley’s Museum of Vertebrate Zoology, 1907–39”, Social Studies of Science, Vol. 19, pp. 387-420. Taylor, F.W. (1911), The Principles of Scientific Management, Harper, New York, NY. Wenger, E. (1998), Communities of Practice: Learning, Meaning, and Identity, Cambridge University Press, New York, NY. Yates, J. and Orlikowski, W.J. (1992), “Genres of organizational communication: a structurational approach to studying communication and media”, Academy of Management Review, Vol. 17 No. 2, pp. 299-326.

Interaction management

353

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

BPMJ 9,3

354

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

Experiences from knowledge management implementations in companies of the software sector Dimitris Apostolou Planet Ernst & Young, Athens, Greece

Gregoris Mentzas Department of Electrical and Computer Engineering, National Technical University of Athens, Zografou, Greece Keywords Knowledge management, Software, Knowledge workers, Networks Abstract The Know-Net knowledge management solution, that includes a theoretical framework, a consulting method and a software tool, is based on a knowledge asset-centric design that innovatively fuses the process or human approach with the product or content approach of knowledge management. This paper describes how Know-Net was applied in four companies of the software sector. It outlines how specific business areas such as an R&D unit, the bid management process and collaboration between geographically-dispersed teams can benefit from knowledge management. It outlines the role specific Know-Net components played within the transformation of existing business processes and structures, and provides key recommendations based on this experience.

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 354-381 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477939

Introduction The management of organisational knowledge has drawn the attention of academics, consultants and practitioners as a key lever for improving performance, boosting productivity and creativity and facilitating innovation in organisational settings. The methods, tools and the actual knowledge management (KM) implementations in various companies have mainly followed one of two perspectives, which in this paper we call process-centric and product-centric approaches (see, for example, Hansen et al., 1999; Ku¨hn and Abecker, 1997; Spek and Spijkevert, 1997): . The process-centred approach mainly understands KM as a social communication process. In this approach, knowledge is closely tied to the person who developed it and is shared mainly through person-to-person contacts. The main purpose of information technology in this approach is to help people communicate knowledge, not to store it. This approach is also referred to as the “personalisation” approach. . The product-centred approach focuses on knowledge documents, their creation, storage and reuse in computer-based corporate memories. This approach is also referred to as “content-centred” or “codification” approach.

The Know-Net KM solution[1] (Mentzas et al., 2000) is built around a Knowledge knowledge asset-centric approach that is a unique fusion of the knowledge-asmanagement a-product (content) and knowledge-as-a-process (context) perspectives to KM. implementations This paper gives a brief overview of the Know-Net solution, highlights the results of the four implementations in the software sector and provides recommendation based on the practical experiences and lessons learned from 355 these initiatives. A strategic perspective to knowledge assets The main motivation of the Know-Net solution was the development of a KM platform that would integrate the two approaches, not only from a technological perspective, but in all constituents of the solution (software tool, consulting methodology, measurement system, etc.). A conceptual, theoretical foundation was needed; one that would guarantee this fusion and that would be underlying every aspect of the solution. Both the process and the product-based approaches aim to support the identification, managing and leveraging of knowledge, through better managing of the organisation’s knowledge assets. In essence, KM is working to better manage the content, quality, value and transferability of knowledge assets. The focus of the Know-Net approach is on knowledge assets as the critical strategic resources of the firm. Such a focus is in line with the recent trend in the strategic management literature to leverage the internal resources of the firm in order to create value. The following paragraphs briefly outline two commonly used approaches to strategic management: the competitive strategy approach and the resource-based view, and proceed with analysing the knowledge asset-centric approach as the proposed theoretical solution to integrating the product- and process-centric approaches. The competitive approach to strategy The competitive strategy perspective posits that competitive advantage is derived from the firm’s environment, more precisely from the industry in which it competes (Porter, 1980; Spender, 1996; Belohlav, 1996; Teece et al., 1997). The competitive strategy perspective can be seen as an “outside-in” approach to understanding the basis of competitive advantage. This approach stresses that an effective competitive strategy calls for the firm to take offensive or defensive action to create a defendable position against competitive forces. Thus, industry conditions determine the rules of the game when it comes to the nature of competition and the strategies available to firms. A major assumption of this perspective is that all relevant, industry-specific resources are distributed homogeneously and are perfectly mobile. That is, the basis for competition is not derived from the firm as such, but rather from the characteristics of the industry. Consequently, superior performance in an

BPMJ 9,3

356

industry or strategic group results from this environmentally-derived competitive advantage. A second assumption of this approach is that both demand and supply conditions are known, and, consequently, market conditions are relatively stable. In a stable demand environment, competition is viewed as a zero-sum market share rivalry between existing and potential firms. Also, because the demand side of the market is known or predictable, competitive advantage stems from the supply side. Consequently, selecting the competitive advantage that yields the highest levels of economic performance requires intensive analysis of the industry structure, of suppliers, buyers, new entrants, and threats from substitutes, as discussed in depth by Porter (1980) and other authors within the industrial organization paradigm. The resource-based view of strategy From the resource-based perspective the firm is seen as a portfolio of resources. What a firm can do to create competitive advantage is not simply a function of the opportunities in the environment (industry) but also of what resources the firm can assemble (see, for example, Wernerfelt, 1984). One of the principal insights of the resource-based view (RBV) is that not all resources are of equal importance or possess the potential to be a source of sustainable competitive advantage. Much attention has focused, therefore, on the characteristics of advantage-creating resources and various approaches have been followed in analysing the characteristics of advantage-creating resources. For example, Barney (1991) proposes that advantage-creating resources must meet four conditions, namely, value, rareness, inimitability and non-substitutability, while Grant (1991) argues that the levels of durability, transparency, transferability and replicability are important determinants. Although the resource-based view recognizes the importance and role of knowledge in firms achieving a competitive advantage, we argue that the RBV does not go far enough. Specifically, the RBV treats knowledge as a generic resource, rather than having special properties, and subsequently, does not make any distinction between different types of knowledge-based capabilities. Intrinsic characteristics of knowledge assets Knowledge assets are different from other organizational resources (see, for example, Glazer, 1991; Day and Wendler, 1998). Knowledge assets are not easily divisible or appropriable. This means that the same information and knowledge can be used by different economic entities at the same time. Moreover, knowledge assets are not inherently scarce (although they are often time-sensitive). This implies that they are not depletable.

Knowledge assets are essentially regenerative. This means that new Knowledge relevant knowledge may emerge from a knowledge-intensive business process management as additional output, besides products and services. implementations Knowledge assets may not exhibit decreasing returns to use, but will often increase in value the more they are used. This characteristic is of crucial importance for senior management (see, for example, den Hartigh and 357 Langerak, 2001). Most assets are subject to diminishing returns, but not knowledge. The bulk of the fixed cost in knowledge products usually lies in creation rather than in manufacturing or distribution. Once knowledge has been created, the initial development cost can be spread across rising volumes. Network effects can emerge as knowledge assets are used by more and more people. These knowledge-users can simultaneously benefit from knowledge and increase its value as they add to, adapt, and enrich the knowledge base. In traditional industrial economics, assets decline in value as more people use them. By contrast, knowledge assets can grow in value, as they become a standard on which others can build. As knowledge assets grow, they tend to branch and fragment (Drucker, 1997). While knowledge assets that become standards can grow more and more valuable, others, like expiring patents or former trade secrets, can become less valuable as they are widely shared. A successful company must therefore continually refresh its knowledge base. The rapid and effective re-creation of knowledge can represent a substantial source of competitive advantage. Conceptual foundation of Know-Net Our perspective is that “knowledge assets” can be human – such as a person or a network of people, structural – such as a business process, or market – such as a brand name of a product. Naturally the product-centric approach is more concerned with accessing and organising knowledge assets, while the process approach makes direct connections between the organisational knowledge assets – both explicit and tacit. Both approaches, however, are using some form of knowledge representation as a means of packaging and transferring knowledge, either from a person to a system and vice versa, or between people. If we define “knowledge objects” as the means of representing knowledge, then the following statement outlines the relation between knowledge assets and knowledge objects: A knowledge asset creates, stores and/or disseminates knowledge objects

For example: . A person is a knowledge asset that can create new ideas, learnings, proposals, white papers (knowledge objects). . A community of interest is a knowledge asset that can create new ideas, best practices (knowledge objects).

BPMJ 9,3

.

.

358

A process is a knowledge asset that can create and/or store and disseminate best practices, company standards, R&D material (knowledge objects). A vision is a knowledge asset that can create a new mission statement, strategic plan, goals (knowledge objects).

A knowledge object represents the information required to be processed by humans and transformed into knowledge (see Figure 1). Knowledge derives from information through knowledge-creating activities, that take place within and between humans. Typical knowledge-creating activities include (Davenport and Prusak, 1998): . Comparison: how does information about this situation compares to other situations known? . Consequences: what implications does the information have for decision and actions? . Connections: how does this bit of knowledge relate to others? . Conversation: what do other people think about this information? The knowledge objects aim to facilitate and leverage such knowledge-creating activities, by providing to humans the information need. A knowledge object has the following characteristics: (1) It acts as a catalyst, enabling the fusion of knowledge flows between people, with knowledge content discovery and retrieval, through technology. That is to say, a knowledge object acts, amongst other things, as the primary connecting node for all key components in a KM system – strategy, people, process, content, technology – “the KM glue”. (2) It facilitates the knowledge transfer from person to person, or from information to person.

Figure 1. How the “knowledge object” relates to the process- and productcentric approaches

(3) A knowledge object is created and maintained by a KM process. Knowledge (4) A knowledge object is used to search, organise and disseminate management knowledge content. implementations We conclude that the knowledge assets and knowledge objects are the common unifiers of the holistic Know-Net KM solution that incorporate and integrate 359 process and content. We have used these concepts as the “resultant manifestation” in the design of the Know-Net solution that fuses the processcentric approach with the product-centric approach. Integrating the product- and process-centric approaches in the Know-Net solution The practical application of the approach described so far is the Know-Net KM solution that comprises three components: (1) A holistic conceptual framework that can be used by managers as a roadmap for ensuring integrity of the KM effort. (2) A KM methodology that helps organisations define and document their KM strategy, audit and design business processes that enhance and facilitate corporate learning, establish related organisational roles, facilitate knowledge sharing between people in the organisation, and explicitly measure and evaluate the quality and business value of the organisation’s intellectual capital. The method is designed to be modular so that an organisation can choose to start at different levels depending on its readiness, needs and requirements. In Stage I of “strategic planning for KM” an organisation determines: the vision and readiness for a KM initiative; and the scope and feasibility of the project. The main steps of Stage I are: . provide leadership; . link KM strategy with corporate strategy; . perform knowledge analysis; . assess risk and change readiness; . develop the case for KM; . obtain top management approval. In Stage II of “developing the knowledge organisation” the structure and the design of a holistic solution (that covers processes, people and technology) are iteratively developed, tested and reviewed. Stage III is the company-wide implementation of the KM initiative, while the measurement part of the method aims to provide consistent support for measuring the creation, sharing and use of knowledge assets within the company.

BPMJ 9,3

360

(3) An intranet-based tool that supports the collection and categorisation of internal and external information, the re-use of stored knowledge using flexible and customisable knowledge navigators and advanced search mechanisms that include keyword-based as well as concept-based searching (the latter supported by a graphical visualisation of the concepts organising the information space), and the collaboration via online workspaces that allow people to work together from different locations. The three components of the Know-Net total KM solution – i.e. the framework, the method and the tool – have clear and consistent interdependencies which exploit the knowledge-asset-centric nature of the solution and facilitate the amalgamation of the process-centric and product-centric perspectives to KM. Figure 2 highlights the overall interdependencies of the Know-Net framework, method and tools. The Know-Net framework (Mentzas, 2000) can be used as an awareness tool as well as for developing a common language among the people of a company. The framework is useful as an enabler for discussing which are the critical knowledge areas, which ones are under-developed and should be further enhanced, and which are already valuable and should be protected and cultivated. In addition, the elements of the framework (strategy, structure, processes, systems) help share a first draft picture of the level of knowledge awareness within the company. Finally, the first discussions about KM should

Figure 2. Interdependencies of the Know-Net framework, method and tool

focus on the various levels for leveraging knowledge assets (i.e. individual, Knowledge team, organisational and inter-organisational levels) in order to help identify management possible areas for intervention. These rough analyses can provide useful input implementations for the application of the Know-Net method (especially Stage I: strategic planning for KM), as well as facilitate the smooth introduction of the concepts and terms used in the Know-Net tool. 361 The application of the method may be tightly linked to the use, customisation and roll-out of the Know-Net tool. In any case, however, each one of the two (method and tool) is also self-sustained and can be independently applied – actually, the method has already been applied to companies without the use of the tool. The coupling between the method and the tool is mainly accomplished through the eight modules of Stage II (developing the knowledge organisation) of the Know-Net method, as outlined in Table I. Each module of Stage II is a self-contained, value-adding entity and therefore not all modules are mandatory in a KM implementation. Ideally, however, just as the knowledge assets are the main unifiers of our approach, module 7 (develop the knowledge asset schema) acts as the frame of the Know-Net method that is being constructed with input from the “audit” modules 1, 3, and 5, while it supports the consistent execution of the “design/implement” modules, 2, 4, and 6. All “audit” modules, among other issues, aim to identify in detail the knowledge assets, and corresponding knowledge objects and their attributes. Module 1 (analyse business processes) for instance, produces processes maps that depict key knowledge assets that are being used or created in selected business processes. Module 7 (develop the knowledge asset schema) collects this information, along with similar information, from modules 3 and 5, arranges possible overlappings, logically groups content, and creates the formal schema (knowledge asset schema) on which the “design/implement” modules are based. Consider another example: module 4 (leverage knowledge networks) designs and organises communities of practice and interest around the core knowledge assets of the organisation and proposes the already specified knowledge objects as units for knowledge representation within these communities. Table I broadly describes how each module relates to the integration of the process- and product-centric approaches and to the functionalities and characteristics of the Know-Net tool. Case summaries of the Know-Net application in IT services companies In the following section we present a brief description of Know-Net implementations in four companies: Delta-Singular, MDA, AlphaNova and Debus IT. The objective in all four cases was to undertake KM initiatives at the business unit, function, or operational process level, where they expected to have the most immediate impact on their performance. Within this context, the

Design of formal business processes and roles for KM (process view) Incorporation of knowledge objects, relations, and attributes (metadata, classification, indexing, etc.) to business processes, e.g. who creates a knowledge object, which metadata should be used, how is the indexing being done, etc. (product view) Detail identification of knowledge assets and objects involved in informal communication/collaboration of employees (product view) Comprehension of patterns of informal knowledge flow, collaboration habits and requirements for improvement (process view)

2. Leverage This module helps the enhancement knowledge in of KM within existing business business processes and the design of new processes business processes and corresponding roles for managing specific knowledge assets

This module helps analyse the informal flow of knowledge within networks of people in the organisation

3. Analyse knowledge networks

Detail identification of knowledge assets and objects involved in business processes (product view) Detail identification of the knowledge flows within business processes. (process view)

This module helps the knowledge audit of business processes. In order to understand how business processes, people, systems and content are related in order to reveal who needs what knowledge and when they need it

Relation to process and product integration

1. Analyse business processes

Table I. Relation of KnowNet methodology modules to the product- and process-centric views and support by the Know-Net tool

Description

Provides requirements for the creation/customisation of KM applications. Emphasis here is on the groupware support of the tool, e.g. applicability of specific collaboration requirements such as video/audio conferencing, customisation of community/team support applications of the tool, etc. Profiles people and roles involved in networks (to be used in personalised facilities of the tool) Provides a “first cut” of the knowledge objects that correspond to the knowledge assets (continued)

Definition of access rights for knowledge objects/applications Customisation of KM applications to support business processes

Provides requirements for the creation/customisation of KM applications (including functionality, user rights, etc.) Profiles people and roles involved in processes (to be used in personalised facilities of the tool) Provides a “first cut” of the knowledge objects that correspond to the knowledge assets

Relation to Know-Net tool

362

Module

BPMJ 9,3

This module helps the design and stimulation of knowledge networks using a set of “non-management techniques”. The module addresses CSFs factors for the network, content quality assurance issues, motivation and reward issues, etc.

Assesses the current state of IT in the organisation and identify existing information sources

Presents the technology element in KM This module guides the design of knowledge asset schema that comprises: † The knowledge object store (for the formal knowledge organisation) † The knowledge ontology (for conceptual/logical structuring organisation of knowledge objects) Synthesises and integrates all changes done in different modules in one working solution, seamlessly incorporated in the existing business environment

4. Leverage knowledge networks

5. Analyse the technology

6. Leverage the technology 7. Develop the knowledge asset schema

8. Integrate the KM architecture

Description

Module

Not applicable

Product view only, as discussed in Chapter 3

Not applicable

Design of informal networks (communities of practice/interest) in support of KM (process view) Incorporation of knowledge objects, relations, and attributes (metadata, classification, indexing, etc.) to knowledge networks, e.g. who has ownership of which knowledge object in the network, which metadata are suitable for the network, filtering rights, etc. (product view) Mostly related to the product view, looks at knowledge artefacts that are stored in information systems

Relation to process and product integration

Not applicable

Direct mapping to the knowledge server. During this module the object store and the ontology are customised and configured for the organisation

Examines the applicability of the Know-Net tool from a technical perspective (network bandwidth, geographical distribution, etc.) Identifies existing information sources within and outside the organisation Not applicable

Definition of access rights for knowledge objects/applications Customisation of KM applications to support knowledge networks. Emphasis on groupware applications

Relation to Know-Net tool

Knowledge management implementations 363

Table I.

BPMJ 9,3

364

Figure 3. Focus of KM in the four IT services companies

application of the Know-Net solution in each of the four companies focused on a specific area of the software development cycle (either in customised software projects or in the development of packaged software solutions; see Figure 3). Debus IT Debus IT specialises in the development, customisation and training of enterprise resource planning software and systems. It is a Czech company with Dutch management and investment. The specialist nature of the company’s business requires that it gathers technical, marketing and other commercial information from diverse sources, to reconfigure and reuse that information for commercial and competitive advantage. Debus IT has acknowledged as critical factors for future success, the potential ability to speed-up the development and related cycle times through the application of KM principles. To do so there was a need for improved communication and collaboration between Debus IT and its customers and improved quality of Debus IT training courses. The KM solution created and applied at Debus IT consisted of a number of inter-acting elements, all of which have been applied to the area of creating “customised training courses”. Debus has had as its major objectives the ability to improve its communication and collaboration, and to reduce the cycle times associated with the delivery of customised training courses. This has been achieved through the following: . The effective use of the contacts database has enabled the company to log key skills, knowledge and competencies. . The application of ideas capture and management has enabled the company to aggregate critical skills and knowledge, leading to the creation of best practices and best knowledge databases, thereby leading to the re-use of knowledge. . The knowledge asset measurement techniques incorporated within Know-Net allowed Debus to appreciate the value of its KM initiatives.

A distinct marketing advantage was gained as Debus has the potential to Knowledge deliver better quality client/business solutions through the ability to access and management evaluate a greater number of technical options, and the ability to access implementations information from a wider and more diverse range of sources. Furthermore, the ability to integrate existing databases, marketing and sales records and information has been critical. Through having a central repository of 365 marketing and sales intelligence the company is able to lever the business generation process. Solution highlight – introducing KM to the company. By applying the strategic planning stage of the Know-Net method, a strong business case was developed to support and define the KM strategy for Debus IT. The key to successfully implemented KM in Debus was to ensure that the KM strategy was able to manage knowledge to enhance product and service capability. Debus realised that as a consultancy/development company, it is driven by continuous market requests for technology and business knowledge. This knowledge is very dynamic, diverse and constantly expanding, therefore there was a clear need for a mechanism to collect, retrieve and share knowledge throughout the whole company. Leadership. Debus IT has successfully gained leadership support for the KM initiative. Management was educated on the need and the benefits of KM to gain buy-in from the top, before being presented at a company level. An initiation team, consisting of members from different departments, was created to introduce the KM solution and methodology to the rest of the company. The focus of these meetings and presentations was to ensure that all employees gained a clear understanding of the KM concepts and how KM could be integrated into their daily work. Gaining leadership support at this early stage made the implementation of the KM initiative a smoother transition and more acceptable by the company. Linking KM strategy with corporate strategy.Taking into account Debus IT’s mission to be both a technology and consultancy oriented company and its objective to remain profitable in the emerging markets of Central and Eastern Europe, it is vital that it can constantly provide valuable solutions to meet market needs; a KM strategy needs to be developed in line with this. Based on this, the critical success factors (CSFs) of the company have been identified as follows: recruitment and retention of the right people; business generation; new product development; management of technical knowledge. Perform knowledge analysis.A knowledge orientation analysis was undertaken with the results shown in Figure 4. The knowledge orientation matrix shows that Debus was quite “knowledge aware” in terms of KM. It had management support, a knowledge sharing culture and some means to gauge how KM benefits the company. In terms of infrastructure, some level of strategy, processes and systems existed; however, there were no clear roles and responsibilities defined within the company to support these. The company as

BPMJ 9,3

366

Figure 4. Knowledge orientation matrix for Debus IT

a whole was accustomed to sharing knowledge, not just at an individual or team level, but also at an organisational level. Assess risk and change readiness. As a company, Debus IT was not resistant to change and realised that to smooth the path on implementing KM, good preparation for the base of change was required to improve and build on the existing processes, rather than going for a radical change. In order to introduce and smooth the implementation of KM, a gradual process of change was undertaken, through education, by first gaining the acceptance from management, then the project/department leaders. Once these people were on board the company staff were involved. Debus continuously conducted education sessions for the whole company throughout the implementation stage. The change was also made more acceptable as Debus involved the whole company in KM and it was introduced almost simultaneously to all company staff. It was identified that the impact of KM would affect the majority of people within the company, owing to the impact on existing information flow and processes. Debus ensured that any impact would be minimised through management support and training. From applying Stage I of the KM method to Debus, a strong business case was put forward to senior management for the need for KM. In particular, it

focused on the need to improve the existing tools and techniques, and initiatives to improve the existing business processes. Delta-Singular Delta-Singular specialises in standardised business software covering market requirements in financial, accounting, distribution, logistics, retail, human resource management, and e-commerce systems. In addition, the company provides solutions integration in large-scale projects for public utilities, telecommunication, aerospace and defence, transportation, and public administration organizations. The company operates in all markets of south and Eastern Europe, the Middle East and Asia market. One of the most valuable commodities for Delta-Singular is the knowledge created within the R&D unit. The innovation originating from the R&D unit drives to a high degree the evolution of the company’s products and services. Members of the R&D unit, as well as other employees from other units of the company, form communities of practice (CoPs) that are formed dynamically in order to resolve specific technical problems. There was a need to develop a strategy for learning from the team experience, and organise the knowledge created in a methodological manner. The challenge was to realise effective mechanisms for KM without imposing overheads to these highly dynamic, highly creative teams. The technical infrastructure had to be highly dynamic and adoptable, integrating with existing systems and information repositories. The solution focused on CoPs as a collaborative structure that facilitates the creation and transfer of knowledge. The project addressed not only the creation of new CoPs from the ground up, but also it built upon existing, informal, working networks. New roles, such as subject experts, have been assigned to take responsibility for providing expert opinion and identifying the knowledge assets stemming from the common effort. Mechanisms for information collection during CoP lifetime were put in place, ensuring quality and consistent documentation of knowledge. These mechanisms provided for knowledge transparency, i.e. ability to re-use knowledge identified in a specific project to other projects. Furthermore, the R&D unit put in place mechanisms and regulations for reviewing knowledge so that “aging” knowledge items are collected and disposed of. The Know-Net methodology was used to support the launch of viable communities, the creation of a relevant knowledge base, and the provision of “care and feeding” for the communities’ growth. One year after the launch of the initiative, the creators are positive about their work, since KM has become an integral part of the strategy of DeltaSingular. The KM infrastructure put in place allowed for: . Savings in cost, e.g. by re-using knowledge on how to tackle specific software development problems. . Increased returns, e.g. by maximising re-use and exploitation of software approaches, components and products.

Knowledge management implementations 367

BPMJ 9,3

368

Figure 5. Networks related to the singular R&D unit

.

The alignment of research and development with product development requirements, by recognising important trends and developments worldwide.

Solution highlights – the R&D community. Two levels of knowledge networking have been identified in the singular R&D unit as pertinent to the proposal preparing process (Figure 5): at the unit level, collaboration happens within teams that work in specific R&D projects and there is informal social networking based on common background; at the company level, networking takes place between employees from other units who are considered subject matter experts. Knowledge leveraging within the two types of networks was addressed in two phases. At the R&D unit level, during the first phase, the team was organised around a common repository that was used as a place where information regarding the unit’s work was stored and maintained. A procedure was adopted for users to deposit information that could be later retrieved from colleagues in accordance with their access rights to the document repository. A separate procedure was adopted for team members to ask questions on matters that were not found in the document repository. In this case, they sent their question to an e-mail account where the initiative coordinator assessed the question content and, if considered appropriate, it forwarded it to the subject matter expert who was responsible for that thematic area. In order for the mechanism to be of real value, a number of steps had to be taken before installing the process. These steps included the identification of the Delta-Singular universe of knowledge and the selection of a group of knowledge providers who would be responsible for answering questions in their specific thematic area. Department heads identified the Delta-Singular universe of knowledge within their departments and this knowledge segmentation was used for recognizing an expert in each field (i.e. a subject matter expert). Once a question had been submitted to the e-mail account the

initiative coordinator forwarded it to the appropriate subject matter expert. The Knowledge expert responded to the initiative coordinator who had the responsibility of management forwarding the response to the person who asked the question and implementations simultaneously he/she updated the knowledge repository by posting the item in question or by making public the question published along with its response – in a frequently asked questions format. The initiative coordinator maintained 369 a log-keeping mechanism that accounts for all knowledge requests submitted. The subject matter expert provided descriptive keywords that were used from the initiative coordinator as file metadata for storing the information at the knowledge repository. The initiative coordinator updated the request database with the question submitted and its response according to the expert provided keywords. Finally, the bookkeeping mechanism was updated accordingly (i.e. response supplied within the pre-set time limit, etc.). Based on the information stored in the database during the first phase, the second phase was implemented in the following steps: a new knowledge-brokering mechanism was established and the information recorded in the database was made available through a company-wide intranet that made use of the Know-Net tool. The knowledge-brokering system utilized in the second phase abides by the same general characteristics as those adopted during the first phase. It is the platform functionality offered by the Know-Net tool that enhanced user access to the new corporate portal. The systems general characteristics are briefly outlined as: (1) The knowledge stored in the database is placed in a company-wide intranet that is implemented with the Know-Net platform. (2) It is categorized according to knowledge assets identified through the implementation of the Know-Net method. (3) The identification of all subject matter experts along with their area of expertise has become available through the corporate portal. (4) Users who do not find the response to their request in the corporate knowledge base are encouraged to submit it directly to the appropriate subject matter expert via e-mail. (5) Users are advised to consult first the knowledge base before addressing questions to the subject matter experts. (6) Responses not existing in the knowledge base are deposited to it from the subject matter experts who have the responsibility for maintaining the knowledge base content. (7) The need for a bookkeeping mechanism has been eliminated. Three roles have been assigned to the network members in relation to their normal activities. These roles include the author, who is a subject matter expert and is responsible not only for providing expert opinion, but also for

BPMJ 9,3

370

identifying the knowledge assets stemming from the common effort. All information collected is delegated to her/him, who is in charge of documenting it according to pre-specified quality, consistency, and design standards. Apart from authors there are also viewers, who have the right to examine the documents deposited within the common workspace and receive messages every time content of their interest has been updated. When they feel that certain pieces of information should be included in the repository they forward them to the author who is responsible for the specific thematic area and she/he decides on publishing the document provided. There is also a system administrator who is responsible for maintaining the system integrity and ensuring its availability to the users. The adopted approach facilitates the dynamic re-evaluation of the firm’s knowledge assets. It introduces the notion of value-added processes that create experience (i.e. knowledge assets) to the company personnel. These elements are code samples, deliverables, technical troubleshoots, case studies, designs, and methodology components that are applicable in various projects. Making these knowledge assets identified in a specific project re-usable to other projects requires that the documented material is transparent. The technical infrastructure implemented provides the technical means so that stored items are indexed and retrieved using a full-text search option to retrieve content and any associated properties (metadata searching). In addition, a semi-automatic categorization process assists users in depositing and retrieving documents by associating them with specific categories that group similar documents. The same process was proven immensely useful when the documents already stored in the corporate file-structure were migrated to the Web storage facility. It provided an automatic importation mechanism that expedited the migration process. Finally, a subscription service notifies users about new, or updated, information on topics that match their interests and have found their way into the corporate depository. This mechanism works through a user profile where the independent user declares her/his preferences.

MDA MDA operates in Turkey and specialises in DBMS application development. MDA is a member of the ORACLE business alliance program as well as the value added reseller of the PROGRES corporation. MDA is organised to deliver turnkey systems, and from this perspective it portrays a company which can be described as a systems integrator. MDA also provides IT consultancy services. MDA wanted to improve the effectiveness of the bid preparation process. The company has been experiencing the “wheel re-invention” syndrome all too often during bid and proposal preparation. They needed a system to improve collaboration and organisation of bid work.

The most important areas that MDA focused on were the effective capture Knowledge and use of contacts from outside the company and the centralisation and management collaboration on new bids. The bid management system application of the implementations Know-Net solution was utilized. This allowed for the effective capturing of contacts and it supported employees to work and collaborate centrally on bids. Furthermore, the ideas capture component was utilized. This allowed for new 371 ideas on the best way to carry out the bid process to be logged and processed effectively. Also installed at MDA are best practices, best knowledge and a knowledge assets/objects directory. The system put in place allows MDA to quickly match the requirements of its customers to the abilities of the staff and specialist associates across the strategic partnerships. Furthermore, the company has improved the bid preparation process effectively, reducing time to prepare a bid, reducing time to locate a relevant bid, improving the success rate of bids, and reducing time to train new employees participating in bid preparation. Solution highlights – the “bid preparation” process. By applying the business process analysis module of Stage II of the Know-Net method, we were able to identify the critical knowledge requirements of the bid preparation business process. Important information needed include financial data related to the funding of the project outlined in the bid, people and external contacts-related data, bid-specific information such as deadlines and official documentation, and information about previous bids that could assist in the preparation of the current bid. In considering an improved bid process, we identified the following key areas for improvement: Knowledge about interest rates. Knowledge about interest rates and the accurate forecast of interest rates are critical to this process, because they determine the safety and feasibility of borrowing money. In order to help with this problem area, Web links were set up in the MDA Know-Net implementation to the best resources and knowledge bases on interest rates and interest rate forecasting. These include the national Central Bank or Foreign Exchange company sites that are being accessed regularly. Speed of suppliers’ response.The speed of response by suppliers to a bid was usually too slow for the process to be carried out effectively. Suppliers put together the technical and marketing documents, and these can take a long time to arrive. After they arrive, MDA often does not have enough time to study and prepare answers to specifications to the quality that they want to. To help speed-up of this communication with suppliers with this issue, a collaborative work area incorporating discussion forums and a central place for working on bids was utilized. The objective was to keep all of the work and communication with suppliers in the one space and increase the speed of supplier response. The common repository can be accessed easily through a Web browser at any time, so suppliers can quickly access the main work area and current discussions on bids.

BPMJ 9,3

372

Figure 6. “Bid preparation” business process

Reusing best practice knowledge. In MDA there was no organised reuse of knowledge in the bid preparation process, so that previous bids and experience are available in a codified, searchable format. To ensure that the company could start to manage its knowledge effectively in the bid process, a “best knowledge” process was introduced. The effective reuse of important knowledge in the process required a new role within the company; for this to be effective a knowledge manager needed to be appointed and trained in the new KM processes. The knowledge manager oversees the harvesting of knowledge from within the system and ensures that it is transformed into best knowledge for effective reuse in the future. Harvesting is done by ordinary knowledge workers, who also make nominations for the best knowledge base. Figure 6 depicts the “bid preparation” business process of MDA. The grey nodes are the knowledge-leveraging steps that have been introduced after the

Know-Net implementation. Below we describe these knowledge-leveraging Knowledge steps. management At the start of any new bid process, the knowledge worker should review the implementations best knowledge process for the current best available codified knowledge to conduct the bid process. If there is specific information/knowledge that the knowledge worker is looking for, then they can use the search function to 373 search the databases, or the entire system. The point of this is to encourage the knowledge worker to access codified knowledge, rather than tie up the time of the experts. If none of these methods can help the knowledge worker, as a last resort they can access the best people/expert locator database to contact the relevant expert. The next new step in the process is a “bid/no bid” decision step (step 4), just after the bids are reviewed. The advantage of this step, at this early stage, is to prevent any further work being carried out if the bid does not fall within the capability of MDA as a company (e.g. technical capability, financing, and so on). It is important to capture the reasons as to why the bid was not accepted, so that, over time, a checklist for bid acceptance can be defined, so that at the review bid (step 3) a formal and knowledgeable process for decision making does not rest on the individual, but on the experience of all. The knowledge worker now has the ability to consult external Web links for interest rates information (step 10) and is able to make more informed decisions. It is vital that MDA incorporates this to ensure that it knows about interest rate status for financing of bids. Virtual meetings with suppliers can be carried out in order to speed-up response from suppliers (step 11). This stage is critical in that it helps both MDA and suppliers to have a central place to discuss and work on specific bids, to ensure that a thorough response to the bids can be created. After the result of the bid is known and the bid has been completed, a review of the bid (step 16) carried out by the knowledge manager is a prerequisite for capturing learning and ideas (step 17). A key part of introducing KM to the process is the continual capture of learnings and ideas at every stage of the process. With any new initiative, it is not a natural process for individuals to submit ideas and learnings into the ideas and learnings repositories as and when they conduct their work, therefore, formalised reviews need to be conducted in order for these valuable learnings and ideas to be captured. At the end of each month, the knowledge manager is responsible for conducting a knowledge harvest of the KM system. The knowledge manager will review all ideas and learnings entered by the knowledge workers, as well as trawl the various discussion databases set up, to identify any valuable nuggets of information that may be useful ideas and learnings. The knowledge manager is then required to submit all the ideas, learnings and discussion nuggets (known as “best knowledge nominations” – step 18) to the best knowledge owner of the process to improve the current best knowledge process.

BPMJ 9,3

374

The best knowledge owner reviews each knowledge nomination (step 19) and makes a decision as to whether it is a valuable contribution to improve the current best knowledge process. If not, the best knowledge nomination is rejected or returned to the knowledge worker for further clarification. In addition to the “best knowledge” database, a “best people” (or “expert locator”) database was introduced to address the tacit dimension of knowledge. It is very difficult to be able to codify all knowledge in an organisation and some knowledge will inevitably remain tacit. The best people/expert locator captures the expertise and skills of individuals and allows the knowledge worker to search for experts in order to contact them directly for further information. The key for MDA is to always provide a current best knowledge bid process for knowledge workers to refer to. It is critical that the best knowledge bid process is always being reviewed and updated through the process described above. Knowledge workers need to actively contribute ideas, learnings and participate in collaborative discussions as part of their day-to-day work, in order for others to benefit accordingly. There must also be adequate systems and technology to support and facilitate the capture of ideas, learnings, collaborative discussions and best knowledge and best people databases, so knowledge workers can work effectively. AlphaNova AlphaNova is a global provider in customer relationship management software and solutions. AlphaNova is supporting companies to take advantage of the changing business environment by using alliances, multiple channels and customer focus to increase growth and profitability. Due to rapid expansion and high employee intake, it has become imperative for AlphaNova that the knowledge of existing employees should be harnessed and maintained in an effective manner. AlphaNova has offices in the UK, Greece and Cyprus; communicating, collaborating and sharing knowledge within offices is critical to the success of the company. The company was faced with the challenge to facilitate and encourage employees to share their knowledge and adopt new methods of work practices in terms of collaboration. The KM solution focused on the interaction and exchange of knowledge between the product design group and a selected team of people from the sales and consultancy groups. Using the Know-Net methodology, a formalised mechanism of managing flow of knowledge was put in place to ensure that validated and quality knowledge is available to those who need it. The firm also used the awareness creation modules of the methodology (that support education, workshops and presentations) to raise levels of awareness and knowledge amongst the employees of AlphaNova. AlphaNova utilized various components of the Know-Net tool to develop their KM infrastructure, such as: discussion forums and libraries (to store

commonly used information), AlphaNova Directory (containing information Knowledge about employees contact details and, more importantly, key skills), Ideas management Management, Learnings, Best Knowledge and, Knowledge Assets Directory. implementations The overall result was improved communications and collaboration owing to the use of collaborative workspaces, shared discussion forums and learnings. Solution highlight – measuring knowledge assets. AlphaNova applied the 375 Know-Net knowledge measurement approach in order to investigate how KM is benefiting the company, in terms of business measures looking at the success of the KM initiative; and knowledge asset measures. To derive the specific metrics and indicators we started top-down: we identified the organisation’s vision, strategy, CSFs, the business performance measurements for those CSFs and the key knowledge assets that should be associated with the CSFs. We then switched to a bottom-up process: starting with the knowledge assets identified for each of the key CSFs we developed the knowledge asset measurements for them. The three CSFs for AlphaNova are as follows: . improve the efficiency and effectiveness of cross-team communications; . improve the efficiency and effectiveness of inter-office communications; and . facilitation of such communication with an overall KM system. Figure 7 provides a summary of the knowledge assets identified for AlphaNova and links them to CSFs and measurements.

Figure 7. Linking knowledge assets and their measurements to critical success factor

BPMJ 9,3

376

Integration issues and concluding remarks The integration of the product- and process-centric approaches can be seen in the cases presented. The case of Delta-Singular demonstrates how both the process approach (i.e. dealing with communities of practice) and the product approach (i.e. dealing with knowledge as an object that is consumed or produced during R&D projects) can interoperate in practice. For instance, one can see how knowledge developed during an R&D project by a programmer can be codified so that it is accessible not only by another programmer in another project team, but also by a subject matter expert, and through him/her, to a wider community within the company. It is important to stress that our approach aims to ensure that both the process-centric view and the product-centric perspective can inter-operate, in the sense that they are not isolated from one another and that one can make use of and add value to the other. The aim is not – and should not be – for all organisations to try to excel in both approaches at equal proportions. Such an attempt may not be in line with the business environment and could be overwhelming (in terms of resources and organisational and cultural changes needed) for an organisation (Hansen et al., 1999). In practice, in all four cases, there is a varying emphasis on one of the two approaches. For instance, DeltaSingular focused on enhancing collaboration and knowledge sharing within informal networks of people that relate to the R&D unit (process approach), while MDA looked at the knowledge as an artefact produced during bid preparation (product approach). Nevertheless, the holistic approach followed during the design of the KM infrastructure ensures that any additional or future KM initiatives, no matter which approach they lean towards, will not be independent from the existing KM infrastructure. Practical experience from working with SMEs in the software development sector has revealed the need for a good practice guide for KM: a set of predefined and proven guidelines covering all aspects of KM (strategy, structures, people and technology). The Know-Net tool, for instance, contains a library of KM applications/processes covering most usual business areas. The tool could be enhanced with predefined knowledge assets, objects and their attributes, from which the user could select as appropriate. Furthermore, the method could include indicative structures of business processes that could serve as templates. Such a predefined KM infrastructure is particularly appealing to SMEs that do no whish to invest heavily in consulting and customised solutions. Some additional recommendations that were drawn from the four cases presented in this paper include the following: (1) Focus on exploiting knowledge sources. Fundamental to the principles of knowledge, asset management is the utilisation of existing knowledge sources. In keeping with this, Debus IT linked the KM tool with the existing software systems and a variety of external databases.

(2) Facilitate knowledge exchange in communities. Organizational structure, Knowledge particularly geographic dispersion, exerts an effect on the management management of knowledge. The basis for the transfer of rich, tacit knowledge is implementations personal relationships and geographical proximity. KM seeks to extend the immediacy of this natural contact beyond its boundaries by taking advantage of technology. In so doing, it extends the formerly untapped 377 and often invaluable, benefits of local groups, to form communities of practice, unhindered by geographical boundaries, as was the case in Delta-Singular and AlphaNova. (3) Put emphasis on client needs and requirements. For the successful implementation of KM projects in sales organisations, or those that work closely with clients such as Debus IT, the needs and requirements of the clients must form the backdrop against which all other activities are performed. By paying close attention to the clients, high business valueadding KM solutions can be deployed. (4) Integrate KM within normal business processes. In order to fully exploit the potential of KM environments, knowledge asset management has to be integrated into standard business processes, like in the case of MDA. At any point in a process, knowledge workers should be able to learn from the experiences of their colleagues, as well as to provide their own experiences as lessons from which others can learn. Finally, the four cases verified the importance of finding the right balance between IT solutions for capturing explicit codified knowledge and leaving enough room to allow for direct personal knowledge exchanges. Note 1. The Know-Net knowledge management solution was the result of the Know-Net research project (Know-Net consortium, 1999), that ended in 2000. Since then, the Know-Net solution has become a commercial product and service, offered world-wide by Knowledge Associates, a network of KM specialists, and Planet Ernst & Young. During the Lever project (Lever consortium, 2001), Know-Net provided the main platform for enabling KM in four software development companies. References Applehans, W., Globe, A. and Laugero, G. (1999), Managing Knowledge. A Practical Web-based Approach, Addison-Wesley Information Technology Series, Addison-Wesley, Reading MA. Barney, J. (1991), “Firm resources and sustained competitive advantage”, Journal of Management, No. 17, pp. 99-120. Belohlav, J. (1996), “The evolving competitive paradigm”, Business Horizons, March-April, pp. 11-19. Cap Gemini (1999) “Managing your knowledge for sustainable competitive advantage”, White Paper.

BPMJ 9,3

378

Davenport, T. and Prusak, L. (1998), Working Knowledge: How Organisations Manage What They Know, Harvard Business School Press, Boston, MA. Day, J. and Wendler, J. (1998), “The new economics of organisation”, The Mckinsey Quarterly, Vol. 1, p. 19. den Hartigh, E. and Langerak, F. (2001), “Managing increasing returns”, European Management Journal, Vol. 19 No. 4, pp. 370-8. Drucker, P. (1997), “The future that has already happened”, Harvard Business Review, September-October. Glazer, R. (1991), “Marketing in an information-intensive environment: strategic implications of knowledge as an asset”, Journal of Marketing, Vol. 55, pp. 1-19. Grant, R. (1991), “The resource-based theory of competitive advantage: implications for strategy formulation”, California Management Review, Vol. 33 No. 3, pp. 114-35. Hansen, M.T., Nohria, N. and Tierney, T. (1999), “What’s your strategy for managing knowledge?”, Harvard Business Review, March-April, pp. 107-16. Know-Net Consortium (1999), The Know-Net WWW site, available at: www.know-net.org KPMG (1999), “The knowledge journey”, White Paper. Ku¨hn, O. and Abecker, A. (1997), “Corporate memories for knowledge management in industrial practice: prospects and challenges”, Journal of Universal Computer Science, Vol. 3 No. 8, pp. 929-54. Lever Consortium (2001), The Lever WWW site, available at: www.kmlever.com Mentzas, G.N., Apostolou, D., Young, R. and Abecker, A. (2000), “Knowledge networking: a holistic approach, method and tool for leveraging corporate knowledge”, Journal of Knowledge Management, Vol. 5 No. 1, pp. 94-106. Newbern, D. and Dansereau, D.F. (1995), “Knowledge maps for knowledge management”, in Wiig, K.M. (Ed.), Knowledge Management Methods, Schema Press, Arlington, TX. Ovum (1999), “Knowledge management: building the collaborative enterprise”, market report. Porter, M. (1980), Competitive Strategy: Techniques for Analyzing Industries and Competition, The Free Press, New York, NY. Schreiber, G. (1999), Knowledge Engineering and Management: The CommonKADS Methodology, MIT Press, Cambridge, MA. Spek, R. and Spijkevert, A. (1997), Knowledge Management, Dealing Intelligently with Knowledge, The Knowledge Management Network, CIBIT, Utrecht. Spender, J. (1996), “Making knowledge the basis of a dynamic theory of the firm”, Strategic Management Journal, Vol. 17, pp. 45-62. Staab, S., Studer, R., Schnurr, H.-P. and Sure, Y. (2001), “Knowledge processes and ontologies”, IEEE Intelligent Systems, January/February. Teece, D., Pisano, G. and Shuen, A. (1997), “Dynamic capabilities and strategic management”, Strategic Management Journal, Vol. 18, pp. 509-33. Wernerfelt, B. (1984), “A resource-based view of the firm”, Strategic Management Journal, Vol. 14, pp. 4-12. Wiig, K.M. (1995), Knowledge Management Methods, Schema Press, Arlington, TX.

Appendix. Overview of KM methodologies (Table AI) The KM method of Ernst and Young The KM practice of Ernst and Young (pre-merger) provided a range of knowledge services and solutions that focus exclusively on strategy, process and change management. It does not cover system integration or development services – which are usually outsourced to another part of the organisation. Ernst and Young (E&Y) advocated a “pilot-first” approach that consists of the three delivery phases: architect, integrate, operate; see also Ovum (1999). The “architect” phase aligns the KM strategy and architecture with the organisation’s business objectives. To a large extent, the services in this phase are aimed towards providing a “blue-print” for piloting and implementing KM solutions. It reflects making choices about where, when and how each type of available solution must be applied in order to realize the intended benefit. The “integrate” phase involves piloting specific knowledge-based solutions, such as community enablement, which focuses on the creation of communities of interest and practice; content management, which focuses on managing content generated externally (typically, this solution addresses the need to match content with business needs for research, analysis and business intelligence capabilities); decision making, which addresses the needs of selected high-level decision-makers and is achieved through a combination of content management and the development of explicit decision models, decision workflow processes and data warehouse integration; and KnowledgeWeb, which provides a user-centric knowledge brokering system. Finally, the “operate” phase involves wide-scale deployment of the KM system throughout the enterprise, and the development of additional pilot solutions outlined in the integrate phase. The KM method of Cap Gemini The management consulting company Cap Gemini (pre-merger) offered KM services based on the applied knowledge management framework (Cap Gemini, 1999). This framework distinguishes between phases of operation and focused streams of activity. The phases of operation provide the logical structure on which the phases of the KM programme may be built. These can be effectively divided into two dominant phases of activity. The initial or “scoping” phase, which includes business vision and business readiness phases. The second or “delivery” phase includes the iterative phases of solution design, solution development and solution deployment plus operate and evaluate. In “scoping”, the business vision phase helps to identify the scale of the opportunity for a client and create a vision for a knowledge-enabled organisation. The business readiness phase helps to establish the ability of an organisation to deliver the vision, and determines the effective ambition level for delivery phases to implement elements of the vision. In solution design, the structure and design of the solution is finalised, building upon the design outlines in the scoping phases. Solution development creates the new functional capabilities laid out in the design phase. Key issues in solution deployment are to ensure that the dependent functional streams deploy their solutions in a logical sequence to ensure that the programme maintains its desired impact. In solutions operation and evaluation, the impact of the delivery is measured against the forecast benefits; any insights or new knowledge captured during the programme are evaluated and measured. The focused activity streams help to define the activities of the KM programme. Each stream exists in every phase of the programme and addresses the key functional elements. In addition, streams convey the means to overcome issues created when dealing with a business programme on this scale. The KM method of KPMG KPMG focuses on optimising the seven key knowledge processes within an organization – creation, application, exploitation, sharing, encapsulation, sourcing and learning (see KPMG, 1999). Its services cover awareness raising, strategy, systems integration and development, business process engineering and change management. KPMG regards KM as an ongoing process, consisting of a number of integrated projects, phased over time, rather than a single

Knowledge management implementations 379

BPMJ 9,3

380

discrete project. KPMG favours a programme-management approach that involves the three principal service components. During strategic planning the company utilizes a mixture of standard, in-house and specialist techniques to: . confirm strategic objectives and the vision for KM using its KM framework tool; . analyse and benchmark the current status of the organisation’s KM infrastructure, using knowledge audit and other analysis techniques; . identify problems and opportunities, and agree on a measurement system for evaluating the effectiveness of, and the business benefits associated with, KM; and . plan a series of quick win projects and longer-term initiatives based on a clear business case. KPMG’s implementation activities cover IT implementation and change management. KPMG does not develop its own software products for KM. Rather, its KM solutions are built using tools from Microsoft, with whom the company has a formal alliance, and from several specialist KM tool vendors. The KM method of IBM Global Services The relation of Global Services to the rest of IBM and its position as a relative newcomer in the management consultancy business gives it a different perspective to its main competitors. It has no single prescriptive method for KM projects, but it uses several techniques and metrics to help an organization understand and develop its knowledge assets and intellectual capital (Ovum, 1999). A key concept promoted by IBM is the knowledge map. The principle behind knowledge maps is that a company needs a high-level view of its existing knowledge before it can decide on a programme of business improvement. The overall approach is based on identifying an organisation’s key tacit and explicit knowledge assets and its current approaches for managing knowledge processes. These approaches are mapped to types of knowledge in order to identify knowledge “gaps”. KM solutions, processes and tools are then clearly prioritised with reference to each approach and type. The result is an overall knowledge performance study that serves as the basis for system design. The KM method of J. D. Edwards Applehans et al. (1999) developed a KM method that is based on the practical experience gained from the development of Web-based KM systems for J.D.Edwards. Their approach focuses on the design and development of “content centres” as the cornerstone of the knowledge architecture. The method supports the translation of content centres into a networked organisation, including navigation strategies and other issues surrounding the deployment of people, content, and technology. Concepts and tools used in this approach include the audit, the content portfolio and the knowledge architecture. The purpose of the audit is to help break KM down into manageable projects without losing sight of the “big picture”. It supports the identification of the success factors of the organisation and the relevant business processes; the important points in these business processes were actions or decisions are taken, the people that act or use content and the content itself. The content portfolio represents the specific pieces and types of content that the organisation must effectively package and deliver to people, who can act on them as knowledge. This can include documents, competitive intelligence, product specifications, case studies, etc. The knowledge architecture represents the organisation’s formal recognition that it has important experience and expertise that it must preserve and use to its advantage. The knowledge architecture identifies the scope of the investment that will be made in managing knowledge in terms of people, technology and content.

These concepts and tools are utilized in order to design an infrastructure that organises around knowledge by combining content and people. Two additional concepts that are key to this approach are the knowledge storyboarding and the knowledge networks. The purpose of the knowledge storyboarding is to identify the relationships among people, processes and knowledge. It consists of four steps: specification of a business process and related steps within the process; identification of information leveraging points within the business process; identification of organisational roles and people that use information in each step; and identification of detailed content used in each step. An essential objective of this tool is to identify the information needs of users within specific business processes. This is accomplished with user-profiling techniques.

Knowledge management implementations 381

Other KM methods and techniques There exist a large number of techniques for auditing, surveying, eliciting, analysing and modelling knowledge. Although these techniques are not stand-alone or complete methodologies, they are used during KM implementations mainly for field knowledge analysis and modelling purposes. Wiig (1995) provides an extensive overview of such methods. Table AI presents a summary of the most characteristic knowledge analysis and modelling techniques. Method

Key characteristics

Task environment analysis TEAs consist of in-depth investigations of how knowledge (TEA); see Wiig (1995) workers perform business tasks and the conditions under which they work. The focus is on knowledge, its presence and use of knowledge, how the task is performed at present, what its inputs are, what its deliverables are. TEA provides the added perspective of knowledge flows and uses Basic knowledge analysis Basic knowledge analysis refers to an analysis and (BKA); see Wiig (1995) characterization of the knowledge in the task environment. It focuses on how knowledge is held, used in decisions and other knowledge-intensive tasks Knowledge mapping is used to develop concepts maps as Knowledge mapping hierarchies or networks. Knowledge mapping systems are used for (Kmap); see Newbern and identifying relevant information from workers, displaying this Dansereau (1995) information and presenting it for training, communicating, planning, problem-solving or decision-making purposes Knowledge use and KURA is performed to explicate knowledge use and proficiency requirements analysis requirements. The focus is on the use of knowledge in problem (KURA); see Wiig (1995) solving, decision making and other knowledge-intensive processes within the target business area Knowledge scripting and Knowledge scripting and profiling is used for the detailed profiling (KS&P); see Wiig description of knowledge-intensive processes, tasks and scripts (1995) Knowledge flow analysis Knowledge flow analysis is used to gain overview of knowledge (KFA); see Wiig (1995) exchanges, losses, or inputs to the business process or the whole enterprise. It also determines characteristics, strengths and weaknesses of existing and potential knowledge exchanges These methods focus on the application-oriented development of CommonKADS; see ontologies and support all phases from the early stages of setting Schreiber et al. (1999); up a KM project to the final roll-out and maintenance of the Knowledge metaprocess; ontology-based KM application see Staab et al. (2001)

Table AI. Knowledge analysis and modelling techniques

The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister

BPMJ 9,3

382

The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/1463-7154.htm

The potential use of knowledge management for training A review and directions for future research Joseph R. Muscatello Infiniti Systems Group, Wickliffe, Ohio, USA Keywords Knowledge management, Training, Information exchange Abstract The powerful competitive traits of a learning organization are well documented. Companies in the future will have to provide new and innovative techniques if they hope to survive and prosper. These techniques will revolve around knowledge management. However, research on using knowledge management as a tool to increase learning is minimal. Knowledge management systems provide a new vehicle to effectively and efficiently disseminate knowledge, skill and expertise throughout an organization. Knowledge management is an outgrowth of the “information society”. Currently, research and implementation of knowledge has centered on the outside customer service and outside customer relationships. Little has been written about the use of knowledge to serve the training needs of the firm. This study combines the powerful competitive traits of a learning organization with knowledge management traits to potentially increase the positive effects of both.

Business Process Management Journal Vol. 9 No. 3, 2003 pp. 382-394 q MCB UP Limited 1463-7154 DOI 10.1108/14637150310477948

Introduction Organizations are constantly looking for new, innovative ways to increase their competitive advantage. To find these areas of advantage we need to examine what successful organizations are doing in today’s world. These organizations are the high performers. High performance organizations have several key elements that enable them to grow. Bullinger (1999) listed them as: . Core competence. High competitiveness, achieved through a process which concentrates itself on the actual competence and precedes the process of growth. . Networks and cooperations. Demands on the company are becoming more and more complex. In order to be able to concentrate on core competence, the companies have to look for partners and cooperate with them industry wide. . Process orientation. Decentralized process and product-oriented organizational structures are generally more suitable for growth than central, functional-oriented ones; it is easier to connect and disconnect profit centers and independent business areas in the form of a company organization, than to inflate and reduce functional departments. . Free margins. The creation of growth cells in the company, e.g. with creative teams, corresponding structural freedom and a variable schedule, so that every employee can think about processes, products, structures, problem fields and improvement potentials in general. In this case, it is required to reduce formalisms in workflow organization and leadership.

.

.

Learning organizational structures. The capacity of learning is becoming Knowledge a core capacity in innovative, growth-oriented companies. Learning is management for necessary in order to improve oneself, to grow more productive and to training gain the ability to adapt oneself to changes – and this faster than the competition. Knowledge management and information technology. Knowledge and 383 information are a basis for creativity and the capacity to learn. Management of knowledge does not only take place in the company, but is also accomplished in a comprehensive way. To achieve this, a suitable information technology has to be created. Particularly, Internet and intranet technologies would provide a good solution.

The purpose of this study is to combine the powerful competitive traits of a learning organization with knowledge management traits to determine the positive impact the combined traits may have on an organization. The first five competitive elements have been known and debated for years and their overall usefulness as management tools has been established. Knowledge management and information technology tools have not had the same scrutiny. However, the potential positive impact of knowledge management on the performance of companies is becoming clearer every day. The Internet and intranet have made an enormous impact on our lives. As the lessons learned from Internet services are applied to different business settings, the performance advantage will become very clear. Companies in the future will have to provide new and innovative techniques if they hope to survive and prosper. These techniques will revolve around knowledge management. Knowledge management is an outgrowth of the “information society”. Currently, research and implementation of knowledge has centered on the outside customer service and outside customer relationships. Little has been written about the use of knowledge to serve the training needs of the firm. This leads to the primary objectives of this paper. (1) Review the current relevant literature on knowledge management and training. (2) Present the technical issues associated with database development and knowledge engines. (3) Combine knowledge management and training traits and develop theories that may positively impact on an organization. To meet our objective of discussing knowledge management, we need to discuss what knowledge is since it is highly nebulous at this time. Menon and Varadarajan (1992) realized this need in their paper on “A model of marketing knowledge use within firms” and wrote: The need to discuss the problems central to conceptualization and measurement of knowledge utilization before expounding any model of the organization and informational

BPMJ 9,3

384

factors affecting knowledge utilization is predicated on the logic that, before we can even begin to discuss how to increase knowledge use, we need to address what knowledge use is.

Research and review of knowledge It is well documented that knowledge and the innovative use of knowledge are strengths needed for companies to remain competitive (Sveiby, 1997, European Commission, 1995; Nonaka, 1994; Nonaka, 1991). Companies are realizing that the basis for their competitive advantage is their knowledge base (Sveiby, 1997) and this knowledge should be widely distributed throughout the organization (Candlin and Wright, 1992). Knowledge management should be concerned with the following tasks (Candlin and Wright, 1992): (1) Analyzing the role of knowledge, skills and expertise in the organization. (2) Understanding how they support the organization in meeting its business objectives. (3) Identifying knowledge sources, knowledge flows and knowledge bottlenecks and where the organization is vulnerable to knowledge attrition through loss of staff. (4) Determining the scope for improving performance through the better acquisition, utilization and communication of knowledge. (5) Implementing the systems and organization necessary to achieve the improved performance and business effectiveness. Candlin and Wright believe that analysis and thought must be precursor to building a knowledge-based company. A company needs to understand how to use information before it can be of benefit. However, “the consensus seems to be that our current understanding of how such information is used within organizations, and of the major detriments of and impediments to knowledge utilization within firms, is less than adequate” (Menon and Varadarajan, 1992). “These developments are indicative of the growing recognition within both the academic and the business communities of the importance of the study of the knowledge utilization process in marketing and other administrative disciplines” (Menon and Varadarajan, 1992). Those other administrative disciplines that Menon and Varadarajan are referring to certainly include the use of knowledge as a training and skill-building tool. The uses of knowledge management in specific business settings, such as training, have not been fully developed. In this paper we try to understand the knowledge needs of employees, students and others that might benefit from its use as a training device. Knowledge management policies and practices It is not the intention of this article to discuss the most appropriate definition of knowledge management, since no clear consensus has yet emerged (Nonaka,

1991). Nor do we seek to explain the “how to” gathering and dissemination of Knowledge data in great detail. Rather, the article briefly discusses common knowledge management for management definitions. training Knowledge, learning and cognition are classical terms that have been rediscovered in the context of the information technology and knowledge management revolutions. Beckman (1997) compiled a number of useful and 385 relevant definitions of knowledge and organizational knowledge (Cayayannis, 1999): . Knowledge is organized information applicable to problem solving (Woolf, 1990). . Knowledge is information that has been organized and analyzed to make it understandable and applicable to problem solving or decision making (Turban, 1992). . Knowledge encompasses the implicit and explicit restrictions placed on objects (entities), operations and relationships, along with general and specific heuristics and inference procedures involved in the situation being modeled (Sowa, 1999). . Knowledge consists of truths and beliefs, perspectives and concepts, judgements and expectations, methodologies and know-how (Wiig, 1993). . Knowledge is the whole set of insights, experiences, and procedures which are considered correct and true, and which therefore guide the thoughts, behaviors, and communication of people (van der Spek and Spijkervet, 1997). . Knowledge is reasoning about information to actively guide task execution, problem-solving and decision-making in order to perform, learn and teach (Beckman, 1997). . Organizational knowledge is the collective sum of human-centered assets, intellectual property assets, infrastructure assets and market assets (Brooking, 1996). . Organizational knowledge is processed information embedded in routines and processes, which enable action. It is also knowledge captured by the organization’s systems, processes, products, rules and cultures (Myers, 1996). Knowledge management is viewed as a sociotechnical system of tacit and explicit business policies and practices (Carayannis, 1999). It can also be expanded to include systemic knowledge, hidden knowledge and relationship knowledge (Johannessen et al., 1999). What knowledge management is not, is information. Many people make the mistake of assuming that information is the same as knowledge, it is not. Rather, it is the effective use of information that creates knowledge management. Regardless of the definition, knowledge management is useful when the company gains a critical advantage that its competitors do not have. Companies that will reap the highest benefits are the ones that realize

BPMJ 9,3

386

knowledge is a valuable resource of competitive advantage (Nonaka, 1991) and the dissemination of data, in a usable fashion, to all levels of the organization, is necessary to gain that advantage (Nonaka and Takeuchi, 1995). The ability to use knowledge as an aid to make decisions allows a user to increase their personnel knowledge base. Therefore, the knowledge server without the use of a human expert trains them. The knowledge servers may replace the need for human interaction, thus allowing the user to train and educate on their own. Since the above definitions all describe knowledge as the effective use of knowledge to resolve an organization’s issues and make them more effective, it stands to reason that the effective use of knowledge improves an organization’s ability to train and educate. Characteristics of training Nearly all employees receive some form of training during their careers. Indeed, individuals rely on training to improve their current skills and to learn new skills (Mathieu et al., 1992). A lot of research has called for the examination into why training works. With companies spending as much as $200 billion annually on workforce training (McKenna, 1990) there is little wonder why. Given the significance of formal training programs for organizational effectiveness, it is imperative that organizations design and implement training programs in the most effective manner and they understand the factors that contribute to training effectiveness (Facteau et al., 1995). Many studies have found that pretraining motivation may influence important training outcomes (Baldwin et al., 1991) such as performance appraisals, and another study found that pretraining was related to classroom performance (Baldwin and Karl, 1987). These studies indicate that pretraining motivation has an important influence on the extent to which trainees actually learn the material presented to them during a training program (Facteau et al., 1995). Learning organizations know that much valuable learning takes place outside of formal training and education programs. For instance, proponents of “organizational learning” (e.g. Argyris and Schon, 1978) and “learning organizations” (e.g. Bullinger, 1999) have emphasized the need to analyze learning activities of all kinds, to set up many different structures, and to create a culture of continuous personal development (Birdi et al., 1997). This type of organization empowers its employees to learn and grow. It allows employees to manage their careers to some extent. The employee or student should also have an interest and responsibility to keep their education current. Rosow and Zager (1988) have explored the suggestion that individuals should accept greater responsibility for their own development. Practical innovations in this area include employee development programs, in which organizations provide financial and other resources for learning that is not only voluntary, but also take place outside working hours (Birdi et al., 1997). This voluntary training taking place outside of working

hours may protect the employee and the organization against skill atrophy Knowledge (Muscatello, 1999). Many people at work have little exposure to training and management for there is growing recognition of the need to rectify that imbalance. training Mentoring relationships have been shown to positively affect the careers of employees. Mentors are individuals with advanced experience and knowledge, who are committed to providing upward support and mobility to their 387 prote´ge´s’ careers (Hunt and Michael, 1983; Kram, 1985). Mentoring can be informal or formal, and personality, gender and other traits can have an effect on how much or how little mentoring you receive. The use of a knowledge management system may produce some interesting results when combined with training. Some or all of the issues mentioned may be resolved. Developing the knowledge-based vision for training In the knowledge-based company, the value added for the customers (people using the knowledge system would be internal customers) comes from the communication among the members of the organization (Howells, 1996). The knowledge-based training vision should be explicitly understood by all employees and driven by a firm to be a core competency. This requires a company to have a knowledge-based vision that ensures the knowledge server is kept up to date with the latest available information. The latest available information should be the knowledge that has been applied at some other area of the organization, that will educate or enhance the user’s ability to make business decisions that will positively affect the outcome of a firm. The vision must also include a clear understanding of why employees must feed the knowledge management system. One of the most important roles managers will have when they introduce knowledge management systems is motivating their employees to keep the system current. Without implementing a vision, employees may see the knowledge server as a fad or an infringement on their turf. Vision leads to knowledge creation, integration and application, which supports organizational innovation (Johannessen et al., 1999). The knowledge database will be built with a lot of tacit knowledge that is based on experience and trial and error. This knowledge will allow an organization to sub-optimize various tasks, which usually leads to an increased optimization of a whole process. This type of knowledge transfer follows a master-apprentice approach. With knowledge management, the master need not be available for consultation after the knowledge is entered into the system, provided that the apprentice can access the information in a useable format and successfully complete the task. Problems with organizing a knowledge management database The development of a knowledge management system is not simple. Aside from the fact that technical hardware problems can and do arise, even the best

BPMJ 9,3

388

human experts can have “off” days or become tired or irritable. They can make mistakes, forget important points and apply their knowledge in a selective manner, depending on their own perceptions and bias (Candlin and Wright, 1992). This requires a strong person to accept the role of knowledge manager and facilitate the experts, at all levels of the organization, to reach a consensus. Then the knowledge manager must compile, build and incorporate the opinion into the knowledge server. Without quality being built into the knowledge management system, the use of tacit knowledge as a training aid will fail. When building the knowledge management server a firm must focus on the value of the information. For knowledge management to become a competitive advantage, its value must be associated with durability (the rate at which it becomes obsolete), transparency (the speed with which other firms can develop the same knowledge), transferability (how easily firms can transfer and share it) and replicability (how easily firms can reproduce and use it) (Grant, 1991). Firms wishing to develop strong knowledge management systems need to take a strong interest and stand on the use of the information. “It is risky to assume that knowledge is ‘naturally’ at the right place, let alone that it is the ‘right’ knowledge. Specific efforts are needed to ensure its presence and to guide its recreation. Thus it is necessary to ensure that the right knowledge gets to the right place and so increases the innovation power of the organization and its knowledge workers. Both require knowledge management” (Hendriks and Vriens, 1999). Poor constructs of a knowledge database could provide poor training information, resulting in the need to retrain an employee or student. Also, poor knowledge could render the training ineffective and the organization would suffer a loss of money, as well as a training instrument. Technology and knowledge management Information technology is making it easier to input, store and retrieve specific information. Database management systems (DBMS) and knowledge base systems (KBS) combine a database with domain specific knowledge in the form of “if-then” rules or other constructs such as frames, objects and semantic nets, to create a knowledge base (Basu, 1988). These technologies can give a great contribution to the firm’s ability to transfer and share valuable knowledge, without geographical limits (Bolisani and Scarso, 1999). The systems are designed to make more “intelligent” decisions based on rules and procedures stored in the knowledge base (Basu, 1988). The company may have the expert ability on staff, however, even if a human expert has complete knowledge of a domain, the expertise may not be available at the time or place of the demand (Candlin and Wright, 1992) for a solution. Furthermore, the current staff, confronted by the problem, may not know the proper terminology in order to input the correct question into the knowledge server. What is needed is instant expertise and logic. Given that these problems can be nebulous and hard to put into words, expert systems, using an inference

engine, have been developed to reduce the burden on the user. An inference Knowledge engine is a computer program that determines which of the rules within the management for knowledge base should be selected for evaluation, interprets the rules in the training light of the data supplied and provides conclusions (Basu, 1998). The use of inference engines can enable instantaneous learning to occur. This can greatly reduce the learning curve for people and help facilitate continuous training 389 throughout the organization. Financial returns of knowledge management The investment in knowledge management is subjected to the same financial return calculations as any other advanced IT system. However, it is highly likely that the investment in a knowledge management server will meet or exceed the needed financial returns if implemented correctly. In advanced information economies where information and knowledge work account for 60 percent of GDP, it is only logical to conclude that revolutionary improvements in IT, coupled with exponential increases in IT investments within factories and offices, will most certainly lead to significant and widespread gains in organizational productivity (Laudon and Marr, 1994). Knowledge management and training: potential positive impacts on an organization and theories for future research Since the use of knowledge management techniques is relatively new for training applications, this paper relies on the combined work of knowledge management and training researchers to derive its potential. Knowledge management can be used as a short-term information source or a long-term knowledge and skill builder. Either way the knowledge server provides a new and innovative way to access information for training. Information should be available in the knowledge server to be accessed by all members of an organization who can benefit from it. An expert system can present several advantages as defined by Candlin and Wright (1992). These advantages confirm the use of a knowledge server as a short-term or long-term training device: . Instant expert advice across the organization leads to quicker system response time. . Instant expert advice encourages more requests for advice and increases the amount of expertise used within the organization. . Expertise may be employed by lower paid staff. . The task of knowledge retrieval can be removed from the expert, allowing the expert more time for the development of knowledge. The use of a knowledge management system may decrease the learning curve owing to instantaneous expert information, the ability to employ expertise at all levels and increase expert development efforts.

BPMJ 9,3

390

Pretraining motivation has been found to influence training effectiveness. Social support variables and attitudinal variables of intrinsic incentives, training reputation, organizational commitment, and compliance are the most highly related factors to pretraining motivation. Managers who perceived more intrinsic reasons to attend training, who were less likely to attend training because it was required, reported higher levels of motivation to attend and learn from the agency’s supervisory and managerial training programs. (Facteau et al., 1995). The use of a knowledge system in training could provide more intrinsic reasons to attend training, show a commitment to training, build a quality training program and create a less threatening environment. This may positively effect the pretraining motivation of employees and help support the learning organization. Organizations know that a learning structure has to be innovative. Developing a knowledgeable staff is more than formal training and education. It requires the transfer of information of all kinds. Details that only experience can provide need to be incorporated into the training environment. A knowledge system could support a learning organization by reducing the amount of time required to transfer knowledge from a master to an apprentice. Employees need to take advantage of training and education to maintain or increase their skill set. The motivation for increasing their skills can be lacking because they do not believe the company values education and training. This may be because the company has not provided an incentive or a vehicle that meets their particular needs. Training in non-working hours can be difficult if the employee has other demands on their time. A knowledge management training system can allow an employee to increase their training and education by working at their own pace in off-hours. Mentoring is a powerful force when it comes to promotability and financial increases. Mentors prepare prote´ge´s for advancement by nominating them for desirable projects, lateral moves, and promotions (sponsorship); providing the prote´ge´ with assignments that increase visibility to organizational decision makers and exposure to future opportunities (exposure and visibility); sharing ideas, providing feedback, and suggesting strategies for accomplishing work objectives (coaching); reducing unnecessary risks that might threaten a prote´ge´’s reputation (protection); and providing challenging work assignments (Noe, 1988). Clearly, the role of the mentor can be extremely strong in advancing a career. Conversely, not having access to a mentor can be detrimental to a career, regardless of ability, education and work ethic. The use of a knowledge system could change the mentoring role in many ways. First, the master-apprentice relationship could be incorporated into a training database to ensure knowledge is available to everyone. Case studies and decision support tools could be used to provide “real world” examples. This eliminates some of the “favoritism” of mentoring. Second, feedback can be

provided from multiple sources that have entered information into the Knowledge database. This can help form a stronger conclusion. Third, coaching and management for protection could become more “informal” if the knowledge server provided training strategic and managerial feedback on business issues. Conclusion Knowledge management has now established itself as a key competitive strategy that can give multiple advantages to a company. One of those advantages is training. Training in both the short-term and long-term can have positive effects on the performance of a company. A knowledge server allows access to information at all levels of an organization and provides a vehicle for people to improve themselves in a formal and informal way. It can reduce the amount of structure required, eliminate excuses and de-politicize the organization, while empowering people to learn on their own. Much research needs to be done in this area to prove or disprove the benefits of using knowledge management as a training aid. This study points out the potential advantages of merging the key competitive traits of knowledge management and training and offers theories for future research. References Argyris, C. and Schon, D.A. (1978), Organizational Learning: A Theory of Action Perspective, Addison-Wesley, Reading, MA. Baldwin, T.T. and Karl, K.A. (1987), “Transfer of training: a review and directions for future research”, Personnel psychology, Vol. 41, pp. 63-105. Baldwin, T.T., Magjuka, R.J. and Loher, B.T. (1991), “The perils of participation: effects of choice of training on trainee motivation and learning”, Personnel Psychology, Vol. 44, pp. 51-66. Basu, A. (1998), “Perspectives on operations research in data and knowledge management”, European Journal of International Research, Vol. 111, pp. 1-14. Beckman, T. (1997), “A methodology for knowledge management”, International Association of Science and Technology for Development (IASTED) AI and Soft Computing Conference, Banff, Canada. Birdi, K., Allen, C. and Warr, P. (1997), “Correlates and perceived outcomes of four types of employee development activity”, Journal of Applied Psychology, Vol. 82 No. 6, pp. 845-57. Bolisani, E. and Scarso, E. (1999), “Information technology management: a knowledge-based perspective”, Technovation, Vol. 19, pp. 209-17. Brooking, A. (1996), Introduction to Intellectual Captal, The Knowledge Broker Ltd, Cambridge. Bullinger, H.J. (1999), “Turbulent times require creative thinking: new European concepts in production management”, International Journal of Production Economics, pp. 9-27. Candlin, D.B. and Wright, S. (1992), “Managing the introduction of expert systems”, International Journal of Production Management, Vol. 12 No. 1, pp. 46-59. Carayannis, E.G. (1999), “Fostering synergies between information technology and managerial and organizational cognition: the role of knowledge management”, Technovation, Vol. 19, pp. 219-31. European Commission (1995), Green Paper on Innovation, Brussels.

391

BPMJ 9,3

392

Facteau, J.D., Dobbins, G.H., Russell, J.E., Ladd, R.T. and Kudish, J.D. (1995), “The influence of general perceptions of the training environment on pretraining motivation and perceived training transfer”, Journal of Management, Vol. 21 No. 1, pp. 1-25. Grant, R.M. (1991), “The resource based theory of competitive advantage: implications for strategy formulation”, California Management Review, Spring, pp. 114-35. Hendriks, P.H.J. and Vriens, D.J. (1999), “Knowledge-based systems and knowledge management: friends or foes?”, Information & Management, Vol. 35, pp. 113-25. Howells, J. (1996), “Tacit knowledge, innovation and technology transfer”, Technology Analysis & Strategic Management, Vol. 8 No. 2, pp. 91-105. Hunt, D.M. and Michael, C. (1983), “Mentorship: a career training and development tool”, Academy of Management Review, Vol. 8, pp. 475-85. Johannessen, J.A., Olsen, B. and Olaisen, J. (1999), “Aspects of innovation theory based on knowledge management”, International Journal of Information Management, Vol. 19, pp. 121-39. Kram, K.E. (1985), Mentoring at Work: Developmental Relationships in Organizational Life, Scott Foresman, Glenview, IL. Laudon, K.C. and Marr, K. (1994), “Productivity and the enactment of a macro culture”, Proceedings of the Fifteenth International Conference on Information Systems, Vancouver, Canada, pp. 243-61. McKenna, J.F. (1990), “Take the ‘A’ training”, Industry week, Vol. 239, pp. 22-9. Mathieu, J.E., Tannenbaum, S.I. and Salas, E. (1992), “Influences of individual and situational characteristics on measures of training effectiveness”, Academy of Management Journal, Vol. 35 No. 4, pp. 828-47. Menon, A. and Varadarajan, P.R. (1992), “A model of marketing knowledge use within firms”, Journal of Marketing, Vol. 56, pp. 53-71. Muscatello, J. R. (1999), “ERP training and education: do your employees have the skills?” Presentation at the Cleveland Advanced Manufacturing Society User Group. Myers, P. (Ed.) (1996), Knowledge Management and Organizational Design, ButterworthHeinemann, Oxford. Noe, R.A. (1988), “An investigation of the determinants of successful assigned mentoring relationships”, Personnel Psychology, Vol. 41, pp. 457-79. Nonaka, I. (1991), “The knowledge-creating company”, Harvard Business Review. Nonaka, I. (1994), “A dynamic theory of organizational knowledge creation”, Organizational Science, Vol. 5 No. 1, pp. 14-37. Nonaka, I. and Takeuchi, H. (1995), The Knowledge-Creating Company, Oxford University Press. Rosow, J.M. and Zager, R. (1988), Training: The Competitive Edge, Jossey-Bass, San Francisco, CA. Sowa, J.F. (1999), Knowledge Representation, Brooks Cole Publishing Co., Monterey, CA. Sveiby, K.E. (1997), The New Organizational Wealth: Managing and Measuring Knowledge Based Assets, Berrett-Koehler, San Francisco, CA. Turban, E. (1992), Expert Systems and Applied Artificial Intelligence, Macmillan, Basingstoke. Van der Spek, R. and Spijkervet, A. (1997), “Knowledge management: dealing intelligently with knowledge”, in Liebowitz, W. (Ed.), Knowledge Management and Its Integrative Elements, CRC Press, Boca Raton, FL. Wiig, K. (1993), Knowledge Management Foundation, Schema Press, Arlington, TX. Woolf, H. (Ed.) (1990), Webster’s New World Dictionary of the American Language.

Further reading Albino, V., Garavelli, A.C. and Schiuma, G. (1999), “Knowledge transfer and inter-firm relationships in industrial districts: the role of the leader firm”, Technovation, Vol. 19, pp. 53-63. Barney, J.B. (1991), “Firm resources and sustained competitive advantage”, Journal of Management, Vol. 17 No. 1, pp. 99-120. Beer, M., Eisenstat, A. and Spector, B. (1990), The Critical Path to Corporate Renewal, Harvard Business School Press, Boston, MA. Black, J.A. and Boal, K.B. (1994), “Strategic resources: traits, configurations and paths to sustainable competitive advantage”, Strategic Management Journal, Vol. 15 (special issue), Summer, pp. 131-48. Brynjolfsson, E. (1994), “Technology’s true payoff”, Informationweek, Vol. 10, pp. 34-6. Brynjolfsson, E. and Hitt, L. (1996), “Paradox lost? Firmlevel evidence on the returns to information system spending”, Management Science, Vol. 42 No. 4, pp. 541-58. Collis, D.C. and Montgomery, C.A. (1995), “Competing on resources: strategy in the 1990s”, Harvard Business Review, Vol. 73, pp. 118-28. Evans, P.B. and Wurster, T.S. (1997), “Strategy and the new economics of information”, Harvard Business Review, pp. 71-84. Grant, R.M. (1996), “Toward a knowledge based theory of the firm”, Strategic Management Journal, Vol. 17, Winter (special issue), pp. 109-22. Grover, V., Teng, J., Segars, A. and Fiedler, K. (1998), “The influence of information technology diffusion and business process change on perceived productivity: the IS executives perspective”, Information and Management, Vol. 34, pp. 141-59. Hart, S.L. (1995), “A natural resource-based view of firm”, Academy of Management Review, Vol. 20 No. 4, pp. 986-1014. Johannessen, J.A. (1997), “Aspects of casual processes”, Kybernetes, Vol. 26 No. 1, pp. 30-52. Kalseth, K. (1990), “Strategic uses of information: challenges for the information services department, American Society for Information Science”, Bulletin of the American Society of Information Science, Vol. 17 No. 1, pp. 22-3. Kanter, R.M. (1986), “Creating the creative environment”, Management Review, Vol. 75, pp. 11-12. Kilbane, D. (1999), Knowledge Management: What You Don’t Know Can Help You, Automatic I.D. News, Cleveland, OH. Lewis, J.D. (1995), The Connected Corporation, The Free Press, New York, NY. Lines, A.H. (1981), “The microcomputer in management”, The Journal of the Operations Research Society, Vol. 32, pp. 297-301. Lundvall, B.A. and Johnson, B. (1994), “The learning economy”, Journal of Industry Studies, Vol. 1 No. 2, pp. 23-42. Malone, T.W., Crowston, K., Lee, J. and Pentland, B. (1999), “Tools for inventing organizations: toward a handbook of organizational processes”, Management Science, Vol. 45 No. 3, pp. 425-43. Mintzberg, H. (1987), “Crafting strategy”, Harvard Business Review, Vol. 64, pp. 66-75. Mirchandani, D. and Pakath, R. (1999), “Four models for a decision support system”, Information & Management, Vol. 35, pp. 31-42. Nakashima, K. (1999), “A design for a management information system with consideration for stochastic variability”, International Journal of Production Economics, Vol. 60-61, pp. 171-6.

Knowledge management for training 393

BPMJ 9,3

394

Nonaka, I., Takeuchi, H. and Umemoto, K. (1996), “A theory of organizational knowledge creation”, International Journal of Technology Management, Vol. 11, pp. 833-45. Parsons, G.L. (1983), “Information technology: a new competitive weapon”, Sloan Management Review, Vol. 3, p. 14. Ragins, B.R. and Cotton, J.L. (1999), “Mentor functions and outcomes: a comparison of men and women in formal and informal mentoring relationships”, Journal of Applied Psychology, Vol. 8 No. 4, pp. 529-50. Truran, W.R. (1998), “Pathways for knowledge: how companies learn through people”, Engineering Management Journal, Vol. 10 No. 4, pp. 15-20. Whitehill, M. (1997), “Knowledge-based strategy to deliver sustained competitive advantage”, Long Range Planning, Vol. 30, pp. 385-91.