Mobile Computing and Sustainable Informatics: Proceedings of ICMCSI 2023 [1 ed.] 9819908345, 9789819908349

Table of contents :
Preface
Contents
About the Editors
Measuring the Technical Efficiency of Thai Rubber Export Using the Spatial Stochastic Frontier Model Under the BCG Concept
1 Introduction
2 Literature Review
3 Data and Methodology
3.1 Data
3.2 Conceptual Framework and Methodology
3.3 Spatial Stochastic Frontier Analysis Models
4 Empirical Results
4.1 Descriptive Information for Observed Spatial Data
4.2 The Result of an Appropriate Model
5 Conclusion and Policy Recommendations
Appendix 1
Appendix 2
Appendix 3
References
Analysis of Digital Data Consumption of Video Streaming Platforms During COVID-19
1 Introduction
2 Methodology
2.1 Inclusion Criteria
2.2 Exclusion Criteria
3 Results
3.1 Background on the Positioning of Streaming Platforms in the Marketplace
3.2 Audience Behaviour with the Platform
3.3 Consumption Drivers
3.4 Netflix Case: Audience Capture Strategies
3.5 Streaming Platforms in Times of Confinement
4 Discussion
5 Conclusion
References
A Prototype of a Wireless Power Transmission System Based on Arduino
1 Introduction
2 Methodology, Design, and Results
3 Conclusion
References
Short Review on Blockchain Technology for Smart City Security
1 Introduction
2 Selection of Papers by PRISMA
2.1 Search String
2.2 PRISMA 2020 Flow Diagram
3 Literature Review
3.1 Networking in a Smart City
3.2 Security Architecture Overview
3.3 A Decentralized Approach to Security Using Blockchain
3.4 Mitigation with AI
3.5 Implementation Feasibility
3.6 Authentication Conditions in Blockchain
3.7 P2P Authentication Workflow
3.8 Security at the Communication Layer
4 Hectic Security Threats in Smart City Environment
4.1 Security at the Communication Layer
4.2 How to Mitigate Security Threats?
5 Case Studies of Security Breaches in Smart Cities
5.1 The Existing Technology of Security in Smart Cities
5.2 The Four Pillars of a Smart City
6 Merits and Demerits of Existing Technology in Smart Cities
6.1 Theoretical Merits of Smart City
6.2 Practical Limitations of Smart City
6.3 Considerations for Building a Smart City
6.4 The Future of Smart Cities
6.5 What Does the Future Hold for Smart Cities?
7 How Blockchain Can Empower Smart Cities?
7.1 Why Use Blockchain for Smart Cities?
7.2 What Applications Can Blockchain Technology Have in a Smart City?
8 Advantages of Decentralizing Security
9 Future Directions
10 Conclusion
References
Efficient Analysis of Sequences of Security Problems in Access Control Systems
1 Introduction
2 State of the Art
3 Problem Statement
4 General Idea of Our Solution
5 Solution
5.1 The Encoder Component
5.2 The Speedup Component
5.3 The Learning Component
6 Experiments
7 Conclusions
References
Artificial Intelligence in Agriculture: Machine Learning Based Early Detection of Insects and Diseases with Environment and Substance Monitoring Using IoT
1 Introduction
2 Literature Review
3 Methodology
3.1 Research Significance
3.2 Algorithm Development
4 Conclusion and Future Studies
References
Design Concepts for Mobile Computing Direction Finding Systems
1 Introduction
2 Development of Direction-Finding Computing Systems
2.1 Concept of the Phase Interferometer
2.2 Development of a Computer System for a Correlation Interferometer
2.3 Operation Algorithm of the Correlation-Interferometric Meter
2.4 Design of a Single-Channel Correlation-Interferometric Meter with Computer Processing of Results
3 Signal Processing Methods for Active Location
3.1 Concepts Comparison of Methods and Problems in Processing Radar Signals
3.2 Estimation of Gain in Relation to Signal/Noise, Resulting from the Compression of a Narrowband Signal
4 Discussion
5 Conclusions
6 Future Scope
References
A Hybrid Machine Learning Model for Urban Mid- and Long-Term Electricity Load Forecasting
1 Introduction
2 Forecasting Model
2.1 Framework and Functions
2.2 Feature Selection
2.3 Ets
2.4 XGBoost
2.5 Ensembling
2.6 Computational Complexity
3 Experiments and Results
3.1 Basic Setup
3.2 Comparison of Forecast Accuracy
4 Conclusion
References
Optimizing Long Short-Term Memory by Improved Teacher Learning-Based Optimization for Ethereum Price Forecasting
1 Introduction
2 Background and Related Works
2.1 Long Short-Term Memory (LSTM) Artificial Neural Network
2.2 Swarm Intelligence and Literature Review
3 Proposed Method
3.1 Teacher Learning-Based Optimization
3.2 Improved TLBO Algorithm
4 Experimental Findings and Results
5 Conclusion
References
A Sophisticated Review on Open Verifiable Health Care System in Cloud
1 Introduction
2 Literature Review
3 Overview of IoMT Network and Its Importance
3.1 Specifications of IoMT
4 Review of Regulations Related to Healthcare Devices
5 Review of IoMT Security Protocols
6 Review of Privacy Preserving Methods for IOMT
7 Review of IoMT Authentication Mechanisms
8 Conclusion
References
Fuzzy Metadata Augmentation for Multimodal Data Classification
1 Introduction
2 Background and Related Work
3 Methodology
3.1 Datasets
3.2 Models and Fuzzy Metadata Augmentation
4 Results
4.1 One Exact Expert Opinion
4.2 Fuzzy Expert Opinion
5 Discussion
6 Conclusions
References
Development of an Information Accuracy Control System
1 Introduction
2 Literature Review and Problem Statement
3 Problem Solution
4 Conclusions
References
Modelling an Efficient Approach to Analyse Clone Phishing and Predict Cyber-Crimes
1 Introduction
2 Related Works
3 Methodology
3.1 Dataset Acquisition
3.2 Deep Multi-view Feature Network Model (DNet)
4 Numerical Results and Discussion
5 Conclusion
References
The Application of Mobile Phones to Enable Traffic Flow Optimisation
1 Introduction
2 Methodology: Systematic Literature Review
3 Results
3.1 Road Surface Conditions
3.2 Classification Approaches
4 Discussion
5 Challenges and Recommendations
6 Conclusion
References
A Hypothesis on Cloud Sourcing Sharing Users' Mobile Devices Through Virtualization
1 Introduction
2 Literature Review
2.1 Mobile Device and User Classification
2.2 Speed of Network Connection
2.3 Battery Life
2.4 Device Hardware Classification
3 Proposed Idea
4 Efficiency of the Idea
5 Conclusion
6 Future Plan
References
An Analytical Assessment of Machine Learning Algorithms for Predicting Campus Placements
1 Introduction
2 Related Work
3 Dataset Formation
4 Proposed Approach
4.1 Artificial Neural Network (ANN)
4.2 AdaBoost Classifier
4.3 Random Forest Classifier
4.4 Logistic Regression
5 Results
6 Conclusion
References
Building Hindi Text Dataset on Stock Market Tweets and Sentiment Analysis Using NLP
1 Introduction
2 Background
3 Architecture of the System
4 Corpus Creation and Description
5 Implementation
6 Results and Discussion
7 Conclusion and Future Scope
References
Fine-Tuning of RoBERTa for Document Classification of ArXiv Dataset
1 Introduction
2 Related Works
3 Methodology
4 Results and Analysis
4.1 Experiments Performed
4.2 Cross-Validation
5 Conclusions and Future Work
References
Comparative Analysis of Using Event Sourcing Approach in Web Application Based on the LAMP Stack
1 Introduction
2 Materials and Methods
2.1 Hardware and System Software to Perform the Experiment
2.2 Experiment Description
3 Experiment Results
3.1 Creating the Records Using Serial Sequence of Requests and Concurrent Requests
3.2 Getting the Aggregated Value from Transactions Using Different Approaches
4 Results Discussion and Conclusions
4.1 Results Discussion
4.2 Conclusions
4.3 Further Research Directions
References
Profitability Improvement for a Distributor Company with Modified Work Posture and Workflow Using REBA and Modelling Simulation
1 Introduction
2 Methodology
2.1 Research Model
3 Result and Discussion
3.1 Work Posture
3.2 Line Balancing
3.3 Activity Cycle Diagram (ACD)
3.4 Causal Loop Diagram
3.5 Simulation
4 Conclusion
References
Demand Forecasting Using Time Series and ANN with Inventory Control to Reduce Bullwhip Effect on Home Appliances Electronics Distributors
1 Introduction
2 Literature Review
2.1 Supply Chain
2.2 Pareto Analysis
2.3 Bullwhip Effect
2.4 Forecasting
2.5 Forecasting Verification and Validation
2.6 Safety Stock
2.7 Distribution Requirement Planning
3 Methods
3.1 Data Collection Technique
3.2 Data Analysis
3.3 Simulation Analysis
4 Data Collection
4.1 Pareto Analysis
4.2 Bullwhip Effect
5 Results and Discussion
5.1 Numerical Results
5.2 Graphical Results
5.3 Implementation of Proposed Improvements
6 Conclusion
References
Memory Malware Identification via Machine Learning
1 Introduction
2 Overview of Malware Types
3 Related Work
4 Research Methodology
4.1 Data Set
4.2 Used Tools
4.3 Data Pre-processing
4.4 Features Extraction
4.5 Machine Learning Algorithms
5 Empirical Results
6 Conclusions and Future Work
References
Basketball Shot Conversion Prediction Using Various ML Techniques and Its Analysis
1 Introduction
1.1 Objectives
1.2 Organization of Paper
2 Literature Survey
3 Methodology
3.1 Discovery
3.2 Data Preparation
3.3 Data Analysis
3.4 Model Planning and Building
3.5 Psuedo-code for Building the Stacking Classifier
3.6 Results
4 Conclusion and Future Work
References
Progressive Web App Implementation in Omah Wayang Klaten Website
1 Introduction
2 Progressive Web App
3 WordPress REST API
4 Methodology
4.1 Establish Prototype Objectives
4.2 Define Prototype Functionality
4.3 Develop Prototype
4.4 Evaluate Prototype
5 Implementation and Discussion
6 Conclusion and Future Studies
References
Use of AI in Cloud-Based Certificate Authentication for Travel Concession
1 Literature Review
2 Introduction
3 Authentication System
3.1 Local Level Authentication
3.2 Government Level Authentication
4 System Design
4.1 Testify Table
4.2 Reference Table
4.3 Flag Table
4.4 Process Model
5 Experimental Results
5.1 Front-End Dashboard
5.2 Front-End Search
5.3 Front-End Reports
6 Summary
7 Conclusion and Future Work
References
Media Player Controller Using Hand Gestures
1 Introduction
2 Literature Review
3 Architecture Design
4 Modules and Implementation
5 Methodology
6 Libraries and Tools
6.1 Visual Studio Code
6.2 Pillow
6.3 Open CV
6.4 Pandas
6.5 NumPy
7 Results and Discussions
8 Limitations
9 Future Scope
10 Conclusion
References
The Detection of Abnormal Behavior in Healthcare IoT Using IDS, CNN, and SVM
1 Introduction
1.1 The Importance of Medical IoT
1.2 The Role of AI in IoHT
2 Related Works
3 Methodology
3.1 Search String Strategy
3.2 Research Questions
3.3 Exclusion and Inclusion Criteria
4 Results
5 Discussion
5.1 Abnormal IoT Behaviors in Health care
5.2 Security Issues That Cause IoHT to Misbehave
5.3 Methods That Are Available to Detect the Abnormal Behavior of IoT in Health Care
5.4 AI Techniques Used to Counter the Abnormal Behavior of IoT in Health Care
6 Conclusion and Recommendation
References
Assessment of Lung Cancer Histology Using Efficient Net
1 Introduction
1.1 Efficientnet
1.2 Related Works
2 Methodology
2.1 Implementation
3 Results
4 Conclusion
References
Baldness Recognition Using Transfer Learning Method Based on Android Application
1 Introduction
2 Literature Review
3 Research Method
4 Results
5 Discussion
6 Conclusion
References
Study of Hybrid Cryptographic Techniques for Vehicle FOTA System
1 Introduction
1.1 Evolution
2 Challenges in FOTA
2.1 Processor Shortcomings
2.2 Power Usage
2.3 Scant Utility
2.4 Untrustworthy Network
2.5 Security
3 Related Work
4 Architecture of FOTA
5 FOTA Update Modes
5.1 Pull Mode
5.2 Push Mode
6 Threats Based on Targets
6.1 IoT Devices
6.2 Mediators
6.3 Manufacturers and Suppliers
7 Analysis
8 Conclusion
References
An Extensive Survey on Machine Learning-Enabled Automated Human Action Recognition Models
1 Introduction
2 General Structure of HAR System
3 Design Issues of HAR Process
4 Review of Recently Developed HAR System
5 Performance Analysis
6 Discussion and Open Issues
7 Conclusion
References
Energy Optimisation in a Cloud Infrastructure Using Ant Colony Optimiser
1 Introduction
2 Related Works
2.1 Energy Consumption
2.2 Load Balancing in Cloud Computing
3 Materials and Methods
3.1 Ant Colony Optimisation (ACO) Algorithm
3.2 The Problem Formulation
3.3 Pseudocode for the Proposed ACO
3.4 Simulation Environment Settings
4 Results and Discussion
4.1 Experiment 1: Energy Consumption
4.2 Experiment 2: CPU and Memory Utilisation
4.3 Experiment 3: Best Path Taken
5 Conclusion
5.1 Threat to Validity
References
The Model of Server Virtualization System Protection in the Educational Institution Local Network
1 Introduction
2 Review and Analysis of the Literature
3 Research Models and Methods
3.1 Information Security Model of the Educational Institution Distributed Network
3.2 Creating an Environment for Network Simulation
4 Experimental Studies of a Secure Network of an Educational Institution
5 Discussion of the Results of Experimental Studies of the Protected Network of the Educational Institution
6 Conclusion
References
Modelling & Optimization of Signals Using Machine Learning Techniques
1 Introduction
2 Background on Machine Learning
2.1 Clustering
2.2 The Principal Component Analysis
2.3 The Naive Bayes
2.4 Support Vector Machines
2.5 Artificial Neural Networks
3 ML in RF Circuit Modelling and Synthesis
4 Conclusion
References
A New Solution for Cyber Security in Big Data Using Machine Learning Approach
1 Introduction
2 Review of the Related Works
3 Methodology
3.1 Performance Evaluation
4 Results and Discussions
5 Conclusion
References
Adaptive Authentication System Based on Unsupervised Learning for Web-Oriented Platforms
1 Introduction
2 Related Work
2.1 Datasets and User Attributes (Features)
2.2 Ensemble Learning
2.3 User Behavior Patterns
2.4 Relevance of the Problem
3 Methodology and Description of the Proposed Outlier Detection System
3.1 Dataset
3.2 Data Vectorization
3.3 Data Preprocessing
3.4 Expanding the Collected Features
3.5 Outlier Detection System on a Website
4 Testing Results and Discussion
4.1 Dataset Analysis
4.2 Training and Test Datasets
4.3 Standard Metrics of the Classifiers with Default Parameters
4.4 Adjustment of OneClasSVM Parameters
4.5 Adjustment of IsolationForest Parameters
4.6 Adjusting the Parameters of EllipticEnvelope
4.7 Identifying the Best Classifier
5 Discussion
6 Conclusion
References
Automated Contactless Attendance System
1 Introduction
2 Literature Survey
3 Existing Systems
4 Proposed System
5 Hardware Setup
6 Proposed System Design
7 Results
8 Conclusion
References
Mobile-Web System for Public Emergency Medical Services Request and Management in Urban and Rural Areas
1 Introduction
2 State of the Art
3 System Architecture
4 Methodology and Design
5 Development and Implementation
5.1 Connection Between Both Modules
5.2 Mobile App Screens
5.3 Admin Panel Screens
6 Results
6.1 The Web Administration Panel
6.2 Mobile App
7 Conclusions
References
Prediction of Dropout Students in Massive Open Online Courses Using Ensemble Learning: A Pilot Study in Post-COVID Academic Session
1 Introduction
2 Related Work
3 Methodology
3.1 Dataset
3.2 Execution of Course
3.3 Analysis of Log Data
3.4 Stacked Ensemble Model for Dropout Prediction
3.5 Performance Metrics
4 Results
5 Discussion
6 Conclusion and Future Work
References
Predictive Maintenance for Remote Field IoT Devices—A Deep Learning and Cloud-Based Approach
1 Introduction
2 Literature Review
3 Data Simulation
4 Data Preprocessing
4.1 Exploratory Data Analysis
4.2 Data Scaling
5 Predictive Analytics Model
5.1 Autoencoder
5.2 Principal Component Analysis (PCA)
6 Tools Description
6.1 Python
6.2 C-SHARP or C#
6.3 Azure IoT Central
6.4 Azure Databricks
6.5 Azure Kubernetes Services
6.6 Azure Event Hub
6.7 Azure Function App
6.8 Azure Blob Storage
6.9 Microsoft Power Automate
7 Implementation Using Tool
7.1 Setting up an Azure Account and Creating Resources
7.2 Configuring the Resources
7.3 Data Simulation Using C# and Azure IoT Central
7.4 Deploying Model into Azure ML Containers
7.5 Exporting Data Using Azure Event Hub
7.6 Creating Azure Function to Predict Pump Failure
8 Performance Measures
9 Results
9.1 Dashboard
9.2 Notification Queue
10 Gap Analysis
11 Conclusion
References
Ubiquitous Learning Environment with Augmented Reality to Stimulate Motor Coordination
1 Introduction
2 State of the Art
2.1 ICT
2.2 Augmented Reality
2.3 Ubiquitous Learning Environment
2.4 Physical Activity
3 Methodology
4 Results
5 Conclusions
References
Identification of Face by Using CNN and Deep Learning Algorithms
1 Introduction
2 Literature Survey
3 Methodologies
3.1 Convolution Layer
3.2 Pool Layer
3.3 Relu Layer
3.4 Fully Connected Layer
4 Proposed Work
5 Results
6 Conclusion and Future Scope
References
Examining the Distribution of Keystroke Dynamics Features on Computer, Tablet and Mobile Phone Platforms
1 Introduction
2 Methodology
3 Result
4 Conclusion
References
Certain Investigations on Eye Diagram Observation for Different Frequencies in Printed Circuit Board
1 Introduction
2 Literature Review
3 Proposed Methodology
4 Results
4.1 Statistical Analysis
5 Conclusion
References
Direct Comparative Analysis of Nature-Inspired Optimization Algorithms on Community Detection Problem in Social Networks
1 Introduction
2 Related Work
3 Nature-Inspired Optimization Algorithms
3.1 Gray Wolf Optimizer (GWO)
3.2 Sine-Cosine Algorithm (SCA)
3.3 Moth-Flame Optimization (MFO)
3.4 Whale Optimization Algorithm (WOA)
4 Community Detection Problem
5 Direct Comparative Analysis
6 Experimental Analysis
6.1 Algorithm Parameter Settings
6.2 Average Isolability
6.3 Result Analysis
7 Conclusion
References
A Handheld Visitor Guidance Device
1 Introduction
2 Literature Survey
3 Methodology
4 Result
5 Conclusion
References
Examination of Different Network Security Monitoring Tools
1 Introduction
2 HTTrack
3 Maltego
4 Nmap
5 Wireshark
6 Burp Suite
7 Bettercap
8 Msfvenom
9 Metasploit
10 Conclusion
References
Designing Climate Control with Fuzzy Logic for Smart Home Systems
1 Introduction
2 Control by Fuzzy Logic
3 Conclusion
References
Detection of Defects in the Railway Tracks Based on YOLOv5
1 Introduction
2 Related Works
3 Components Used
3.1 DC Motors
3.2 Motor Controller
3.3 Types of Motor Controllerz
3.4 Ultrasonic Sensors
3.5 Raspberry Pi
4 Existing System
5 Proposed System
5.1 Yolov5 Algorithm
6 Experimental Analysis
7 Conclusion
References
Game to Ride: Gamification to Salvage Carbon Footprints for Sustainable Development
1 Introduction
2 Literature Review
2.1 Rideshare Platform
2.2 Motivation
2.3 Gamification
3 Research Methodology
4 Research Objective
5 Result and Analysis
6 Discussion
7 Conclusion and Limitations
References
A Comprehensive Survey on Internet of Things Security: Challenges and Solutions
1 Introduction
1.1 Various Threats to Internet of Things Data
2 Security Solutions
2.1 Data Integrity
2.2 Capabilities for Encryption
2.3 Privacy Issues
2.4 Network Security and Security Framework
2.5 Blockchain (BC) Technology
3 Existing Works on Internet of Things Security
4 Surveys of Internet of Things Security Solutions
4.1 Security Solutions Using BC
4.2 Security Solutions Using Machine Learning
4.3 Security Solutions Using Trust and Reputation
4.4 Security Solutions Using Cryptography
5 Comparison of Security Solutions
6 Research Challenges
7 Conclusion and Future Directions
References
CNN-Based Detection of Cracks and Moulds in Buildings
1 Introduction
2 Related Works
3 Algorithms Used
3.1 Inception-v3
3.2 VGG-16
3.3 Resnet-50
4 Proposed Process
4.1 Dataset Collection and Pre-processing
4.2 Implementation
5 Results and Discussion
5.1 Accuracy and Metrics
5.2 Class Activation Mapping
6 Conclusion
References
Biogeography-Based Optimization for Lifetime Enhancement in Wireless Sensor Network
1 Introduction
2 Related Work
2.1 Energy-Efficient Clustering Protocols
2.2 Biogeography-Based Optimization
3 System Architecture
3.1 Three-Level Heterogeneous Wireless Sensor Network
3.2 Energy Model of a Sensor Node
3.3 Objective Function
4 Proposed Biogeography-Based Optimization for Energy-Efficient Clustering Protocol in HWSN
4.1 General Description
4.2 Island Encoding
4.3 Island Population Initialization
4.4 Migration Operator
4.5 Mutation Operator
5 Simulation Results and Discussion
6 Conclusion and Perspectives
References
Internet of Things and Smart Intelligence-Based Google Assistant Voice Controller for Wheelchair
1 Introduction
2 Literature Survey
3 Existing Methodology
4 Proposed Methodology
4.1 Implementing Techniques
5 Results and Discussions
5.1 IFTTT Applets and Output
5.2 Google Assistant Input and Its Response
5.3 Adafruit.IO Output Result
5.4 Optimized Working System
6 Conclusion
7 Future Enhancement
References
Increasing the Immunity of Information Transmission and Fault Tolerance of the Path
1 Introduction
2 Literature Review and Problem Statement
3 Solution of the Problem
4 Conclusions
References
PLAN: Indoor Positioning Using Bluetooth with Received Signal Strength Indicator
1 Introduction
2 Related Works
2.1 Indoor Localisation
2.2 Accuracy Estimation
3 Overview of PLAN
4 Localisation Using PLAN
5 Crowdsourcing the Radio Map
5.1 The PLAN Identifier
6 Accurate Positioning and Navigation
6.1 The PLAN Director
7 Adding Buildings and Navigation
7.1 The PLAN Planner
7.2 Information Exploration in PLAN
7.3 Information Stack
8 Observation of the Indoor Mappings
8.1 The PLAN Observer
9 Results
10 Conclusion
References
Author Index

Citation preview

Lecture Notes on Data Engineering and Communications Technologies 166

Subarna Shakya George Papakostas Khaled A. Kamel   Editors

Mobile Computing and Sustainable Informatics Proceedings of ICMCSI 2023

Lecture Notes on Data Engineering and Communications Technologies Volume 166

Series Editor Fatos Xhafa, Technical University of Catalonia, Barcelona, Spain

The aim of the book series is to present cutting edge engineering approaches to data technologies and communications. It will publish latest advances on the engineering task of building and deploying distributed, scalable and reliable data infrastructures and communication systems. The series will have a prominent applied focus on data technologies and communications with aim to promote the bridging from fundamental research on data science and networking to data engineering and communications that lead to industry products, business knowledge and standardisation. Indexed by SCOPUS, INSPEC, EI Compendex. All books published in the series are submitted for consideration in Web of Science.

Subarna Shakya · George Papakostas · Khaled A. Kamel Editors

Mobile Computing and Sustainable Informatics Proceedings of ICMCSI 2023

Editors Subarna Shakya Pulchowk Campus Institute of Engineering Tribhuvan University Kathmandu, Nepal

George Papakostas MLV Research Group Department of Computer Science International Hellenic University Kavala, Greece

Khaled A. Kamel Department of Computer Science Texas Southern University Houston, TX, USA

ISSN 2367-4512 ISSN 2367-4520 (electronic) Lecture Notes on Data Engineering and Communications Technologies ISBN 978-981-99-0834-9 ISBN 978-981-99-0835-6 (eBook) https://doi.org/10.1007/978-981-99-0835-6 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

We are privileged to dedicate the proceedings of ICMCSI 2023 to all the participants and editors of ICMCSI 2023.

Preface

This Conference Proceedings volume contains the written versions of most of the contributions presented during the 4th International Conference on Mobile Computing and Sustainable Informatics (ICMCSI) 2023 held in Tribhuvan University, Nepal, 11–12, January 2023. The conference provided a setting for discussing recent developments in a wide variety of topics including Mobile Computing, Cloud Computing and Sustainable expert systems. The conference has been a good opportunity for participants coming from various destinations to present and discuss topics in their respective research areas. ICMCSI 2023 Conference tends to collect the latest research results and applications on Mobile Computing, Cloud Computing and Sustainable expert systems. It includes a selection of 56 papers from 305 papers submitted to the conference from universities and industries all over the world. All accepted papers were subjected to a strict peer-review system by 2–4 expert referees. The papers have been selected for this volume because of their quality and relevance to the conference. ICMCSI 2023 would like to express our sincere appreciation to all authors for their contributions to this book. We would like to extend our thanks to all the referees for their constructive comments on all papers, especially, we would like to thank the organizing committee for their hard work. We thank the keynote speaker Dr. Danilo Pelusi, Faculty of Communication Sciences, University of Teramo, Italy, for his valuable thoughts. Finally, we would like to thank Springer publications for producing this volume. Kathmandu, Nepal Kavala, Greece Houston, USA

Prof. Dr. Subarna Shakya Dr. George Papakostas Dr. Khaled A. Kamel

vii

Contents

Measuring the Technical Efficiency of Thai Rubber Export Using the Spatial Stochastic Frontier Model Under the BCG Concept . . . . . . . . Chanamart Intapan and Chukiat Chaiboonsri Analysis of Digital Data Consumption of Video Streaming Platforms During COVID-19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lizeth Aracelly Lopez-Orosco, Valeria Alexandra Solano-Guevara, Adriana Margarita Turriate-Guzman, and Luis-Rolando Alarcón-Llontop A Prototype of a Wireless Power Transmission System Based on Arduino . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Md. Rawshan Habib, Ahmed Yousuf Suhan, Md. Shahnewaz Tanvir, Abhishek Vadher, Md. Mossihur Rahman, Shuva Dasgupta Avi, and Shabuj Dasgupta Ami Short Review on Blockchain Technology for Smart City Security . . . . . . . Alanoud Alquwayzani and M. M. Hafizur Rahman

1

23

37

45

Efficient Analysis of Sequences of Security Problems in Access Control Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anh Tuan Truong

67

Artificial Intelligence in Agriculture: Machine Learning Based Early Detection of Insects and Diseases with Environment and Substance Monitoring Using IoT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Gnana Rajesh, Yaqoob Yousuf Said Al Awfi, and Murshid Qasim Mohammed Almaawali

81

Design Concepts for Mobile Computing Direction Finding Systems . . . . Juliy Boiko, Oleksıy Polıkarovskykh, Vitalii Tkachuk, Hanna Yehoshyna, and Lesya Karpova

89

ix

x

Contents

A Hybrid Machine Learning Model for Urban Mid- and Long-Term Electricity Load Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Xianghua Tang, Zhihui Jiang, Lijuan Zhang, Jiayu Wang, Youran Zhang, and Liping Zhang Optimizing Long Short-Term Memory by Improved Teacher Learning-Based Optimization for Ethereum Price Forecasting . . . . . . . . . 125 Marija Milicevic, Luka Jovanovic, Nebojsa Bacanin, Miodrag Zivkovic, Dejan Jovanovic, Milos Antonijevic, Nikola Savanovic, and Ivana Strumberger A Sophisticated Review on Open Verifiable Health Care System in Cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Mandava Varshini, Kolla Akshaya, Katakam Krishna Premika, and A. Vijaya Kumar Fuzzy Metadata Augmentation for Multimodal Data Classification . . . . . 157 Yuri Gordienko, Maksym Shulha, Yuriy Kochura, Oleksandr Rokovyi, Oleg Alienin, and Sergii Stirenko Development of an Information Accuracy Control System . . . . . . . . . . . . . 173 A. M. Mehdiyeva, I. Z. Sardarova, and Z. A. Mahmudova Modelling an Efficient Approach to Analyse Clone Phishing and Predict Cyber-Crimes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Bondili Sri Harsha Sai Singh, Mohammed Fathima, Mohammad Sameer, Thota Teja Mahesh, A. Dinesh Kumar, and K. Padmanaban The Application of Mobile Phones to Enable Traffic Flow Optimisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 T. Shilowa, J. P. van Deventer, and M. J. Hattingh A Hypothesis on Cloud Sourcing Sharing Users’ Mobile Devices Through Virtualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Nazmus Sakib and Al Hasib Mahamud An Analytical Assessment of Machine Learning Algorithms for Predicting Campus Placements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 S. Bala Dhanalakshmi, Raksheka Rajakumar, Swetha Shankar, R. Sowndharya Rani, N. Deepa, and C. Subha Priyadharshini Building Hindi Text Dataset on Stock Market Tweets and Sentiment Analysis Using NLP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Choudhary Anushka, Gupta Mohit, and S. K. Lavanya Fine-Tuning of RoBERTa for Document Classification of ArXiv Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Kshetraphal Bohara, Aman Shakya, and Bishal Debb Pande

Contents

xi

Comparative Analysis of Using Event Sourcing Approach in Web Application Based on the LAMP Stack . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Marian Slabinoha, Stepan Melnychuk, Vitalia Kropyvnytska, and Bohdan Pashkovskyi Profitability Improvement for a Distributor Company with Modified Work Posture and Workflow Using REBA and Modelling Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 Louis Valentino, Lina Gozali, Frans Jusuf Daywin, and Ariawan Gunadi Demand Forecasting Using Time Series and ANN with Inventory Control to Reduce Bullwhip Effect on Home Appliances Electronics Distributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 Stiven Tjen, Lina Gozali, Helena Juliana Kristina, Ariawan Gunadi, and Agustinus Purna Irawan Memory Malware Identification via Machine Learning . . . . . . . . . . . . . . . 301 Maysa Khalil and Qasem Abu Al-Haija Basketball Shot Conversion Prediction Using Various ML Techniques and Its Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 Sanyam Raina, Shreedhar Bhatt, Vaidehi Shah, Heem Amin, Vinay Khilwani, and Samir Patel Progressive Web App Implementation in Omah Wayang Klaten Website . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Budi Susanto, Gloria Virginia, Umi Proboyekti, and Jeysy Carmila Dewi Ester Use of AI in Cloud-Based Certificate Authentication for Travel Concession . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349 Dangwani Avinash, Jetawat Ashok Kumar, and Rawat Chandansingh Media Player Controller Using Hand Gestures . . . . . . . . . . . . . . . . . . . . . . . 363 Vijay Mane, Harshal Baru, Abhishek Kashid, Prasanna Kshirsagar, Aniket Kulkarni, and Prathamesh Londe The Detection of Abnormal Behavior in Healthcare IoT Using IDS, CNN, and SVM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 Oluwaseun Priscilla Olawale and Sahar Ebadinezhad Assessment of Lung Cancer Histology Using Efficient Net . . . . . . . . . . . . . 395 Vishal Giraddi, Shantala Giraddi, Suvarna Kanakaraddi, and Mahesh Patil Baldness Recognition Using Transfer Learning Method Based on Android Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 Ary Kurniadi Irawan, Rangga Atmajaya, Fajrin Hardinandar, and Ari Satriadi

xii

Contents

Study of Hybrid Cryptographic Techniques for Vehicle FOTA System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417 Manas Borse, Parth Shendkar, Yash Undre, Atharva Mahadik, and Rachana Patil An Extensive Survey on Machine Learning-Enabled Automated Human Action Recognition Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431 Lakshmi Alekhya Jandhyam, Ragupathy Rengaswamy, and Narayana Satyala Energy Optimisation in a Cloud Infrastructure Using Ant Colony Optimiser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445 Ebenezer Owusu, Godwin Banafo Akrong, Justice Kwame Appati, and Solomon Mensah The Model of Server Virtualization System Protection in the Educational Institution Local Network . . . . . . . . . . . . . . . . . . . . . . . . 461 V. Lakhno, B. Akhmetov, B. Yagaliyeva, O. Kryvoruchko, A. Desiatko, S. Tsiutsiura, and M. Tsiutsiura Modelling & Optimization of Signals Using Machine Learning Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477 P. Shanthi, Adish, Bhuvana Shivashankar, and A. Trisha A New Solution for Cyber Security in Big Data Using Machine Learning Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495 Romil Rawat, Olukayode A. Oki, K. Sakthidasan Sankaran, Oyebola Olasupo, Godwin Nse Ebong, and Sunday Adeola Ajagbe Adaptive Authentication System Based on Unsupervised Learning for Web-Oriented Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507 Andrey Y. Iskhakov, Yana Y. Khazanova, Mark V. Mamchenko, Roman V. Meshcheryakov, Anastasia O. Iskhakova, and Sergey P. Khripunov Automated Contactless Attendance System . . . . . . . . . . . . . . . . . . . . . . . . . . 523 B. Bhuvaneshwari, H. Ranga Krishna Prasadh, P. Sathish, V. Anuja, and A. Mythily Mobile-Web System for Public Emergency Medical Services Request and Management in Urban and Rural Areas . . . . . . . . . . . . . . . . . 533 Juventino Aguilar-Correa, Raúl Meneses-Leal, David G. Pasillas-Banda, Rodolfo Omar Domínguez-García, Yehoshua Aguilar-Molina, and Miriam González-Dueñas Prediction of Dropout Students in Massive Open Online Courses Using Ensemble Learning: A Pilot Study in Post-COVID Academic Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 549 Rizwan Alam, Naeem Ahmad, Sana Shahab, and Mohd Anjum

Contents

xiii

Predictive Maintenance for Remote Field IoT Devices—A Deep Learning and Cloud-Based Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567 A. Kannammal, M. Guhanesvar, and R. R. Venketesz Ubiquitous Learning Environment with Augmented Reality to Stimulate Motor Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587 German Sailema-Lalaleo and Cristina Páez-Quinde Identification of Face by Using CNN and Deep Learning Algorithms . . . 599 Hari Padmavathi Madala, Dharani Daparti, Arepalli Gopi, and P. V. Sivarambabu Examining the Distribution of Keystroke Dynamics Features on Computer, Tablet and Mobile Phone Platforms . . . . . . . . . . . . . . . . . . . . 613 Olasupo Oyebola Certain Investigations on Eye Diagram Observation for Different Frequencies in Printed Circuit Board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621 A. Shan and V. R. Prakash Direct Comparative Analysis of Nature-Inspired Optimization Algorithms on Community Detection Problem in Social Networks . . . . . . 629 Soumita Das, Bijita Singha, Alberto Tonda, and Anupam Biswas A Handheld Visitor Guidance Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643 S. R. Bhagyashree, A. Jain Sukrutha, R. Pooja, N. Sheetal, and B. S. Swathi Examination of Different Network Security Monitoring Tools . . . . . . . . . 653 Syed Maaz, Deepak Kumar Sinha, and Garima Sinha Designing Climate Control with Fuzzy Logic for Smart Home Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667 Ahmad Valiyev, Rahib Imamguluyev, and Rena Mikayilova Detection of Defects in the Railway Tracks Based on YOLOv5 . . . . . . . . . 677 T. Sangeetha, M. Mohanapriya, and P. Prakasham Game to Ride: Gamification to Salvage Carbon Footprints for Sustainable Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695 Swati Tayal, K. Rajagopal, and Vaishali Mahajan A Comprehensive Survey on Internet of Things Security: Challenges and Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 711 Nilima Karankar and Anita Seth CNN-Based Detection of Cracks and Moulds in Buildings . . . . . . . . . . . . . 729 V. Maheysh and S. Kirthica

xiv

Contents

Biogeography-Based Optimization for Lifetime Enhancement in Wireless Sensor Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 745 Chahla Mansour, Houssem Eddine Nouri, and Olfa Belkahla Driss Internet of Things and Smart Intelligence-Based Google Assistant Voice Controller for Wheelchair . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 761 K. Muthulakshmi, C. Padmavathy, N. Kirthika, B. Vidhya, and M. A. P. Manimekalai Increasing the Immunity of Information Transmission and Fault Tolerance of the Path . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775 A. M. Mehdiyeva, I. N. Bakhtiyarov, and S. V. Bakhshaliyeva PLAN: Indoor Positioning Using Bluetooth with Received Signal Strength Indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 785 K. Deepika and Prasad B. Renuka Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 797

About the Editors

Subarna Shakya is currently Professor of Computer Engineering, Department of Electronics and Computer Engineering, Central Campus, Institute of Engineering, Pulchowk, Tribhuvan University, and Coordinator (IOE), LEADER Project (Links in Europe and Asia for engineering, eDucation, Enterprise and Research exchanges), ERASMUS MUNDUS. He received M.Sc. and Ph.D. degrees in Computer Engineering from the Lviv Polytechnic National University, Ukraine, 1996 and 2000, respectively. His research area includes E-government system, computer systems & simulation, distributed and cloud computing, software engineering and information system, computer architecture, information security for E-government, and multimedia system. George Papakostas received the Diploma degree in electrical and computer engineering in 1999 and the M.Sc. and Ph.D. degrees in electrical and computer engineering in 2002 and 2007, respectively, from the Democritus University of Thrace (DUTH), Greece. From 2007 to 2010, he served as Adjunct Lecturer with the Department of Production Engineering and Management, DUTH. He currently serves as Adjunct Assistant Professor with the Department of Computer and Informatics Engineering, Technological Educational Institution, Eastern Macedonia and Thrace, Greece. In 2012, he was elected as Full Professor in the aforementioned Department of Computer and Informatics Engineering. He has co-authored more than 70 publications in indexed journals, international conferences, and chapters. His research interests include pattern recognition, computer/machine vision, computational intelligence, machine learning, feature extraction, evolutionary optimization, and signal and image processing. He served as Reviewer in numerous journals and conferences, and he is Member of the IAENG, MIR Labs, EUCogIII, and the Technical Chamber of Greece. Khaled A. Kamel is currently Chairman and Professor at Texas Southern university, College of Science and technology, Department of Computer Science, Houston, TX. He has published many research articles in refereed journals and IEEE conferences. He has more than 30 years of teaching and research experience. He has been xv

xvi

About the Editors

General Chair, Session Chair, TPC Chair, and Panelist in several conferences and acted as Reviewer and Guest Editor in referred journals. His research interest includes networks, computing, and communication systems.

Measuring the Technical Efficiency of Thai Rubber Export Using the Spatial Stochastic Frontier Model Under the BCG Concept Chanamart Intapan and Chukiat Chaiboonsri

Abstract The Bio-Circular-Green economy (BCG) concept was originally started by the Thai government to promote national development and post-pandemic recovery in 2021. This study concentrates on the ways to increase the effectiveness of Thailand’s natural rubber exports. The primary goal is to evaluate Thailand’s natural rubber exports to ASEAN nations, in terms of their technical efficiency rankings. In order to achieve the main objectives based on the BCG concept (BCG policy measures in number 10: “investing in infrastructure”), the spatial dataset is applied with a stochastic frontier analysis model, which is called the panel spatial stochastic frontier analysis model estimation. The empirical results of this study to improve the technical efficiency of Thailand’s rubber export found that the infrastructure, especially the logistic system requirements of CLMV countries, needs to be addressed first. This is because the mixed spatial matrix (mixed-wij ) represents significantly the level of logistics system development, which plays an important role in sustainably improving the technical efficiency. Therefore, the government and private sectors can use these empirical findings to promote policy recommendations in agricultural economics, especially the investment in logistic systems’ aspects of low carbon emissions and using renewable energy, which is the BCG concept. Keywords BCG · Thailand’s natural rubber exports · ASEAN countries · Infrastructure · Agricultural · Economics · Spatial · Technical efficiency

1 Introduction The BCG (Bio-Circular-Green Economy) concept was originally started by the Thai government to promote national development and post-pandemic recovery in 2021.1 1

https://www.bcg.in.th/eng/background/.

C. Intapan · C. Chaiboonsri (B) Modern Quantitative Economic Research Centre (MQERC), Faculty of Economics, Chiang Mai University, Chiang Mai, Thailand e-mail: [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 S. Shakya et al. (eds.), Mobile Computing and Sustainable Informatics, Lecture Notes on Data Engineering and Communications Technologies 166, https://doi.org/10.1007/978-981-99-0835-6_1

1

2

C. Intapan and C. Chaiboonsri

The BCG concept consists of the three element pillars such as Bioeconomy, Circular Economy, and Green Economy. In addition, the BCG model emphasizes the use of science, technology, and innovation to turn Thailand’s comparative advantage in biological and cultural diversity into a competitive advantage, with a focus on four strategic sectors: agriculture and food, wellness and medicine, energy, materials, and biochemicals, and tourism and creative economy. However, this research article focuses on the first four strategic sectors (agriculture only) regarding the main issues in the promotion of agriculture product especially Thailand’s rubber export to ASEAN countries by technical efficiency improvement based on spatial analysis. In the Royal speech of His Majesty, the King said “Our economy has always been dependent on the agriculture sector. The income of the country that has been used to create prosperity in various fields is mostly income from agriculture. So, it can be said that the prosperity of the country depends on the prosperity of agriculture. Moreover, the work of all parties can progress because our agriculture is prosperous” (Source: Office of the Royal Development Projects Board (ORDPB)). The previous statement clearly shows the importance of the agricultural sector to the nation and the people of Thailand. This is due to various reasons that the agricultural sector is important to the prosperity of the country. Agriculture has been the basic occupation of Thai people of every age. About two-thirds of the population is in the agricultural sector. Agricultural development has always been an important goal of national development. In addition, the field of agriculture is a highly important subject in every issue of the National Economic and Social Development Plan. Besides the importance of agricultural sector within the country, it also plays an enormous role in international trade. This is because the majority of the income generated from international trade, especially from exports, is income from the agricultural trade. Figure 1 shows the export value of Thai agricultural products. It was discovered that the growth rate for Thailand’s agricultural exports from 1995 to 2020 was, on average, between 16 and 17% each year (Source: Bank of Thailand). From Fig. 1, it can be seen that the value of Thai exports, especially agricultural exports, has created enormous value. It also tends to grow further in the future. Then, when studying more deeply about the situation of exporting Thai agricultural products, it was found that the main market for exporting agricultural products is the ASEAN market. The agricultural market in ASEAN is considered the future of Thai agricultural products due to its continuous growth. The agricultural market accounted for the largest export share of 24% of the total export market. At the same time, the government has policies to promote farmers such as reducing production costs, promoting large plantings, focusing on the integration of production groups, including the policy of developing water resources. From the policy to support all farmers, agricultural products have a bright future (Source: Department of Agriculture Extension). Figure 2 displays the value of Thai agricultural exports to ASEAN-9, with an average annual growth rate of 13–14% from 1990 to 2020 (Source: Bank of Thailand). Additionally, Fig. 2 indicates that the trend line for the value of Thai agricultural exports to ASEAN is likely to increase steadily. As for agricultural trade with ASEAN in 2020 compared with the same period of 2019, it was found that Thai agricultural

80,000.00 60,000.00 40,000.00 20,000.00 0.00

3

01-1995 03-1996 05-1997 07-1998 09-1999 11-2000 01-2002 03-2003 05-2004 07-2005 09-2006 11-2007 01-2009 03-2010 05-2011 07-2012 09-2013 11-2014 01-2016 03-2017 05-2018 07-2019 09-2020

THB MN

Measuring the Technical Efficiency of Thai Rubber Export Using …

DATE Fig. 1 The export value of Thai agricultural products from 1995 to 2020 (Unit: THB million, (on average is 30–31 THB:1 US$)). Source Bank of Thailand

10-1990 01-1992 04-1993 07-1994 10-1995 01-1997 04-1998 07-1999 10-2000 01-2002 04-2003 07-2004 10-2005 01-2007 04-2008 07-2009 10-2010 01-2012 04-2013 07-2014 10-2015 01-2017 04-2018 07-2019 10-2020

THB MN

20,000.00 18,000.00 16,000.00 14,000.00 12,000.00 10,000.00 8,000.00 6,000.00 4,000.00 2,000.00 0.00

DATE

Fig. 2 The value of the Thai agricultural products exported to ASEAN-9 from 1990 to 2020 (on average is 30–31 THB:1 US$)). Source Bank of Thailand

trade to ASEAN has expanded, with a total trade value of 421,977 million baht (Source: Ministry of Commerce). Although in the first half of 2020, the world faced the Covid-19 pandemic, it can be seen that the trade in Thai agricultural products in 2020 is still considered a good direction. Despite the new outbreak, consumers are starting to adapt, and countries have eased more restrictions on imports, as well as the readiness of Thailand to control the epidemic has improved. For the reasons mentioned earlier, Thailand has more opportunities and trends in exports as well. Moreover, it was found that Thailand has a trade surplus advantage in agricultural trade with ASEAN, which can be valued at 171,334 million baht. The top three markets in ASEAN where Thailand exported the most agricultural products were Vietnam, Malaysia, and Cambodia. One of the high-value agricultural exports is natural rubber, with an export value of approximately 23,271 million baht (Source: International Agricultural Economics Division). Therefore, as the agricultural sector becomes more involved in the export sector, it will result in the export sector, which is considered the heart of the Thai economy, to generate enormous income for the country each year. The income from

4

C. Intapan and C. Chaiboonsri

Fig. 3 The export value of the Thai natural rubber from 2011 to 2020 (Unit: THB, (on average is 30–31 THB:1 US$)) (Source Department of International Trade Promotion, Bureau of Agricultural and Industrial Trade Promotion of Thailand)

Fig. 4 The export volume of the Thai natural rubber from 2011 to 2020 (Unit: kg) (Source Department of International Trade Promotion, Bureau of Agricultural and Industrial Trade Promotion of Thailand)

the export of these agricultural products will eventually return to the development of the country, creating the welfare of the people, including farmers (Source: Ministry of Commerce). Because rubber is an agricultural export product, it is important to create a great value for Thai exports, which can be seen in Figs. 3 and 4. These figures show the value and volume of Thai rubber exports, respectively. When penetrating into the main market where Thailand has the most rubber exports, it is found that Thailand has a large rubber export value based on the value of its exports to China and ASEAN countries. The value of exports to the ASEAN-8 countries is shown in Fig. 5. In addition, the report on the situation of rubber exports from January to February 2020

Measuring the Technical Efficiency of Thai Rubber Export Using …

5

Fig. 5 The value of the Thai natural rubber exporting to ASEAN-8 since 2010–2020 (Unit: THB million, (on average is 30–31 THB:1 US$)) (Source Department of International Trade Promotion, Bureau of Agricultural and Industrial Trade Promotion)

by the Department of International Trade Promotion has been presented to show that Thailand’s rubber exports have increased. This is due to both the automotive sector that uses rubber as a component to produce parts and tires, as well as the increasing demand for rubber gloves in the global market due to the impact of the Covid-19 epidemic. Rubber gloves are in demand all over the world right now. As a result of this situation, demand for rubber products has continued to expand throughout the world. Business Wire forecasts that by 2027 the size of the global rubber gloves market is expected to grow to $22.1 billion. From 2020 to 2027, the growth rate is expected to grow at 14.7% (Source: “Rubber Gloves Market Size, Share and Trends Analysis Report by Material (Latex, Nitrile, Neoprene), by Type, by Product, by Distribution Channel, by End Use, by Region, and Segment Forecasts, 2020–2027”). From these events that resulted in increased need for rubber, prices tend to improve from the previous year (source: Department of International Trade Promotion, Bureau of Agricultural and Industrial Trade Promotion). From the analysis of the situation of rubber exports that tends to improve, making it possible to clarify the opportunities or strengths of Thai rubber exports that Thai rubber products are recognized worldwide as Thailand is recognized from around the world as the world’s number 1 producer and exporter of processed rubber, which can be observed from both the production volume and the volume of export including the value of rubber exports, in addition, coupled with the fact that Thai rubber is of high quality and standard. In addition, Thailand has taken care of and promoted the rubber industry as a whole supply chain, starting from the farmers themselves who have the knowledge, expertise, and high experience, coupled with the presence of an agency that specifically takes care of the rubber industry, making the supply chain system as the most

6

C. Intapan and C. Chaiboonsri

efficient tire industry. However, the Thai rubber industry is not only an aspect of opportunities or strengths but also an aspect of the problems and obstacles that Thai rubber entrepreneurs will face, because most Thai rubber is mainly export-oriented. As a result, when the world economy slows down, it will directly affect Thai exports. Moreover, at present, many countries have begun to grow more rubber, especially the CLMV countries, causing the global rubber supply to exceed the market demand. From this, although Thailand is known as the number one country in rubber exports, at the same time, there are many obstacles. These obstacles include the area or the country that is an exporting country of Thailand that exports rubber as well. Therefore, from all the above reasons, the organizers have prepared this research to answer the question of whether the rubber exports are spatial factors or not, especially affecting the export of Thai rubber to ASEAN, which is the main market. In other words, this work studies the efficiency of Thai rubber exports to each ASEAN country. Figure 6 shows the value of rubber exports to each ASEAN country, and it can be seen that Thailand has different rubber exports to each ASEAN country. The countries that acquire the highest value of rubber exports from Thailand are Singapore, Malaysia, and the Philippines, respectively. All the above is the reason why the spatial econometric model is crucially needed. In order to confirm the plan number 10 (investment in infrastructure (logistics system based on BCG concept (low carbon emissions and using renewable energy))) to promote Thailand’s rubber export’s sustainable future, this research study uses the spatial analysis to evaluate the technical efficiency to push this export more effectively. The objectives of this research study are to find the best model to support Thailand’s rubber export product in order to increase its efficiency using a proposal

Fig. 6 The value of the Thai Natural Rubber exporting to ASEAN-8 from 2010 to 2020 by country (Unit: THB million (on average is 30–31 THB:1 US$) (Source Department of International Trade Promotion, Bureau of Agricultural and Industrial Trade Promotion of Thailand)

Measuring the Technical Efficiency of Thai Rubber Export Using …

7

from an econometrics model. This analysis uses the stochastic frontier of Thailand’s rubber export product based on three models, such as the non-spatial model, the mixed model, and the spatial model. This research study’s results are an evaluation of all models using the Technical Efficiency (TE) formula. Furthermore, the study’s anticipated results continue to support BCG-based policy recommendations.

2 Literature Review Research studies on the agricultural sector around the world are widely studied, especially on the topic of technical efficiency scores in the agricultural sector. Most of the research has been done using well-known and popular tools such as Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA). The tool called the SFA was used by Latruffe et al. [8] to investigate the technical efficiency of crop and livestock production in Poland. Then, the score of efficiency from the SFA method was compared with the DEA method. The results illustrated that the technical efficiency score of crop farms had a lower average technical efficiency score. This means that the livestock farms are more technically efficient than the crop farms. The results from the SFA method were consistent with the results from the DEA method. Moreover, the results showed that the key factors affecting the crop farms’ efficiency are land and labor. Family labor own land is the key factor affecting the livestock farms. In addition, education is a barrier to efficiency, especially for crop farms. In the case of Thailand, a nonparametric approach was selected for measuring the technical efficiency of rice farms in Central Thailand by Taraka et al. [13]. The work investigated the technical efficiency of rice farms in Central Thailand. Technical Efficiency was evaluated by using the nonparametric approach, the DEA on 400 rice farms in-crop year 2009/2010. The results indicated that most of farmers tend to operate at a lower level of technical efficiency since the range of technical efficiency can be estimated between 0.30 and 100%. In addition, the study also found a positive correlation between farm efficiency and family labor, extension officer’s service, certified seed use, and pest control on weedy rice and insect. After that in 2012, SFA method was employed for measuring the technical efficiency of rice farms by Taraka et al. [14]. The technical efficiency of rice farmers in the central region of Thailand was measured and the factors causing technical inefficiency were identified using SFA approach. The mean of technical efficiency was about 85.35%. Moreover, SFA method can be used to identify the key factors affecting farm’s efficiency. The results found that gender, farming experience, Good Agricultural Practices (GAP), and cropping intensity have positive correlation toward farm technical efficiency in the case of Thailand. Zamanian et al. [15] analyzed the technical efficiency in crop and livestock production. Technical efficiency is estimated with SFA. The SFA results were compared with results using DEA. On average, livestock farms are more technically efficient than crop farms. Moreover, the factors that affect the level of technical efficiency in crop and livestock production were also examined. The results showed that land and

8

C. Intapan and C. Chaiboonsri

labor are important for crop farms, while livestock farms can rely on family labor and own land. Also, education is a constraint to efficiency particularly for crop farms. The latest research study by Melati and Mayninda [11] evaluated the efficiency of East Java rice production by employing the SFA model. The empirical results of the findings indicated that the efficiency of rice production in East Java in 2018 was said to be very efficient in several districts. However, some of the tools which past researchers chose have some flaws. Therefore, the tool called Bootstrapping DEA is used to calculate the technical efficiency scores of Thai natural rubber productions. This is the main difference between previous research and this research. In the agricultural sector, besides the above-mentioned, spatial studies are important and applied in the agricultural sector as well. Langyintuo and Mekuria [9], Maertens and Barrett [10], and Pede et al. [12] proved that in agriculture, neighborhood interactions have primarily been pinpointed for investigating the drivers of technology adoption in agricultural production. Especially, Pede et al. [12] used spatial econometrics to study spatial problems and propose solutions for urban or local centers and regions. Agricultural products must have the most spatial dependence. Spatial dependency has been used to measure regional yields of a large number of commodities, particularly when applied to agricultural commodities. This is because spatial studies are the main factor in decision-making that will affect the yield or the choice of farming for different areas. The spatial studies have been applied to the agricultural sector in the past, such as in Cambardella et al. [4], Bucknell et al. [3], Griffin [6], and Brindha and Elango [2]. Cambardella et al. [3] studied the topic of spatial analysis of soil fertility parameters. The study examined spatial patterns for nine soil chemical properties in two adjacent fields. The findings showed that soil properties with strong spatial correlations and the maximum distance to which those properties were correlated, differed for the two fields. Bucknell et al. [3] investigated land-use change and agriculture in Eastern Canada by using spatial analysis. The aim was to provide a quantitative basis for the discussion of rural policy issues. The results found that rural policy should orient its geographical delineation to scales. It was suggested that applying a rural landscape design to the entire region would help address the sustainability of agriculture and rural communities. Moreover, spatial regression methods were adapted by Griffin [6] to on-farm trials in the context of farm management. The result found that farmers will have more confidence in farm management decisions when they receive a spatial analysis report. Brindha and Elango [2] studied the topic of spatial analysis of soil fertility in India. The soil fertility was analyzed in an agricultural area. This study will help to choose the type of crop that would be suitable for plantation in this area and can help to compose agrochemicals that will be required for this area can also be decided. Furthermore, this study serves as baseline information to improve the soil fertility in this intensively irrigated area. Moreover, the spatial study has also been applied to the most recent situation in the world, the Covid-19 epidemic. Franch Pardo et al. [5] studied about the geographical

Measuring the Technical Efficiency of Thai Rubber Export Using …

9

dimension of the 2019 coronavirus disease pandemic by using geospatial and spatialstatistical analysis. The results found that understanding the spatial analysis of Covid19 is important for solving the problem of Covid-19 because it helps to clarify the scope and impact of outbreaks and can help in decision-making, planning, and implementation. However, the review of the past studies found that spatial studies are not widely applied to the agriculture. This study uses this gap to bring more spatial analysis to the agricultural sector. From the review of the past research, it can be concluded that the spatial study is a hugely useful tool to be applied to research studies in all disciplines, especially in agriculture. Furthermore, the difference of this study from the other studies is that in this study the Bayesian inference concept is also applied to the spatial panel econometrics to determine the efficiency score and the factors affecting the export of Thai natural rubber productions.

3 Data and Methodology 3.1 Data The collected data used in this study are presented in Table 1. In this study, two variables were identified such as independent variables and dependent variables. The dependent variable, natural rubber productivity, was applied to estimate spatial panel time series. The study by Huo [7] discussed the factors on export competitiveness of the agriculture industry. Huo [7] mentioned that there are both positive and negative relationships with the export competitiveness of the agriculture industry. Irrigated land area (ln land) and exchange rate (ln x rate) were found to have a negative relationship with export competitiveness. Labor cost and domestic consumption demand were found to have a negative relationship with export competitiveness. Therefore, in this study, another independent variable used was farming families (ln fam). All the independent variables were considered referenced in a study by Huo [7] as variables affecting the export of agricultural products.

3.2 Conceptual Framework and Methodology Under the BCG idea, the process of spatial analysis for enhancing the TE (technical efficiency) of Thailand’s rubber exports is described in this research study’s conceptual framework shown in Fig. 7. The process under this conceptual framework comprises three steps for proceeding using a spatial approach technique. Both data manipulation and model creation must be started as the first stage. The second stage aims to estimate the three models: non-spatial, mixed spatial, and spatial, in that order. The final stage in using Moran’s I test to improve the TE of Thailand’s

10

C. Intapan and C. Chaiboonsri

Table 1 Data reviews and definitions Variables

Definition

Data source

Dependent variable ln_rubber

The panel samples of Thai natural Office of Agricultural Economics, rubber export to ASEAN countries Bangkok, Thailand (Malaysia, Singapore, Philippines, Indonesia, Laos, Cambodia, Myanmar, and Vietnam) from 2010 to 2020. The data was transformed into a natural log form (The natural log of rubber’s ton to export)

Independent variables Description

Data source

ln_land

The natural log of land (Rai)

ln_labor

The natural log of labor (people)

ln_price

The natural log of price to export (baht)

Office of Agricultural Economics, Bangkok, Thailand, and BOT (Bank of Thailand)

ln_xrate

The natural log of exchange rate (Thai baht/Dollar)

Source From official data

Fig. 7 A conceptual framework for the improvement of technical efficiency under the BCG concept

Measuring the Technical Efficiency of Thai Rubber Export Using …

11

rubber exports is to choose the optimal model within the BCG ideas, which stresses the highest TE comparison across the various models. Additionally, this conceptual framework and methodology attempt to display how to enhance the TE of Thailand’s rubber export by the exploring three models. The weight matrix is the main concept employed in the spatial model, which is the significant fundamental structure used in these models for analysis. Furthermore, the first model, the non-spatial model for Thailand’s rubber exports, is estimated using the Maximum Likelihood Estimator without using the weight matrix to help with the analysis. The weight simulation matrix is also included in the second model for Thailand’s exports, which is estimated by the same estimator. The traditional spatial model is the last model, which uses the weight matrix using the original spatial information that may be found in geographic data to analyze. This research computes all models using the same data but different structure weight matrices, then evaluate them using TE. The highest TE appearance indicates that the finest Thai export model is one that Thailand should prioritize and develop policies to support quickly. The final steps of this conceptual framework are to display the policy suggestions based on the BCG concept (Logistics with low carbon emissions and using renewable energy) along with the research results.

3.3 Spatial Stochastic Frontier Analysis Models The traditional SFA models have been extended with the purpose of taking into account firm-specific heterogeneity. If firm-specific heterogeneity is not accounted for, in fact, a considerable bias in the inefficiency estimates can be endogenously created. The extension of the SFA model is called the Spatial Stochastic Frontier Analysis (SSFA) model. The aim of the SSFA model is to test and depurate the spatial heterogeneity in SFA models by splitting the inefficiency term into three terms: 1. Inefficiency term related to spatial peculiarities of the territory in which every single unit operates 2. An Inefficiency term related to the specific product features, and 3. The inefficiency term is representing the error term. The dependence of spatial refers to how much the level of technical inefficiency of farm i depends on the levels set by other farms j = 1,…, n, under the assumption that part of the farm i inefficiency (u i ) is linked to the neighbor DMU j’s performances ( j = i). yi is the single output of producer i. xi is the vector of inputs. f is a generic parametric function. The production frontier model with normal/half-normal cross-sectional can be presented as the following equation. log(yi ) = log( f (xi ; βi )) + vi − u i  = log( f (xi ; βi )) + vi − 1 − ρ

 i

where

−1 ωi

u˜ i

(1)

12

C. Intapan and C. Chaiboonsri

vi ∼ N (0, σv2 )   ui ∼ N

+

0, 1 − ρ



ωi

i

−2

 σu˜2

u i and vi are independently distributed of each

other and of the regressor  u˜ i ∼ N 0, σu˜2 ωi is a standardized row of the spatial weight matrix. ρ is the spatial lag parameter (ρ ∈ [0, 1]). Thus, the SSFA model used in this research can be written as Eqs. (2) and (3) ln(yit ) = ln( f (xit ; βi )) + vit − u it

(2)

log(yit ) = a + β1log(land) + β2 log(labor ) + β3log( price) + β4 log(exchnge_rate) + vit − u it

(3)

where log(yit ) = the dependent variable is ln(export) for Thai rubber export. log(land), log(labor ), log( price), log(exchnge_rate) = the independent variables of this model. vit = ρvit + ψit = the autocorrelated pure error of spatial stochastic model of Thai rubber export. T E it = u it = 1+exp[−(w1 it γ +zit φ] + εit = Technical efficiency measurement of Thai rubber export. The TE, whose value ranges from 0 to 1, can be used to calculate the technical efficiency measurement of Thai rubber export. Conversely, TE approaches 0 which denotes low efficacy of rubber exports to those nations. If TE approaches 1, Thailand has the greatest export efficacy to those nations. wit is the spatial weight matrix for measurement of spatial distance. z it are some factors to determine the inefficiency and ρ is the spatial lags parameter. ψit the error terms of vit and u it are the measures of inefficiency.

4 Empirical Results 4.1 Descriptive Information for Observed Spatial Data Data visualization and descriptions of the variables are shown in Table 2 to display some descriptive statistics such as mean, median, maximum, and minimum. This statistic is used to describe the fact that there is a scale of verity that is difficult to explain using the only regression model. Therefore, this research needs to convert the variety scale to be the same scale by taking nature log in all variables before

Measuring the Technical Efficiency of Thai Rubber Export Using …

13

Table 2 Description of both dependent variable and independent variables used to estimate by maximum likelihood estimator RUB_EX LAND

LABOR

Mean

12,283.30 218,963.6

16,533.64 32.22364

Median

896.7200

221,100.0

16,089.00 31.70000

63.90000

Maximum

70,279.63 221,100.0

18,290.00 35.28000

148.2800

Minimum

0.040000

14,547.00 30.47000

50.73000

Std. Dev

19,469.20 4029.152

Skewness

1.572716

210,600.0

X-RATE

PRICE 78.03909

1355.552

1.536446

30.95756

−1.544129 0.174417

0.726258

1.076975

Kurtosis

4.190911

3.512563

1.566469

2.196731

2.917161

Jarque–Bera

41.47737

35.93353

7.981225

10.10182

17.03667

Probability

0.000000

0.000000

0.018488

0.006404

0.000200

Panel unit root test (Levin, Lin, and −3.98932 −303.975 Chu t*)

−7.32468 −2.30381 −3.71323

Prob

0.0000

0.0000

0.0000

0.0106

0.0001

Observations

88

88

88

88

88

From The author’s computation

estimation by the maximum likelihood estimator. Furthermore, the panel unit root test proposed by Levin et al. [1] confirms that all variables used to estimate the spatial stochastic frontier are stationary. It is implied that all variables have a mean and variance that are stable over time.

4.2 The Result of an Appropriate Model From Table 2, the mixed spatial model is the appropriate model for Thai natural rubber exports to ASEAN countries (Malaysia, Singapore, Philippines, Indonesia, Laos, Cambodia, Myanmar, and Vietnam) from 2010 to 2020. It is because the technical efficiency of this model is highest when compared with another model. This research study’s mixed spatial model is the main contribution idea for increasing collaboration between the spatial models and the non-spatial models working together. The Moran I statistic (detail in Table 2 and Appendix 1) from the mixed spatial model has confirmed that the spatial effect must provide more informative direction. The Moran’s I2 has simply been valued between −1 and 1, where −1 means perfectly dispersed, which means that they are independent in spatial among ASEAN countries’ importers of Thai’s natural rubber. The Moran’s I approach 1 indicates that Thailand’s natural rubber importers are perfectly clustered in terms of spatial analysis. Furthermore, if the Moran’s I approach 0, which is the randomly distributed or independent among ASEAN countries’ importers of Thai natural rubber. The Moran’s 2

https://gisgeography.com/spatial-autocorrelation-moran-I-gis/.

14

C. Intapan and C. Chaiboonsri

Table 3 Result of the average of technical efficiency based on the three types of stochastic frontier model Technical efficiency of all countries’ importers

Non-spatial model (model without weights matrix)

Mixed spatial model (model with simulation of a weights matrix)

Spatial model (model with weights matrix)

Average of technical efficiency

0.166784

0.178543

0.159989

−0.000455 (p-value = 0.01117)

0.001309 (p-value = 0.2285)

Moran’s I statistic (Ho: – Moran’s I under the null hypothesis of spatial randomization) From The author’s computation

I statistics from the mixed spatial model in Table 2 are approximately −0.000455, indicating that this model is not fitted for being perfectly dispersed or perfectly clustered in terms of spatial analysis. However, the mixed model presented shows that the values of the statistic (P-value = 0.01117) can be rejected as evidence that Thailand’s rubber exports are spatially independent. It means that the infrastructure development especially the logistics system based on BCG still requires high level of international development among ASEAN countries, which needs to be addressed as a priority. Among ASEAN countries, the importers of Thai natural rubber are to be concerned for speedy logistics system development in the CLMV (Cambodia, Laos, Myanmar, and Vietnam) Asian member and Indonesia. It is because those countries attained the less technical efficiency, compared to the main rubber importers of Thailand such as Singapore, Malaysia, and the Philippines (see Appendix 2). The mixed spatial stochastic frontier model has been performed and the results are given in Table 3. In terms of explanation, the logistics system needs to improve connectivity among Thai importers to get more export efficiency. The development of the logistics system is reflected in the mixed spatial model’s simulation of a weights matrix, which is evaluated by an average rise in export technical efficiency. When the weight matrix is close to 0.9, it signifies that the logistics system will improve by around 90%, and when it is close to 0.1, it means that the opposite will occur. According to this study, the weight matrix of 0.5 is the greatest option for increasing the technical efficacy of Thailand’s rubber exports to ASEAN nations. The mixed spatial model’s estimated results for quantifying Thai natural rubber exports to ASEAN nations between 2010 and 2020 are shown in Table 4. This table presented the results of all parameters control estimated by the mixed spatial stochastic frontier model to calculate the technical efficiency based on model selection by the Moran I statistic (Fig. 7). The empirical result of the estimation for this model suggests that land is the important factor to drive in Thailand’s rubber industry for export because it can increase efficiency to support this industry’s expansion. The coefficient of the log(land) is decreasing when Thailand has more exports to ASEAN countries by a statistical significance level of 0.01. The negative direction of log(land)

Measuring the Technical Efficiency of Thai Rubber Export Using … Table 4 Result of estimation for the mixed spatial model to quantify the Thai natural rubber exports to ASEAN countries from 2010 to 2020

15

Spatial stochastic frontier regression model

Mixed spatial model

(Intercept)

95.96*** (3.61)

log(LAND)

−9.72*** (1.53)

log(LABOR)

1.80 (2.39)

log(PRICE)

0.07 (0.62)

log(EX)

5.10 (3.52)

σ2

19.44*** (2.28)

γ

0.003 (0.007)

‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05

in this output implies that Thailand increases export productivity while decreasing rubber plantations and increasing export value3 as well. However, the other factor (Intercept, which is the latent variable that is not considered in this model for estimation) is still to be researched for the purpose of promoting and supporting the export industry’s efficiency. This latent variable has a very high significance level of 0.01 in the positive direction, which means that it has a highly encouraging potential factor to support this export industry. Additionally, the labor, export price, and exchange rate variables estimated by the mixed spatial model between 2010 and 2020 had little impact on the effectiveness of Thai rubber exports to ASEAN countries. At a statistical significance level of 0.001, σ2 is also highly significant, indicating that the technical efficiency can be used to quantify the effectiveness of Thai natural rubber exports to ASEAN nations between 2010 and 2020. The research results gathered initially support that the policymaker needs to promote the spatial development strategy of the Thai agriculture industry, especially rubber export to ASEAN under a logistics system focusing on the BCG concept.4 Thai natural rubber exports to ASEAN countries are expected to increase in the future, if the stakeholders grasp the BCG concept and work to develop a good farm management and logistics management system. The development of good logistics systems, such as railways, roads, and rivers, needs to be focused on by focusing on logistics with low carbon emissions and using renewable energy (the BCG concept for logistics system development). In terms of an appropriate model, the mixed spatial model is presented where the average technical efficiency is equal to 0.178, which is the highest when compared with that of other models. 3 4

https://www.reuters.com/article/thailand-rubber-idUKL4N28E2EW. https://www.bcg.in.th/eng/strategies/.

16

C. Intapan and C. Chaiboonsri

5 Conclusion and Policy Recommendations This research study would still like to achieve its main objective of improving the rubber exports of Thailand based on the BCG concept. The spatial dataset was applied with a stochastic frontier analysis model, which is called the panel spatial stochastic frontier analysis model estimation applied in the analysis of Thailand’s rubber export. The empirical results of this study suggested that improving the technical efficiency of Thailand’s rubber exports by improving the infrastructure, especially the logistic system development of CLMV countries, needs to be addressed first. The mixed spatial matrix (mixed Wij ) of this study represents the level of logistics system development, which plays an important role in the sustained technical efficiency of Thailand’s rubber export. Many factors point to the recent global market trend for rubber as a promising opportunity for Thailand’s rubber export. For example, the pandemic of Covid-19 has resulted in an increasing demand for rubber products worldwide. According to information from Business Wire, the global outbreak of the Covid-19 pandemic in 2020 has increased the demand for personal protective equipment,5 such as gloves, masks, face shields, and gowns. For a variety of reasons, the demand on the global market for rubber is increasing; as a result, Thailand’s rubber export needs to improve its efficiency, particularly in terms of logistic system development, to help this industry contribute to farmers’, entrepreneurs’, and business agriculture sectors’ income on a sustainable basis in the future. Acknowledgements This research was supported by CMU Junior Research Fellowship Program. And also, thanks to Thai Rubber Export for providing the data for the analysis of this research work.

Appendix 1 See Figs. 8 and 9.

5

https://www.pharmiweb.com/press-release/2022-11-11/global-covid-19-personal-protective-equ ipment-ppe-market-supply-demand-and-future-forecasts-2022.

Measuring the Technical Efficiency of Thai Rubber Export Using … Fig. 8 Graphical representation of Moran scatterplot for a spatial mixed model (SFA_2). Source author’s computation

Fig. 9 Graphical representation of Moran scatterplot for a spatial model (SFA_3). Source author’s computation

Appendix 2 See Figs. 10 and 11.

17

18

C. Intapan and C. Chaiboonsri

Highest TE(Technical efficiency) 1.2000 1.0000 0.8000 0.6000 0.4000 0.2000 0.0000 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Malaysia

Singapore

Philippines

Fig. 10 Technical efficiency estimated by spatial mixed model of Thailand’s rubber export to Malaysia, Singapore, and Philippines from 2010 to 2020. Source author’s computation

Potenal TE(Technical efficiency) 0.03500 0.03000 0.02500 0.02000 0.01500 0.01000 0.00500 0.00000 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Indonesia

Laos

Myanmar

Vietnam

Cambodia

Fig. 11 Technical efficiency estimated by spatial mixed model of Thailand’s rubber export to Indonesia, Laos, Cambodia, Myanmar, and Vietnam since 2010 to 2020. Source author’s computation

Measuring the Technical Efficiency of Thai Rubber Export Using …

19

Appendix 3 (A programming algorithm applied in this research article) ########### Simulation study for improving the technical efficiency by the spatial matrix simulation approach ####### install.packages("plot.matrix") install.packages("openxlsx") install.packages("readxl") install.packages("ssfa") library('plot.matrix') library("openxlsx") library("readxl") library("ssfa") ############# 1: Spatial matrix = 0.5 ############################################################### ####### w_sim2