FLEXIBILITY AND RELATED ISSUES IN EVALUATION AND SELECTION OF
MANUFACTURING SYSTEMS
by
B.V. CHOWDARY
Department of Mechanical Engineering
submitted
in fulfilment of the requirements of the Degree of
DOCTOR OF PHILOSOPHY
to the
INDIAN INSTITUTE OF TECHNOLOGY, DELHI HAUZ KHAS, NEW DELHI - 110016, INDIA
September, 1997
TV 7. o)
F
1. I. T. TT:T.
r-F.f
CERTIFICATE
The thesis entitled Flexibility and Related Issues in Evaluation and Selection of Manufacturing Systems being submitted by Mr. B. V. Chowdary to the Indian Institute of Technology, Delhi for the award of the degree of Doctor of Philosophy, is a record of bonafide research carried out by him. He has worked under our guidance and supervision and has fulfilled the requirements for the submission of this thesis, which has attained the standard required for a Ph. D degree of this institute. This work has not been submitted elsewhere for the award of any other degree or diploma.
Dr. Arun Kanda Professor
Department of Mechanical Engineering Indian Institute of Technology, Delhi Hauz Khas, New Delhi -110 016 India
Dr. K. Surya Prakasa Rao Professor
Department of Mechanical Engineering K.L. College of Engineering
Kunchanapalli - 522 502
Nagarjuna University
Andhra Pradesh, India
ACKNOWLEDGEMENTS
It gives me profound pleasure to thank Prof. Arun Kanda and Prof. K. S. P. Rao for their valuable help and guidance throughout the research study. I am extremely indebted to Prof. Arun Kanda and Prof. K. S. P. Rao for the constant encouragement, help, sincere and timely advice and for keeping the spirit high throughout the study.
Thanks are due to my Professors, Dr. U. R K. Rao, Dr. N. K. Tewari, Dr. S. G. Deshmukh, Dr. Prem Vrat, Dr. Sushi!, and Dr. S. Wadhwa who gave me constant encouragement and guidance throughout my research work at Indian Institute of Technology, New Delhi.
I am grateful to Mr. S. Narayana Rao in arranging the necessary data at different case study units while the research was in progress. I extend my thanks to Mr. G. Narasimha Murthy for providing the valuable data and suggestions in carrying out the present research.
I am overwhelmingly indebted to my colleagues Mr. M. Srinivasa Rao, Mr. M. Sridhar, Mr. K. Raja Sekhar, Mr. N. V. Rao and Mr. A. Jagadesh for their help, sincere advice and encouragement all through the strenuous endeavour.
A special word of thanks to Mr. M. Srinivasa Rao who helped me in planning and debugging the computer software programmes. It also gives me an opportunity to thank Mr. G. Bhanu Prasad for sparing his valuable time in manuscript correction of this thesis.
My special thanks are due to my friends Dr. S. Srikrishna, Pandiri Srinivasa Rao, Sarat Chandra, Narasimha Rao, K. Ram Bhoopal Reddy, Ch. Srinivasa Rao, and M. V. Satya Naryana Raju for their care, attention, help and encouragement in bringing out the thesis into this volume.
Finally, a special word of thanks to my wife Mrs. B. Padma, for bearing with me during my arduous task. My thanks to all of my friends and others who helped me directly or indirectly in bringing out this thesis into the present form.
jj‘i&I-Obsu'r (B. V. Chowdary)
ABSTRACT
The main objective of this thesis is to develop models and heuristics for evaluation and selection of manufacturing systems keeping manufacturing flexibility and related issues in view. The models and heuristics proposed in the thesis handle not only conventional but also advanced technologies such as flexible manufacturing systems (FMS). The entity- relationship (E-R) framework has been used to model a variety of manufacturing flexibilities. An E-R modelling approach has been used for deriving various other flexibilities in the context of individual as well as a combination of flexibilities. Manufacturing system evaluation models like integrated manufacturing performance (IMP), multi-criteria evaluation for ranking of manufacturing system configurations, and multi-objective modelling approach through application of goal programming (GP) and interval goal programming (IGP) are proposed. Knowledge based expert systems and back-propagation artificial neural network models are proposed for selection of a flexible manufacturing system or a machining centre. This thesis consists of eight chapters.
Initially, the study introduces the subject, presents an organization of the problems, and brings out the motivation of the study. A consolidated review of the literature is presented with schematic representations in tabular forms at appropriate places.
A tool such as an E-R model has been proposed to model various manufacturing flexibilities. An algorithm for generating alternative combinations of entities for unification of various flexibilities has been presented. Unification of machine and routing flexibilities has been illustrated for a company engaged in discrete parts manufacture.
For evaluation and selection of manufacturing systems, a revised integrated manufacturing performance (RIMP) measure has been proposed. This RIMP measure is tested for its utility in the context of different Indian manufacturing environments.
A conceptual model for evaluation and ranking of manufacturing systems under multiple criteria decision making (MCDM) environment has been developed and tested through various case studies.
Application of a GP model to a real manufacturing situation under multiple conflicting goals is illustrated.
The results of the study are of great significance to the operations manager for long run production planning. The IGP approach has been applied for the manufacturing scenario to incorporate the preferences of a systems designer over prespecified ranges.
ii
Development of a knowledge based expert system for selection of a machining centre (KBFMS) is presented. The developed KBFMS is demonstrated through different performance criteria such as productivity, quality, and flexibility. A decision support system for evaluation and selection of manufacturing systems for flexibility and related issues is proposed to integrate much of the work done in the thesis. A user friendly code in 'C' has been developed for this purpose, and sample sessions are included.
The potential of back-propagation artificial neural network for selection of a machining centre is explored through a three layered neural network model which is trained and tested on an illustrative example. This is an alternative approach to select a machining centre apart from the MCDM and KBFMS models.
Finally a summary of research and the contributions of the research study are highlighted. The limitations and the possible extensions for future research are also suggested.
A list of references of the literature related to this study is included. A substantial part of the research reported in this thesis is presented in various journals. A list of papers presented based on the work reported is given at the end.
iii
CONTENTS
Page No.
ACKNOWLEDGMENTS
ABSTRACT ii
CONTENTS iv
LIST OF FIGURES xi
LIST OF TABLES xiv
NOMENCLATURE xix
CHAPTER I INTRODUCTION
1.1 Manufacturing Systems and Society 1
1.2 Developments/Trends in Manufacturing 7
1.3 Overview of Machine Tool Industry 10
1.4 Motivation Behind the Study 16
1.5 Objectives of the Study 18
1.6 Organization of the Thesis 19
CHAPTER II LITERATURE REVIEW
2.1 Introduction 25
2.2 Basis of the Present Review 25
2.3 Schematic Representation of the Literature Review 29 2.4 Design of Manufacturing Systems for Flexibility 29
2.4.1 Manufacturing Flexibilities 32
2.4.2 Existing Frameworks for Manufacturing Flexibilities 41 2.5 Performance Evaluation of Manufacturing Systems 43
2.5.1 Individual Flexibilities 46
2.5.2 Multiple Flexibilities and other Performance Measures 48 2.5.3 Existing frameworks and other performance measures 51
2.6 Selection of Manufacturing Systems 51
iv
2.6.1 Selection Under Multi-Criteria Decision Making Environment
2.6.1.1 Multi-Attribute Decision Making 2.6.1.2 Multi-Objective Decision Making
2.6.2 Selection of Advanced Machining Centre through Artificial Intelligence Techniques
52 53 62
66 2.6.2.1 Knowledge Based Expert Systems 67
2.6.2.2 Decision Support Systems 69
2.6.2.3 Back Propagation Artificial Neural Networks 71
2.7 General and other Issues 76
2.8 Current Research Trends 77
2.9 Limitations of Existing Approaches 78
2.10 Need for Further Research and Focus of the Study 79
2.11 Conclusions 81
CHAPTER III RELATIONSHIPS AMONG MANUFACTURING
3.1 3.2
FLEXIBILITIES USING ENTITY-RELATIONSHIP MODELS Introduction
Classification and Modelling of Manufacturing Flexibilities
82 83 3.2.1 Relationship among Flexibilities 89 3.2.2 E-R Models of Individual Flexibilities 86
3.3 Derivation of Valid Flexibility Paths 92
3.4 An Algorithm for deriving Valid Flexibility Paths 98 3.5 Interpretation of Derived Flexibility Paths 100
3.6 Discussion of Results 101
3.7 Possible Extensions 106
3.8 Conclusions 106
CHAPTER IV UNIFIED FRAMEWORK FOR DESIGN OF MANUFACTURING SYSTEMS
4.1 Introduction 108
4.2 Motivation Behind the Proposed Unified Framework 109 4.3 Proposed Unified Framework for Design, Evaluation,
and Selection of Manufacturing Systems 109 4.3.1 Uncertainties versus Flexibilities 112 4.4 Development of the Revised Integrated Manufacturing
Performance Measure 115
4.4.1 Total Productivity Measure 115
4.4.2 Total Quality Measure 115
4.4.3 Total flexibility Measure 117
4.4.4 Revised Integrated Manufacturing Performance
Measure 114
4.5 Evaluation and Validation Of Proposed Unified Framework and
Derived RIMP Measure 120
4.5.1 Case Study 1 120
4.5.2 Case Study 2 122
4.5.3 Case Study 3 124
4.5.4 Investigations Under Varying Weightages 128 4.6 Extensions to the Proposed Unified Framework 130 4.7 Limitations of the Proposed Unified Framework 134
4.8 Conclusions 134
CHAPTER V MULTI-CRITERIA EVALUATION FOR SELECTION OF MANUFACTURING SYSTEMS
5.1 Introduction 136
5.2 A Conceptual Model for Manufacturing System Evaluation
and Selection 137
5.2.1 Evaluation of Conceptual Model Through Case Studies 138
vi
5.2.2 Performance Measures and the Modelling Approach for
Manufacturing Systems 141
5.2.3 Multi-Criteria Ranking of Different Manufacturing System
Configurations 159
5.2.4 Interpretation of Results 159
5.3 Flexibilities in the Multi-Criteria Framework 163 5.3.1 A Review of Existing Flexibility Measures 163 5.3.1.1 An Application to Case Study 4 163 5.3.1.2 Merits and Demerits of the Suggested
Flexibility Measures 165
5.3.2 An Alternative Method of Selecting a Flexible
Manufacturing System 169
5.3.2.1 Case Study 4 Revisited 170
5.3.2.2 Formation of Multi-Criteria Decision Table for
Machine Flexibility 171
5.3.2.3 Formation of Multi-Criteria Decision Table for
Routing Flexibility 175
5.3.2.4 Combined Flexibility Measure 175
5.3.3 Comparison of Results 176
5.4 Production Planning in a Precision Machine Tool Manufacturing
Industry 178
5.4.1 Manufacturing System Performance Modelling Under
Multi-Objective Approach 179
5.4.1.1 Formulation of Product Flexibility Coefficients 180 5.4.1.2 Multi-objective Model Building 183 5.4.2 Formulation of the Problem in Goal Programming Format 186 5.4.2.1 Description of the Problem 186
5.4.2.2 Model Assumptions 186
vii
5.4.2.3 Model Formulation 187 5.4.2.4 Sensitivity to Changes in the Goal Priority
Structures 190
5.4.2.5 Model Solution and Results 190 5.4.2.6 Sensitivity to Changes in Model parameters 196 5.4.2.7 Validation of Dynamic Modelling Approach 196 5.4.3 Formulation of the Problem in Interval Goal
Programming Approach 201
5.4.3.1 Sensitivity to Changes in the Goal Priority
Structures 202
5.4.3.2 Comparison of Results Obtained from Goal and
Interval Goal Programming 206
5.5 Conclusions 207
5.6 Limitations of the Work 208
Chapter VI DEVELOPMENT OF A DECISION SUPPORT SYSTEM FOR FLEXIBILITY IN MANUFACTURING
6.1 Introduction 210
6.2 Architecture of the Proposed Expert System for Selection of
Machining Centre 211
6.2.1 Design of Data Base 212
6.2.2 Design of Knowledge Base 212
6.2.3 Inference Engine 216
6.2.4 User interface 218
6.3 Knowledge Based Expert System for Selection of
Machining Centre 220
6.4 Features of KBFMS 224
6.5 Sample Session with KBFMS 225
6.6 Decision Support System for Flexibility in Manufacturing 252 6.6.1 Architecture of the Proposed System 253
6.6.2 Development of the System 259
viii
6.6.3 Salient Features of the System 266 6.6.4 A Sample Session with FMDSS 266 6.6.5 Concluding Remarks and Validation 275
6.7 Conclusions 276
CHAPTER VII AN ALTERNATIVE APPROACH FOR FACILITY SELECTION THROUGH BACK-PROPAGATION ARTIFICIAL NEURAL NETWORKS
7.1 Introduction 277
7.2 Artificial Neural Networks 278
7.3 Description of Machining Centre Selection Problem 284
7.3.1 The BPANN Architecture 284
7.3.2 Preparing the Training and Testing Patterns 287 7.3.3 Supervised Learning and Back-Propagation Training
Method 288
7.3.4 Testing of BPANN Model Output 293
7.4 Results and Discussions 298
7.5 Comparison of Results of BPANN with those of the KBFMS 299
7.6 Validation of BPANN Model 303
7.7 Conclusions 303
CHAPTER VIII SUMMARY OF FINDINGS AND CONCLUSIONS
8.1 Introduction 306
8.2 Summary of the Research and Findings 306
8.3 Significant Research Contributions of the Study 311
8.4 Limitations of the Study 313
8.5 Suggestions for Future Extensions 314
8.6 Concluding Remarks 314
REFERENCES
318
ix
APPENDIX A Process Description of the Computer Based Algorithm
For Combining Flexibilities 334
APPENDIX B Questionnaire for Data Collection 336 APPENDIX C Sample Data Collection
C.1 Case Study 1 341
C.2 Case Study 2 341
C.3 Case Study 3 345
CA Case Study 4 347
APPENDIX D Supporting Calculations for Chapter IV
D. I Rationale in Collecting/Preparing Input Cost Elements 349 D.2 Supporting calculations for Table 4.2 (Shop-floor Performance
Measures of the System) 350
D.3 Supporting Calculations for Table 4.3 (Shop-floor Performance
Measures of the System for the Past and Planned Year) 352 D.4 Supporting Calculations for Table 4.5 (Shop-floor Performance
Measures of the System before and after Changes in the
Model/Product Design Features 356
APPENDIX E Sample Model Formulations of Chapter V
E.1 GP Formulation for Priority Structure I 359 E.2 GP Formulation for Priority Structure II 360 E.3 GP Formulation for Priority Structure III 361 E.4 IGP Formulation for Priority Structure I 362
E.5 Routing Model Formulation 364
APPENDIX F Machining Centre Codes followed while Preparing
the Data Base for KBFMS 367
APPENDIX G Outline of Computer Programs Developed 369
LIST OF PUBLICATIONS FROM THIS WORK 374
CURRICULUM VITAE
