Intelligent Knowledge-Based Systems (eBook)

Title Page 
2 
Copyright Page 
3 
Table of contents 
4 
FOREWORD 6
PREFACE 8
CONTRIBUTORS 11
Title Page 
416 
Copyright Page 
417 
Table of contents 
418 
FOREWORD 420
PREFACE 422
CONTRIBUTORS 425
Title Page 
804 
Copyright Page 
805 
Table of contents 806
FOREWORD 808
PREFACE 810
CONTRIBUTORS 813
Title Page 
1153 
Copyright Page 
1154 
Table of contents 
1155 
FOREWORD 1157
PREFACE 1159
CONTRIBUTORS 1162
Title Page 
1624 
Copyright Page 
1625 
Table of contents 
1626 
FOREWORD 1628
PREFACE 1630
CONTRIBUTORS 1633
VOLUME I. KNOWLEDGE-BASED SYSTEMS 31
PLATFORM-BASED PRODUCT DESIGN AND DEVELOPMENT: KNOWLEDGE SUPPORT STRATEGY AND IMPLEMENTATION 
32 
1. INTRODUCTION 32
2. LITERATURE REVIEW 33
3. PLATFORM-BASED PRODUCT DESIGN AND DEVELOPMENT 36
4. PRODUCT PLATFORM AND PRODUCT FAMILY MODELING 38
4.1. Product family architecture modeling 38
4.2. Product family evolution representation 39
4.3. Product family generation 41
4.4. Product family evaluation for customization 42
5. MODULE-BASED PRODUCT FAMILYDESIGN PROCESS 44
6. KNOWLEDGE SUPPORT FRAMEWORK FOR MODULAR PRODUCT FAMILY DESIGN 
48 
6.1. Knowledge support scheme, challenges and key issues 48
6.2. Product family design knowledge modeling and support 51
6.2.1. Product family design knowledge modeling issues 
51 
6.2.2. Knowledge modeling /representation for product family design 54
7. KNOWLEDGE INTENSIVE SUPPORT SYSTEM FOR PRODUCT FAMILY DESIGN 
57 
8. SUMMARY AND FUTURE WORK 61
REFERENCES 61
KNOWLEDGE MANAGEMENT SYSTEMS IN CONTINUOUS PRODUCT INNOVATION 
65 
1. INTRODUCTION 65
2. KNOWLEDGE AND KNOWLEDGE MANAGEMENT 67
2.1. The concept of knowledge in management literature 67
2.2. Defining a knowledge management system 69
3. LITERATURE REVIEW 71
3.1. Main streams in literature 73
3.2. The literature evolutive trend: towards KM configurations 78
4. THE INVESTIGATION FRAMEWORK 80
5. THE RESEARCH METHODOLOGY 83
6. RESULTS 86
7. IMPLICATIONS FOR MANAGERIAL ACTION AND FUTURE RESEARCH 91
REFERENCES 92
KNOWLEDGE-BASED MEASUREMENT OF ENTERPRISE AGILITY 96
1. INTRODUCTION 96
2. MANAGING AN ADAPTIVE INFRASTRUCTURE 98
3. AGILITY MODELING AND MEASUREMENT FUNDAMENTALS 99
3.1. Dimensions of agility 100
4. MODELING OF AGILITY INFRASTRUCTURES 103
4.1. Production infrastructure 103
4.2. Market infrastructure 104
4.3. People infrastructure 104
4.4. Information infrastructure 105
4.5. Discussion 105
5. AN EXAMPLE 107
6. CONCLUDING REMARKS 110
REFERENCES 111
KNOWLEDGE-BASED SYSTEMS TECHNOLOGY IN THE MAKE OR BUY DECISION IN MANUFACTURING STRATEGY 
112 
1. INTRODUCTION 112
THE MAKE OR BUY DECISION 113
(i) No Formal Methodfor Evaluating the Decision 113
(ii) Inaccurate Costing Systems 114
(iii) The Competitive Implications of the Decision 114
A DESCRIPTION OF THE MAKE OR BUY MODEL 114
Stage 1-Identification of Performance Categories 
116 
Stage2-An Analysis of the Technical Capability Categories 118
Stage 3-Comparison of Retrieved Internal and External Technical Capability Profiles 
118 
Stage 4-An Analysis ofthe Suppliers' Organisations 118
Stage 5-Total Acquisition Cost Analysis 119
THE MAKE OR BUY SYSTEM 119
KNOWLEDGE BASED SYSTMS (KBS) AND CASE-BASED REASONING (CBR) 119
THE REQUIREMENTS 121
SYSTEM DEVELOPMENT 122
Stage 1-Peiformance Criteria Identification and Weighting 
122 
1. Technical Capability Categories 122
2. Suppliers Organisation Categories 
123 
Stage 2-Technical Capability Stage 123
Stage 3-Comparison of Retrieved Internal and External Technical Capability Profiles 
128 
Stage 4-An Analysis of the Suppliers' Organisations 
130 
EVALUATION 133
FURTHER ENHANCEMENTS 135
Dynamic performance analysis 135
A consultancy tool 135
Application of AI techniques 135
CONCLUSION 136
REFERENCES 137
INTELLIGENT INTERNET INFORMATION SYSTEMS IN KNOWLEDGE ACQUISITION: TECHNIQUES AND APPLICATIONS 
139 
1. INTRODUCTION 139
2. RELATED WORK 140
2.1. The Web 140
2.2. Information retrieval 142
2.3. Hyperlink analysis 144
2.4. Information extraction 145
2.5. Data mining and machine learning 145
2.6. Document categorization 148
2.7. Web mining 149
2.8. Intelligent web agent 151
3. THE I3 SYSTEM 151
3.1. The architecture of the I3 system 
151 
3.2. Semantic issues of the I3 system 
152 
4. I3 WEB ANALYZER 
154 
4.1. Web crawler and document parser 155
4.2. Linguistic detector 156
4.3. Structural analyzer 156
4.4. Content analyzer 157
4.5. Summary of I3WA 158
5. I3 METADATA EXTRACTOR 158
6. 13 KNOWLEDGE LEARNER 161
6.1. The ACIRD system 161
6.2. Mining term associations 161
7. I3 APPLICATIONS 
163 
REFERENCES 164
AGGREGATOR: A KNOWLEDGE BASED COMPARISON CHART BUILDER FOR eSHOPPING 
169 
1. INTRODUCTION 169
2. RELATED WORK 172
3. WRAPPERS AS CONCEPTUAL GRAPHS 174
3.1. Conceptual graphs background 174
3.2. Modeling and training wrappers with CGs 176
3.3. Reusing CG-wrappers 180
4. COMPARISON CHART BUILDING WITH CG-WRAPPERS 181
4.1. Locating product specification pages 181
4.2. Collecting and merging specification information 182
5. A FRAMEWORK FOR INFORMATION EXTRACTION WITH CG-WRAPPERS 185
5.1. System architecture 185
5.2. Case study 188
6. CONCLUSIONS AND FUTURE WORK 190
REFERENCES 191
IMPACT OF THE INTELLIGENT AGENT PARADIGM ON KNOWLEDGE MANAGEMENT 
193 
1. INTRODUCTION 193
2. PARADIGM SHIFT: FROM DATA AND INFORMATION MANAGEMENT TO KNOWLEDGE MANAGEMENT 
196 
3. KNOWLEDGE MANAGEMENT: DEFINITIONS AND ARCHITECTURE 198
4. KNOWLEDGE MANAGEMENT SUPPORT 202
5. INTELLIGENT AGENTS AND MULTIAGENT SYSTEMS 204
6. KNOWLEDGE TYPOLOGIES 209
6.1. Notion of Knowledge and Knowledge Possessors 210
6.2. Types of knowledge 213
6.3. Sources of knowledge 215
7. ORGANIZATIONS AS COMMUNITIES OF AGENTS AND PASSIVE OBJECTS 218
8. ORGANIZATIONS AS INTELLIGENT AGENTS 220
9. ORGANIZATIONS AS MULTIAGENT AND KNOWLEDGE MANAGEMENT SYSTEMS 
223 
9.1. Intelligent agents for OKMS "Engine Room" 227
9.2. Agents that provide communications 229
9.3. Personal agents 229
10. CONCLUSIONS 231
ACKNOWLEDGMENTS 232
REFERENCES 232
METHODS OF BUILDING KNOWLEDGE-BASED SYSTEMS APPLIED IN SOFTWARE PROJECT MANAGEMENT 
236 
INTRODUCTION 236
1. PROBLEMS OF MODELLING SPM 237
1.1. Expert knowledge of project management 237
1.2. Methods of supporting management processes 
238 
1.3. Description of project teams 241
1.4. Models for assessing team and project processes 243
2. NEW POSSIBILITIES FOR CREATING THE SPM MODEL 248
2.1. Use of modelling and simulation theories 250
2.2. Application of fuzzy set theory 251
2.3. Application of elements of fuzzy regulator theory 254
3. EXAMPLE OF BUILDING A FUZZY SPM MODEL 256
3.1. The concept of model construction 257
3.2. Construction of the model 264
3.2.1. Fuzzy Models of Knowledge- Based System for Software Project Management 
264 
3.2.2. Hierarchical model 264
3.2.3. Structural model 266
3.2.4. Integrated model 
266 
3.2.5. Tuning of the Fuzzy Model 269
3.2.6. Adaptation of the model to the needs of newprojects 269
3.2.6.1. THE SPM-RFM MODEL AS A SUPPORT FOR SOFTWARE PROJECT MANAGEMENT. 269
4. ASSESSMENT OF EXISTING SOLUTIONS 271
REFERENCES 271
SECURITY TECHNOLOGIES TO GUARANTEE SAFE BUSINESS PROCESSES IN SMART ORGANIZATIONS 
275 
1. INTRODUCTION 275
2. SMART ORGANIZATIONS-ARE THEY THE FUTURE? 277
2.1. Main characteristics of Smart Organizations 277
2.1.1. Definition of Smart Organization 277
2.1.2. Life cycle of networked organizations 278
2.1.3. Human role in smart organization 280
2.2. Organizational form 280
2.2.1. Main characteristics of VE 280
2.2.2. Importance of safe communication in VE 282
2.3. Knowledge technologies and applications 283
2.3. 1. Knowledge technologies 283
2.3.1.1. ARTIFICIAL NEURAL NETWORKS. 
284 
2.3.1.2. ANT ALGORITHMS AND SWARM INTELLIGENCE. 284
2.3.1.3. INTELLIGENT AGENTS. 285
2.3.1.4. KNOWLEDGE SHARING. 286
2.3.2. Knowledge management 286
2.4. Network technologies for smart organizations 287
2.4.1. Trends in information technology 287
2.4.2. Wired network technology 288
2.4.3. Wi-Fi (Wireless Fidelity) technology 289
2.4.4. Mobile technology 290
2.4.5. Powerline communications 292
2.4.6. The Grid computing 
293 
3. 
294 
3.1. The content of business processes 294
3.2. Relation between BPR & information and communication technology
4. SECURITY TECHNOLOGIES 296
4.1. Types and trends of cyber crimes 296
4.2. Computer system and network security 298
4.3. Role of trust 300
4.4. Security services and mechanisms 301
4.5. Tools, methods and techniques for security 302
4.5.1. Achieving confidentiality 302
4.5.2. Security architectures 302
4.5.3. Firewalls 302
4.5.4. Virus defense 303
4.5.5. Identification of persons 
303 
4.5.6. Smart cards 304
4.5.7. Personal trusted device 304
4.6. Application of security technologies in networks 305
4.6.1. Wired network security 305
4.6.2. Security technoloyies for wireless communication 306
4.6.3. Mobile security 308
4.6.4. Security issues in PLC 310
4.6.5. Security in the Grid 311
5. SECURITY APPLICATIONS IN SMART ORGANIZATIONS 311
5.1. Security in distributed environments 311
5.2. Human aspects of security in smart organizations 312
5.3. Application of security in the life-cycle phases 313
6. CONCLUSIONS 314
REFERENCES 315
BUSINESS PROCESS MODELLING AND ITS APPLICATIONS IN THE BUSINESS ENVIRONMENT 
317 
1. INTRODUCTION 317
2. BUSINESS PROCESS MODELLING 319
2.1. The model of an artificial system 319
2.2. Business processes and business process modelling 320
2.2.1. Business process 320
2.2.2. Business process model 320
2.2.2.1. WHAT IS A MODEL? 320
2.2.2.2. BUSINESS PROCESS MODELS AS SPECIFIC TYPES OF ENTERPRISE MODELS. 322
2.2.3. Categories of business process models and business process types 323
2.2.3.1. CATEGORIES OF BUSINESS PROCESS MODELS. 323
2.2.3.2. BUSINESS PROCESS TYPES. 324
2.3. Generalised enterprise reference architecture and methodology (GERAM) 325
2.3.1. GERAM framework 
325 
2.3.2. Generalised enterprise reference architecture (GERA) 327
2.3.3. Business process modelling languages and tools 328
2.3.4. Enterprise reference models 331
2.4. Business process modelling principles 331
2.4.1. Process decomposition 331
2.4.2. The granularity (depth) of process models 
332 
2.4.3. Modelling approach 
332 
2.5. CIMOSA process modelling language 333
2.6. Workflow management 334
2.6.1. Abstraction of process management 
334 
2.6.2. Architecture 335
2.6.3. Design principles and issues 336
2.6.4. Workflow from a data perspective 
336 
3. WHAT ISO 9000:2000 STANDARD REQUIREMENTS MUST BUSINESS PROCESS MODELS SATISFY? 
338 
3.1. Business process modelling related requirements of the ISO 9000:2000 standards 
339 
3.2. Business process interactions 343
3.3. Product realisation and support processes 346
3.4. From business process modelling to enterprise modelling 348
3.4.1. Organisational view related standard requirements 
348 
3.4.2. Resource view related standard requirements 350
3.4.3. Information view related standard requirements 352
3.5. The ISO 9000:2000 and business process reference models 354
4. BUSINESS PROCESS MODELLING IN BUSINESS PROCESS REENGINEERING 355
4.1. Ten-step approach to BPR 356
4.2. How to develop an AS-IS process model 357
4.3. Use documented best practice as an input to the BPR process 359
5. BUSINESS PROCESS MODELLING AND KNOWLEDGE MANAGEMENT 359
5.1. What is knowledge? 360
5.2. Need for knowledge management 361
5.3. The nature of knowledge and its sharing 362
5.4 . The knowledge process and knowledge resources 364
5.5. Business process modelling and knowledge management 366
5.6. The knowledge life-cycle model 367
6. CONCLUSION 371
REFERENCES 372
KNOWLEDGE BASED SYSTEMS TECHNOLOGY AND APPLICATIONS IN IMAGE RETRIEVAL 
375 
1. INTRODUCTION 375
2. KNOWLEDGE REPRESENTATION AND DESCRIPTION LOGICS 376
3. RELATED WORK 378
3.1. Feature-based approaches 378
3.2. Approaches based on spatial constraints 379
3.3. Logic-based approaches 380
4. PROPOSED KNOWLEDGE BASED APPROACH 381
4.1. Syntax 381
4.2. Semantics 382
5. REASONING AND RETRIEVAL 390
5.1. Exact reasoning on images and descriptions 391
5.2. Approximate recognition 397
5.2.1. Spatial similarity 
399 
5.2.2. Rotation similarity 401
5.2.3. Scale similarity 402
6. REPRESENTING SHAPES, OBJECTS AND IMAGES 404
6.1. Image features 404
6.2. Similarity computation 405
7. PROTOTYPE SYSTEM 407
7.1. Knowledge base management 408
8. DISCUSSION 411
REFERENCES 411
VOLUME II. INFORMATION TECHNOLOGY 445
TECHNIQUES IN INTEGRATED DEVELOPMENT AND IMPLEMENTATION OF ENTERPRISE INFORMATION SYSTEMS 
446 
1. INTRODUCTION TO THE INTEGRATED METHODOLOGY FOR ENTERPRISE INFORMATION SYSTEMS 446
1.1. Development of information systems 447
1.2. Previous research 447
1.3. Overview of the integrated methodology for enterprise information systems 
448 
2. TECHNIQUES OF INFORMATION STRATEGIC PLANNING 452
2.1. Overview 452
2.2. Previous researches 452
2.3. Information Strategic Planning Methodology (ISPM) 453
2.4. Framework for evaluation of ISP 454
3. TECHNIQUES FOR THE EVALUATION OF INDUSTRIAL INFORMATION SYSTEMS (EIII) 
455 
3.1. Overview 455
3.2. Previous researches 455
3.3. The improvement model of IS performance 
458 
4. TECHNIQUES OF IS ECONOMIC JUSTIFICATION AND MEASUREMENT 460
4.1. Overview 460
4.2. Previous researches 461
4.3. Framework for economic justification and measurement system (EJMS) 461
5. OTHER TECHNIQUES 464
5.1. Techniques of requirements analysis 464
5.2. UMT (Unified Modeling Tools) and repository 466
5.3. S3IE (Support Systems for Solution Introduction and Evaluation) 468
6. FURTHER WORKS 468
REFERENCES 468
INFORMATION SYSTEMS FRAMEWORKS AND THEIR APPLICATIONS IN MANUFACTURING AND SUPPLY CHAIN SYSTEMS 
470 
1. INTRODUCTION 470
2. INFORMATION SYSTEMS USE IN THE MANUFACTURING INDUSTRY 471
3. INFORMATION SYSTEMS EVOLUTION IN MANUFACTURING 472
3.1. Infrastructure as an element of information systems in manufacturing 474
4. ELECTRONIC COMMERCE AND MANUFACTURING INFORMATION SYSTEMS 
476 
5. VIRTUAL ORGANISATIONS AND MANUFACTURING INFORMATION SYSTEMS 
477 
6. PARADIGMS SHIFTS IN MANUFACTURING 478
6.1. IT and information systems for mass customisation 479
7. DEVELOPMENT OF INFORMATION SYSTEMS IN MANUFACTURING 480
8. EXAMPLES AND HIGHLIGHTS OF INFORMATION SYSTEMS DEVELOPMENTS IN MANUFACTURING 
483 
9. A BROADER SCOPE OF INFORMATION SYSTEMS IN MANUFACTURING 486
9.1. Information systems role in improving manufacturing organisations performance 
486 
10. INFORMATION SYSTEMS ENTERPRISE-WIDE SUPPORT: AN EXAMPLE 488
10.1. Information dependency and intensity 489
10.2. Information flow and operation of the supply chain 491
10.3. Analysis of information accuracy 496
11. ENSURING A POSITIVE CONTRIBUTION OF INFORMATION SYSTEMS TO THE ENTERPRISE 
499 
1. Development of enhanced manufacturing operations based on a sound business strategy 500
2. Definition of an IT strategy to support the business strategy 500
3. Implement an IT strategy to lead the company once it has been possible to improve its manufacturing operations 
500 
12. CONCLUSIONS 503
REFERENCES 504
MODELLING TECHNIQUES IN INTEGRATED OPERATIONS AND INFORMATION SYSTEMS IN MANUFACTURING SYSTEMS 
507 
1. INTRODUCTION 507
1.1. Review of integrated modelling simulation methods or tools for manufacturing systems analysis, design and performance evaluation 
509 
1.2. Research objectives 512
2. THE PCBA SYSTEM 513
3. SIMULATION TOOLS USED 515
3.1. Operational system model development based on ARENA 3.0 516
3.1.1. Operational system model development 
517 
3.1.2. Experimentalframe developmentfor ARENA models 521
3.1.2.1. INPUT DATA ACQUISITION AND ANALYSISFOR STOCHASTIC SYSTEM MODELS. 525
3.1.3. Simulation data analysis 
527 
3.2. Communication system model development based on COMNET III 528
3.2.1. Nctwore description and modelling constructions 528
3.2.1.1. MODELLING OF NETWOHK TOPOLOG IES. 530
3.2.1.2. NETWORK TRAFFIC AND WORKLOAD. 
533 
3.2.2. Network simulation 536
4. INTEGRATED MODEL APPROACH 536
4.1. Establishment of an integrated model 536
4.1.1. Operational system 540
4.1.2. Information processing system 
543 
5. SIMULATION RESULTS, ANALYSIS AND DISCUSSION 547
5.1. Operational system's aspects 
549 
5.1.1. Line-balancing and collecting critical data 
550 
5.1.2. Using animated simulation to investigate system performances 
551 
5.2. Information system's aspects 552
5.2.1. Channel utilisation (%) 
554 
5.2.2. Maximum message delay (ms) 
555 
5.2.3. Comparative dynamic performance of LANs for the PCBA system 
556 
5.2.3.1. CHANNEL UTILISATION (%) AND MAXIMUM MESSAGE DELAY (MS) VS TRANSMISSION RATES (MBPS) . 
557 
5.2.3.2. CHANNEL UTILISATION (%) AND MAXIMUM MESSAGE DELAY (MS) VS MAXIMUM MESSAGE SIZES (Kb). 
558 
6. DISCUSSION AND CONCLUSION 561
REFERENCES 564
TECHNIQUES AND ANALYSES OF SEQUENTIAL AND CONCURRENT PRODUCT DEVELOPMENT PROCESSES 
566 
1. INTRODUCTION 566
2. SEQUENTIAL ENGINEERING 567
2.1. Sequential product development process 
567 
2.2. Characte ristics of sequential engineering 569
3. CONCURRENT ENGINEERING 569
3.1. Concurrent product development process 569
3.1.1. Data transfer between activities in concurrent product development process 
570 
3.1.2. Loops of concurrent product development process 
570 
3.1.3. Team work 574
3.1.3.1. TEAM STRUCTURE IN CONCURRENT PRODUCT DEVELOPMENT PROCESS. 574
3.1.3.2. TEAMS IN BIG COMPANY. 576
3.1.3.3. TEAM STRUCTURE IN SMEs. 577
3.2. Organisational structures 581
3.2.1. Functional organisational structure 581
3.2.2. Project organisational structure 
583 
3.2.3. Matrix organisational structure 
584 
3.2.4. Organisational structure of team work in SME 
585 
3.3. Goals and tools for support of concurrent product development process 588
a.) Considerably shorter newproduct development time 588
b.) Reduced newproduct development costs 588
c.) Better quality of newproducts regarding customer requirements 588
3.3.1. Quality Functions Deployment (QFD) 589
3.3.1.1. HOUSE OF QUALITY STRUCTURE. 591
3.3.1.2. STEPS IN CONSTRUCTING THE HOUSE OF QUALITY. 592
3.3.1.3. EXTENDING THE HOUSE OF QUALITY. 595
3.3.2. value analysis 597
3.3.3. Failure Mode and Effects Analysis (FMEA) 
601 
4. SAMPLE CASE OF INTRODUCTION OF CONCURRENT ENGINEERING IN AN SME 
606 
4.1. Building a house of quality 607
4.2. Project of concurrent product development process 609
4.2.1. Goals of the project and project team 
609 
4.2.4. Time and structural plan of the project 
610 
5. CONCLUSION 618
REFERENCES 619
DESIGN AND MODELING METHODS FOR COMPONENTS MADE OF MULTI-HETEROGENEOUS MATERIALS IN HIGH-TECH APPLICATIONS 
620 
1. INTRODUCTION 620
2. DESIGN METHOD FOR THE COMPONENTS MADE OF MULTI HETEROGENEOUS MATERIALS 
623 
2.1. Design procedure 623
2.2. Material design 627
2.3. How to determine the optimal material properties needed in different regions 
630 
2.4. How to select material constituent composition and microstructure 634
2.4.1. Select material constituent compositions from the database of material constituent composition 634
2.4.2. Select material murostructurcsirom the database of material microstructure 636
2.5. How to generate two material region sets 636
(1) Encode decision variables 637
(2) Determine the size of population 637
(3) Evaluation 637
(4) Selection 639
(5) Crossover operation 639
(6) Mutation 641
(7) Reproduction 642
(8) Stop criterion 642
(9) Construct the regions for different material constituent compositions and material microstructures 
642 
2.6. An example of multi heterogeneous component design 643
(1) Requirements for material design 643
(2) Generate material regions 644
(3) Create optimization model 644
(4) Sensitivity analysis of material properties 644
(5) Search for the optimal material property vector of different regions of the flywheel 
644 
(6) Select material constituent composition and microstructure for each region 645
3. CAD MODELING METHOD FOR THE COMPONENTS MADE OF MULTI HETEROGENEOUS MATERIALS 
646 
3.1. Analyses of the requirements for representing the components made of heterogeneous materials 
646 
3.2. Unified CAD modeling for th e compo nent made of heterogeneous materials 
648 
3.3. Material constituent composition models 
650 
3.4. Material microstructure models 651
3.4.1. Material microstructure modelsfor composite materials (R) 
651 
3.4.2. Material microstructure models for heterogeneous materials with a periodic microstructure (P) 
652 
3.5. Main model for integrating the two types of sub-models 656
3.6. An example of modeling 
657 
4. FINITE ELEMENT ANALYSIS BASED ON THE MODEL 661
5. SUMMARY 663
ACKNOWLEDGEMENTS 664
REFERENCES 665
QUALITY AND COST OF DATA WAREHOUSE VIEWS1 
667 
1. INTRODUCTION 667
2. NON-EQUIVALENT QUERY REWRITINGS 669
3. EFFICIENCY MODEL: QUALITY OF A QUERY REWRITING 670
3.1. Information preservation in rewritings 670
3.2. Information preservation on the view interface 672
3.2.1. Dispensable and replacable attributes 672
3.3. Information preservation on view extent 674
3.4. Metric of quality: Degree of Divergence (D,D) 
675 
3.4.1. Degree of divergence on the query interface (DVattr(Vi)) 
675 
3.4.2. Degree of divergence on the query extent (DDext (Vi)) 
676 
3.4.3. Total degree divergence 
677 
4. EFFICIENCY MODEL: VIEW MAINTENANCE COST OF A LEGAL REWRITING 677
4.1. View maintenance basics 677
4.2. Cost factor based on number of messages exchanged (CFM ) 678
4.3. Cost factor based on bytes of data transferred (CFT ) 678
4.4. Cost factor based on I/O (CFI/O) 
678 
4.5 . Total view maintenance cost for a single data update 679
4.6. Overall efficiency of a legal rewriting 679
5. REVIEW OF THE EVE PROJECT 680
5.1. A relaxed SQL query modcl-E-SQL 
681 
6. IMPLEMENTATION AND EVALUATION 683
6.1. Implementation of the EVE System 
683 
6.2. Evaluation and discussion 684
6.2.1. Influence of relation distribution on view maintenance cost 685
6.2.2. Effect of relation cardinality on QC-value 686
6.2.3. Experiments on accuracy of cost model prediction 690
7. RELATED WORK 694
8. CONCLUSION 697
ACKNOWLEDGMENTS 697
REFERENCES 698
WEB DATA EXTRACTION TECHNIQUES AND APPLICATIONS USING THE EXTENSIBLE MARKUP LANGUAGE (XML) 
702 
1. INTRODUCTION 702
2. WEB DATA EXTRACTION 703
2.1. Why Web data is important 703
2.2. Core technologies behind the World Wide Web 703
2.3. The challenges of web data extraction 704
2.4. Using XML technologies in web data extraction 706
3. FROM WEB TO SYSTEMS 706
3.1. Business requirements 706
3.2. Database-centric data extraction 707
3.3. Crawler-based data extraction 708
3.4. Challenges 712
3.5. Techniques for effective data extraction 716
4. OUTLINE OF A DATA EXTRACTION SYSTEM ARCHITECTURE 718
4.1. Data retriever 719
4.2 . Data extractor 719
4.3 . Data checker 721
4.4 . Data exporter 721
4.5. Scheduler 722
4.6. Administrative interface 722
4.7. Pattern designer 
725 
5. DATA EXTRACTION PRINCIPLES 726
5.1. Extraction templates 726
5.2. Extracting XML data from HTML 728
5.3. Pattern creation 728
5.4 . Sample pattern analysis 729
6. CONCLUSION 733
BIBLIOGRAPHY 733
PRODUCT LIFE CYCLE MANAGEMENT IN THE DIGITAL AGE 
736 
1. INTRODUCTION 736
2. THE NEW PARADIGM OF LIFE CYCLE MANAGEMENT 737
2.1. Partnerships for sustainabl e product life cycles 740
2.1.2. Customer's view 741
2.1.3. Life cycle objectives 741
2.2. Economical assessment of product life cycles 742
3. THE DIGITAL AGE-ACTIVATING HIDDEN PERFORMANCE POTENTIALS 746
3.1. Digital product tracing 748
3.2. Boosting utilization performance 748
3.3. Workplaces on change 750
3.4. Product data management for high data continuity 750
4. ALLIANCES AND LIFE-LONG NETWORKS 753
4.1. Product life-time value 
755 
4.2. Selling benefit instead of usage 756
5. INDUSTRIAL PROTOTYPES OF DIGITALLY NETWORKED PRODUCT LIFE CYCLE MANAGEMENT 
758 
5.1. Example of online process monitoring 758
5.2. Example of the digital factory of the future 759
5.2.1. Structure of the platform 
760 
5.2.2. Data transjer 761
5.3. Example for the web-based control of a technical consumer product 761
5.3.1. System structure 761
5.3.2. Software-/Hardware architecture 
763 
5.3.3. Extension of system bounds-afuture vision 763
6. CONCLUSION AND OUTLOOK 763
REFERENCES 765
PRODUCT REDESIGN AND PRICING IN RESPONSE TO COMPETITOR ENTRY: A MARKETING-PRODUCTION PERSPECTIVE 
767 
1. INTRODUCTION 767
2. MODEL FORMULATION 770
2.1. Model notation 771
2.2. Attraction and market share models 771
2.3. Profit maximization objective 773
3. EXISTENCE OF A NASH EQUILIBRIUM 773
4. PRODUCT AND PRICE RESPONSES TO MARKET ENTRY 775
5. NUMERICAL EXAMPLE AND SENSITIVITY ANALYSIS 777
6. CONCLUSION 780
APPENDIX A 783
APPENDIX B 
786 
REFERENCES 788
KNOWLEDGE DISCOVERY BY MEANS OF INTELLIGENT INFORMATION INFRASTRUCTURE METHODS AND THEIR APPLICATIONS 
790 
INTRODUCTION 790
Neural Online Analytical Processing System (NOLAPS) 791
Neural Fuzzy Model 795
Cross-platform intelligent information infrastructure 799
Conclusion 801
REFERENCES 801
VOLUME III. EXPERT AND AGENT SYSTEMS 833
TECHNIQUES IN KNOWLEDGE-BASED EXPERT SYSTEMS FOR THE DESIGN OF ENGINEERING SYSTEMS 
834 
QUOTATION 834
1. INTRODUCTION 834
2. CHARACTERISTICS OF KNOWLEDGE-BASED EXPERT SYSTEMS 838
2.1. Domain knowledge 839
2.2. Inferential knowledge 841
3. KNOWLEDGE-BASED TECHNIQUES AND THEIR APPLICATION IN ENGINEERING DESIGN 
844 
3.1. Implementation-specific knowledge-based techniques 844
3.1.1. Rule-based representation 
845 
3.1.2. Semantic networks 
846 
3.1.3. Frame-based representation 
846 
3.1.4. Object-onented representation 
848 
3.1.5. Logic-based representation 
850 
3.1.6. Fuzzy logic 
852 
3.2. Generic knowledge-based techniques 
853 
3.2.1. Control strategies 
853 
3.2.2 . Search strategies 
855 
3.2.3. Constraint processing 
858 
3.2.4. Case-based reasoning 860
3.2.5. Blackboard architecture 
861 
4. KNOWLEDGE-BASED APPLICATION IN FUNCTIONAL DESIGN 862
4.1. B-FES functional modeling framework 863
4.2. Acquisition of functional design knowledge through two-level knowledge modeling 
864 
4.3. Knowledge-based functional representation scheme 867
4.3.1. Rule-based representation in rule base 
867 
4.3.2. Fuzzy logic in FMCDM model base 868
4.3.3. Knowledge-basedJunctional representation in an object-oriented behavior base 868
4.4. Knowledge-based functional reasoning strategy 870
4.5. Best-first heuristic search in functional reasoning 872
4.5.1. Weighted performallce rating aggregation of a mechanical device 
872 
4.5.2. Dynamic evaluation index of a design alternative 
874 
4.6. Case study 875
4.6.1. Problem description and user input 875
4.6.2. Automated functional design process and system output 875
5. CONCLUSION 879
REFERENCES 881
EXPERT SYSTEMS TECHNOLOGY IN PRODUCTION PLANNING AND SCHEDULING 
886 
1. INTRODUCTION 886
2. THE EXPERT SYSTEMS TECHNOLOGY 887
3. EXPERT SYSTEMS IN PRODUCTION PLANNING & SCHEDULING
4. EXPERT SYSTEMS RESEARCH IN PRODUCTION PLANNING & SCHEDULING
5. GENESYS: A QUICK CASE STUDY 
895 
5.1. Introduction 895
5.2. Problem analysis 896
5.3. The knowledge base 897
5.4. Construction & features
5.5. Performance evaluation 901
6. CONCLUSIONS/RECOMMENDATIONS 902
REFERENCES 903
APPLYING INTELLIGENT AGENT-BASED SUPPORT SYSTEMS IN AGILE BUSINESS PROCESSES 907
1. INTRODUCTION 907
2. INTELLIGENT AGENT FRAMEWORK 910
2.1. Intelligent agent system environment 911
2.2 . Architecture of an IA 912
2.3. The agent communication 916
2.3.1. The agent communication language 916
2.3.2. The content language 917
2.3.3. The agent conversation policy 
919 
3. AGENT-BASED OBJECT-ORIENTED DESIGN PROCESSES 924
3.1. An object-oriented approach 926
3.1.1. The concept of a design object 
926 
3.1.2. A design process model formalism for DwO 
927 
3.1.3. Modular software components 929
3.1.4. Object-oriented approach in modular software components 931
3.2. Agent-based system in design process 932
4. AGENT-BASED SUPPLY CHAIN PROCESSES 936
4.1. Classification of intelligent supply chain agents 937
4.2. Architecture of agents 943
4.2.1. Message handling process 943
4.2.2. The XML-based contents of the message 
945 
4.2.3. Basic alient architecture 
947 
5. AGENT-BASED SYSTEMS OF KNOWLEDGE MANAGEMENT 953
5.1. The definition of agents 953
5.2. The architecture of the agent-based system 954
5.3. Process 4: Distribute knowledge passively 956
6. CONCLUSIONS 958
REFERENCES 959
THE KNOWLEDGE BASE OF A B2B eCOMMERCE MULTI-AGENT SYSTEM 
963 
1. INTRODUCTION 963
2. RELATED WORK 964
3. AGENT INFERENCE MODEL (AIM) 967
4. CASE STUDY SCENARIO 969
5. CREATING THE KNOWLEDGE-BASE 972
6. ARCHITECTURE 976
7. IMPLEMENTATION & RESULT
8. CONCLUSION 983
REFERENCES 983
FROM ROLES TO AGENTS: CONSIDERATIONS ON FORMAL AGENT MODELING AND IMPLEMENTATION 
985 
1. INTRODUCTION 985
1.1. Complex systems 
985 
1.2. Application protocols 986
1.3. Distributed objects 987
2. AGENT-ORIENTED PROGRAMMING 987
2.1. Sub-protocols 988
2.2. Agent-UML 988
2.3. From objects- to agent-oriented programming 989
2.4. Method and message passing unification 991
2.5. Internal state control of an agent 992
3. ROLES AND SCENARIOS 993
3.1. Definitions 993
3.2. Artifacts for development 995
3.3. Roles and scenarios as programming artifacts 
998 
3.3.1. State control 
998 
3.3.2. State space inheritance 999
3.3.3. Genericity and composability 1001
3.4. On state inheritance 1001
4. THE DHELI TOOL 1003
4.1. The interaction-oriented programming framework 1003
4.2. System runtime interfaces 1004
4.3. Communication interfaces 1006
4.4. The DHELI language 1007
4.4.1. Entities 1007
4.4.2. Communication acts 1008
4.4.3. Variables, role variables and meta-roles 1010
5. CONCLUSION AND FUTURE WORK 1011
REFERENCES 1011
AGENT-BASED eLEARNING SYSTEMS: A GOAL-BASED APPROACH 
1013 
1. INTRODUCTION 1013
2. OVERVIEW OF THE AGENT COMPOSITE GOAL MODEL 1014
3. EXTENSION OF THE AGENT COMPOSITE GOAL MODEL 1017
3.1. Action selection 1017
3.1.1. Bayesian inference 
1017 
3.1.2. Discussion 1018
3.2. Multi-agent modeling 1018
3.2.1. Agent identification 1019
3.2.2. Coordination 1019
3.2.3. Communication 1020
3.2.4. Summary 1020
4. E-LEARNING MODEL 1020
4.1. Goal-based modeling 1020
4.2. E-learning modeling 1021
5. E-LEARNING SYSTEM DEVELOPMENT 1024
5.1. E-learning system architecture 1024
5.2. E-Iearning system development 1026
6. CONCLUSION AND FUTURE WORK 1027
6.1. Conclusion 1027
6.2. Future work 1027
REFERENCES 1027
COMBINING TEMPORAL ABSTRACTION AND DATA MININGMETHODS IN MEDICAL DATA ANALYSIS 
1029 
1. INTRODUCTION 1029
2. TEMPORAL ABSTRACTION AND DATA MINING METHODS 1031
2.1. Temporal abstraction methods 1031
2.2. Data mining methods 1032
3. THE HEPATITIS DATABASE AND A FRAMEWORK FOR COMBININGTEMPORAL ABSTRACTION WITH DATA MINING METHODS 1033
3.1. The hepatitis database and problems 1033
3.2. Preprocessing for hepatitis data 1035
3.2.1. Feature selection and data reduction 
1035 
3.2.2. Extraction of data subsets 1036
4. A TEMPORAL ABSTRACTION METHOD IN THE HEPATITIS DOMAIN 1037
4.1. Determination of typical abstraction patterns 1037
4.1.1. The TA primitives 1037
4.1.2. Observation and determination of absttattion patterns 1038
4.1.3. Relations between TA primitives 1038
4.2. Temporal abstraction algorithms for extracting abstraction patterns 1040
4.2.1. Notations and parameters used in the algorithms 1041
4.2.2. Abstraction of short-term changed tests 1044
4.3 . Abstraction of long-term changed tests 1044
5. MINING ABSTRACTED DATA BY DATA MINING METHODS 1045
5.1. The statistical significance of discovered knowledge 1045
5.2. Mining abstracted hepatitis data with system D2MS and Clementine 1046
6. CONCLUSIONS 1052
7. ACKNOWLEDGMENTS 1052
REFERENCES 1052
DISTRIBUTED MONITORING: METHODS, MEANS AND TECHNOLOGIES 
1054 
1. INTRODUCTION 1054
2. FEATURES AND FUNCTIONS OF MONITORING SYSTEMS 1055
3. MONITORING ARCHITECTURES 1056
3.1. Centralized monitoring 1056
3.2. Hierarchical monitoring 1058
3.3 . Distributed monitoring 1059
4. SOFTWARE PARADIGMS FOR MONITORING 1061
4.1. Client-Server 1061
4.2. Remote Evaluation (or code pushing) 1062
4.3. Code on Demand (or code pulling) 1062
4.4. Mobile Agents 1063
5. SOFTWARE MOBILITY FOR DISTRIBUTED MONITORING 1064
5.1. Location transparency and location awareness 1065
5.2. Mobile code systems 1065
5.3. Mobility mechanisms 1066
6. TECHNOLOGIES AND STANDARDS FOR MONITORING 1066
6.1. Internet/IETF monitoring 1067
6.2. OSI monitoring 1068
6.3. CORBA 1069
6.4. Java 1070
6.5. SOAP 1071
6.6. OSA, Parlay and Jain 1071
7. EMERGING APPROACHES TO DYNAMIC MONITORING 1072
7.1. Management by Delegation (MbD) 1072
7.2. MbD in the context of internet management 1073
7.3. MbD in the context of OSI management 1074
7.4. Mobile agents for distributed monitoring 1075
7.5. Potential benefits of MA-based management 1077
7.6. Open issues of MA-based management 
1078 
8. CONCLUSIONS 1079
ACKNOWLEDGEMENTS 
1080 
REFERENCES 1080
FINDING PATTERNS IN IMAGE DATABASES 
1085 
1. INTRODUCTION 1085
2. RELATED WORK 1086
2.1. Preprocessing 1086
2.2. Pattern discovery 1087
2.2.1. Association mining in image data 
1087 
2.2.2. Clustering in image data 1088
2.2.3. Classification in image data 
1089 
2.3. Image-specific considerations 1090
2.3.1. Semantic information 1090
2.3.2. Spatial relationship 1090
3. VIEWPOINT PATTERN DISCOVERY 1091
3.1. Overview 1092
3.2 . Algorithm ViewpointMiner 1092
4. EXPERIMENTS 1095
4.1. General category images 1095
4.2. Retinal images 1098
4.3. Kitchen plan images 1100
5. CONCLUSION 1101
REFERENCES 1102
COGNITION TECHNIQUES AND THEIR APPLICATIONS 
1104 
1. EVOLUTION OF THE MODEL OF COGNITION 1104
1.1. The cycles of model of cognition 1105
2. SCOPE OF REALIZATION OF THE ACQUISITION CYCLE 1109
3. BUILDING PERCEPTION CYCLE 1116
3.1. Need for map building 1117
3.2. Map building by depth first search 1118
3.2.1. Algorithms for map-building by depth first search 
1120 
3.2.2. A illustration of procedure traverse boundary 
1122 
3.2.3. An illustration of procedure map building 
1123 
3.2.4. Simulation on Superscoutt-ll Linux based graphics interface 
1123 
3.3. Construction of 3D world map by depth first search 1124
4. LEARNING AND COORDINATION CYCLE 1129
4.1. Learning for obstacle avoidance 1130
4.1.1. The constraints in the navigation process 1131
4.1.2. Learning for local guidance through Neural Net 
1133 
4.1.3. Building the Third Neural Net 1136
4.2. Planning by bi-directional associative memory 1139
4.2.1. Temporal associative memory in mobile robot navigation 1140
4.3. Planning using evolutionary algorithm 1144
4.3.1. Simulation for EC planning 1148
5. CONCLUSIONS 1148
REFERENCES 1150
VOLUME IV. INTELLIGENT SYSTEMS 1182
ARTIFICIAL INTELLIGENCE AND INTEGRATED INTELLIGENT SYSTEMS IN PRODUCT DESIGN AND DEVELOPMENT* 
1183 
1. INTRODUCTION 1183
2. OVERVIEW OF EVOLUTION OF PRODUCT-PROCESS DESIGN METHODOLOGIES 
1184 
2.1. Automated & integrated design
1185 
2.2. Concurrent engineering & concurrent design
2.2.1. Concurrent engineering 
1186 
2.2.2. Design for X 
1187 
2.2.3. Product life cycle management 
1188 
2.3. Intelligent computer-aided design 1188
2.4. Virtual prototyping 1189
2.5. Computer supported collaborative design 
1189 
3. ARTIFICIAL INTELLIGENCE IN PRODUCT-PROCESS DESIGN 1190
3.1. Intelligent product design 1190
3.2. Intelligent process planning 1191
3.3. Intelligent production system layout and design 1192
3.4. Intelligent simulation 1193
4. INTELLIGENT SYSTEMS FOR PRODUCT-PROCESS DESIGN 1193
4.1. Symbolic reasoning systems 
1193 
4.2. KBE and coupling intelligent systems 1194
4.3. Artificial neural network systems 1195
4.4. Genetic algorithms and systems 1197
4.5. Case-based reasoning systems 1199
4.5.1. Case-based reasoning model 1200
4.5.2. Analogical reasoning for design problem solving 
1201 
4.5.3. Design prototypes and cases 
1203 
4.6. Integrated & distributed intelligent systems
4.7. Hybrid intelligent systems 1206
5. A GENERIC FRAMEWORK FOR INTEGRATED INTELLIGENT DESIGN 1209
5.1. Issues and requirements for integrated intelligent design 1210
5.2. Framework for integrated intelligent design 1214
5.2.1. Working environment 1214
5.2.2. AI protocol based integrated intelligent design 1214
5.2.3. Architecture of framework 
1216 
(1)Designer Communication Layer 1216
(2) Core System and Control Layer 1217
(3) Application Layer 1218
6. IMPLEMENTATIONS OF INTEGRATED INTELLIGENT DESIGN SYSTEMS 1218
6.1. Integrated distributed collaborative design and assembly planning 1219
6.2. AI-supported internet-enabled virtual prototyping 1221
(1 ) Distributed Artificial Intelligence/Multi-Agent System 1222
(2) Feature-Based Virtual Model Representation 1223
(3)Neural Networks for Modeling Module/System's Dynamics 1223
(4) Optimization with Genetic Algorithms 1223
(5) Case-based Design 1223
6.3. A web-based knowledge intensive design support system 1225
6.3.1. Knowledge-based systems as knowledge servers 
1225 
6.3.2 . WebDMME framework and implementation 
1226 
7. CONCLUSIONS 1228
REFERENCES 1230
INTELLIGENT PATIENT MONITORING IN THE INTENSIVE CARE UNIT AND THE OPERATING ROOM 
1240 
1. INTRODUCTION 1240
2. TEMPORAL PATTERN RECOGNITION 1245
2.1. Template-based methods 1246
2.2 . Signal processing of the ECG 1249
3. REASONING METHODS 1251
3.1. Fuzzy logic 1251
3.2 . Eviden ce-based reasoning 1254
3.3. Bayesian networks 1257
3.4 . Artificial neural networks 1259
4. INTELLIGENT INTENSIVE CARE UNIT MONITORS 1263
5. INTELLIGENT ANAESTHESIA MONITORS 1265
5.1. Intelligent alarms 1268
6. SMART SENSORS 1269
6.1. Detection of esophageal intubation 1270
6.2 . Depth of anaesthesia 1271
6.3. Cardiac output 1274
7. AUTOMATIC CONTROL IN THE ICU AND OR 1276
7.1. Mean arterial blood pressure 1277
7.2. Depth of anaesthesia 1282
7.3. Multiple drug infusion-cardiac output 1283
8. INTERFACE DESIGN 1285
8.1. Display of anaesthesia information 1287
8.2. Data display for intelligent monitors 1288
9. DISCUSSION 1292
ACKNOWLEDGEMENTS 1293
REFERENCES 1293
MISSION CRITICAL INTELLIGENT SYSTEMS 
1300 
1. INTRODUCTION 1300
2. DEFINITIONS 1301
3. REAL TIME EXPERT SYSTEMS 1304
4. FAULT TOLERANCE IN INTELLIGENT SYSTEMS 1307
4.1. Failure detection and recovery in intelligent systems 1309
4.2. An architecture for a dependable intelligent system 1310
5. DISTRIBUTED MISSION CRITICAL INTELLIGENT SYSTEMS (DMCIS) 1312
6. COORDINATION IN DISTRIBUTED INTELLIGENT SYSTEMS 1314
6.1. MINUTE: Multi issue negotiation under time constrained environments 1315
6.2. TRACE-Task and resource allocation in a computational economy 1317
6.3. MAS organization in TRACE 1318
6.4. Task allocation protocol 1318
6.5. Resource allocation protocol 1319
6.6. Experiments 1322
6.6.1. Reduction in decommitments 
1322 
6.6.2. Fairness of resource allocation 1322
6.6.3. Adaptiveness of the TRACE MAS 
1323 
7. CONCLUSIONS 1324
REFERENCES 1324
AN INTELLIGENT HYBRID SYSTEM FOR BUSINESS FORECASTING 
1327 
1. INTRODUCTION 1327
2. PROBLEM STATEMENT 1328
3. OBJECTIVE 1329
4. NEW AREAS FOR IRS APPLICATIONS 1330
5. ARCHITECTURE OF IFS 1330
6. OPERATING PROCEDURE OF IFS 1331
7. INTELLIGENT BUSINESS FORECASTER 1332
7.1. Architecture of IBF 
1333 
7.2. Operating procedure 1336
7.2.1. Self-organised learning 1336
7.2.2. Identification of Fuzzy Rules 
1337 
7.2.3 . Supervised learning 
1338 
7.2.4. Forecasting and retraining 
1339 
7.3. Case studies 1340
7.3.1. IBF vs multiple regression method 
1340 
7.3.1.1. CASE I. 1340
7.3.1.2. COMPARISON WITH A MULTIPLE REGRESSION MODEL. 1342
7.3.2. IBF vs conventional neural networks 
1343 
7.3.2.1. CASE II. 1343
7.3.2.2. CASE III. 1348
7.3.2.3. COMPARISON WITH CONVENTIONAL NEURAL NETWORKS. 
1351 
8. INTELLIGENT SCENARIO GENERATOR 1355
8.1. Truth valued flow inference 
1359 
8.2. Architecture 1360
8.3. Learning algorithm 1363
8.4. An illustrative example 1365
9. JUDGMENTAL ADJUSTMENT OF OBJECTIVE FORECASTS 1370
9.1. The FSD method 1372
9.1.1. The FSD model 
1372 
9.1.2. Formatting the problem 
1375 
9.1.3. Combining Experts' Adjustments 
1376 
a) Construction of the "a" Vector 1377
b) Construction of "R" Matrix 1378
c) Tra nsformat ion 1378
9.2. The fuzzy adjuster 1379
9.3. Validation 1380
9.3.1. Experimentation 
1381 
9.3.2. Experimental results 1381
10. THE IFS SOFTWARE 1385
10.1. System manager 1385
10.2. Intelligent business forecaster 1385
10.2.1. IBF Setting up procedure 
1386 
10.2.2. IBF Operating procedure 1386
10.3. Intelligent scenario generator 1386
10.3.1. ISG Setting up procedure 
1386 
10.3.2. ISG Operating procedure 
1386 
10.4. Fuzzy adjuster 1387
10.5. Database 1387
10.6. Knowledge base 1387
10.7 . User interface 1387
11. CONCLUSIONS 1387
REFERENCES 1388
INTELLIGENT SYSTEMS TECHNOLOGY IN THE FAULT DIAGNOSIS OF ELECTRONIC SYSTEMS 
1392 
1. INTRODUCTION 1392
2. THE DIAGNOSTIC PROCESS 1393
3. A SIMPLIFIED MODEL OF MACHINE INTELLIGENCE 1393
4. TRADITIONAL APPROACHES 1394
4.1. Rule-based systems 1394
4.2. Fault (decision) trees 1395
5. MODEL-BASED APPROACHES 1396
5.1. Fault models (or fault dictionaries) 1396
5.2. Causal models 1398
5.3. Models based on structure and behaviour 1398
5.4 . Diagnostic inference model 1401
6. APPROACHES BASED ON LEARNING 1403
6.1. Case-based reasoning 1403
6.2. Explanation-based learning 1404
6.3. Learning knowledge from data 1405
7. FUZZY LOGIC APPROACHES 1405
8. NEURAL NETWORK APPROACHES 1407
9. HYBRID APPROACHES 1411
10. DIAGNOSTIC STANDARDS 1413
11. COMMENTARY 1413
11.1. Rule-based approaches 1414
11.2. Model-based approaches 1414
11.3. Case-based approaches 
1415 
11.4. Fuzzy logic and neural networks 1415
11.5. Hybrid approaches 1416
12. FUTURE RESEARCH DIRECTIONS 1416
13. TOOLS FOR THE RAPID DEPLOYMENT OF AI-BASED DIAGNOSTIC SOLUTIONS 
1418 
13.1. Knowledge representation 1418
13.2. Diagnostic inference 1421
13.3. Summary 1425
14. CONCLUSIONS 1425
REFERENCES 1426
TECHNIQUES IN THE UTILIZATION OF THE INTERNET AND INTRANETS IN FACILITATING THE DEVELOPMENT OF CLINICAL DECISION SUPPORT SYSTEMS IN THE PROCESS OF PATIENT CARE 
1430 
WHY DO WE NEED IT TO COORDINATE CARE? 1430
Prevalence of chronic disease 1430
Poor quality of care and escalating health care costs 1431
Communication and coordination issues 1431
Promoting support for chronic disease management 1433
Chapter outline 1433
DEFINING KEY CONCEPTS 1434
Evidence-based medicine 1434
Guideline 1434
Protocol 1438
Care plan 1440
Pathway 1441
Workflow 1444
Relationship between the concepts 1447
INTERNET/INTRANET-ENABLED HEALTH INFORMATION NETWORKS 1447
CLINICAL GUIDELINES AND DECISION SUPPORT 1451
INTEGRATING GUIDELINES AND WORKFLOW INTO EHR DESIGN 1456
Instruction 1459
ERR system architecture for CIGs and workflows 1462
Case study 1-hypertension in diabetes 
1465 
Case study 2-early supported discharge 1467
CONCLUSION 1467
ACKNOWLEDGEMENTS 1472
RISK ANALYSIS AND THE DECISION-MAKING PROCESS IN ENGINEERING 
1477 
1. INTRODUCTION 1477
2. THE NEED FO R RISK MANAGEMENT 1477
3. RISK 1478
4. DECISION-MAKING PROCESS 1480
4.1. Basic concepts 1480
4.2. Decision trees 1481
4.3. Defining utility criteria 1482
5. RISK-ANALYSIS BASED DECISION PROCESS 1483
5.1. General framework for integrating risk to the decision making process 1484
5.2. Final remarks 1487
6. ACCEPTABILITY OF RISK 1488
7. OPTIMIZATION 1491
7.1. Basic optimization concepts 1491
7.2. Cost of saving human lives 1493
7.3. Life cycle costing 1495
7.3. 1. General aspects 1496
7.3.2. Basics oflife cycle costing 1497
8. EXAMPLES 1498
8.1. Allocation of resources to transport networks 1498
8.1.1. Basic considerations 1498
8.1.2. Decision criteria 1498
8.1.3. Accessibility 
1499 
8.1.4. Optimization of resource allocation 1500
8.1.5. Case study 
1501 
8.1.6. Summary and final remarks 
1503 
8.2. Design of structural systems 1503
8.2.1. Decision criteria 1503
8.2.2. Probabilistic model of theground motion 
1503 
8.2.3. Model of the probability of failure of the structural system 
1505 
8.2.4. Estimation of cost 1505
8.2.5. Optimization 1506
8.2.6. Summary and final remarks 
1507 
9. CONCLUSIONS 1508
REFERENCES 1508
MECHATRONICS AND SMART STRUCTURES DESIGN TECHNIQUES FOR INTELLIGENT PRODUCTS, PROCESSES, AND SYSTEMS 
1510 
1. INTRODUCTION 1510
2. ANALYSIS OF INDUCED-STRAIN ACTUATION 1512
2.1. Actuator-structure interaction 1512
2.1.1. Displacement analysis 
1513 
2.1.2. Output energy analysis 
1516 
2.2. Induced-strain actuators with compliant support 1517
2.2.1. Displacement analysis 1517
2.2.2. Output energy analysis 1518
2.3 . Displacement-amplified induced-strain actuators 1519
2.3.1. Displacement analysis 1519
2.3.2. Output energy analysis 
1521 
2.3.3. Optimal kinematic gain, G, for a given value of . 
1522 
2.4. Electric response 1525
3. ANALYSIS OF INDUCED-STRAIN ACTUATION FOR DYNAMIC APPLICATION 1528
3.1. Mechanical response 1531
3.2. Electric response 1533
4. DESIGN OF SMART STRUCTURES WITH INDUCED-STRAIN ACTUATORS 1535
4.1. Efficient static design 1537
4.2. Efficient dynamic design 1539
4.3. Quasi-static dynamic operation 1541
4.4. Undamped dynamic operation 1542
4.5. The damped dynamic system 1543
4.6. Design example of induced-strain actuation application 1545
5. DESIGN OF EMBEDDED ULTRASONICS SMART STRUCTURES FOR STRUCTURAL HEALTH MONITORING 
1550 
5.1. PWAS Ultrasonic transducers 1550
5.2. Shear-layer coupling between PWAS and structure 1554
5.2.1. Symmetric case 1556
5.2.2. Antisymmetric case 1557
5.2.3. Shear lag solution 1558
5.2.4. Pin-force model 1563
5.2.5 . Energy transier between the PWAS and the structure 
1565 
5.2.6 . Conditions for optimum energy transfer 
1567 
5.3. Lamb waves excited by PWAS 1568
5.3.1. Lamb wave solution under nonuniform shear-stress boundary excitation 
1569 
5.3.2. Ideal-bonding solution 
1575 
5.4. Pitch-catch PWAS experiments 
1578 
5.4.1. Experimental setup 1578
5.4.2. Excitation signal 1580
5.4.3. Lamb mode tuning 1581
5.4.4. Pitch-catch results 
1583 
6. SUMMARY AND CONCLUSIONS 1583
BIBLIOGRAPHY 1587
ENGINEERING INTERACTION PROTOCOLS FOR MULTIAGENT SYSTEMS 
1589 
1. INTRODUCTION 1589
1.1. Interaction protocols in multiagent systems 
1589 
1.2. Communication protocols 1590
1.2.1. Communication protocols in distributed systems 
1590 
1.2.2. Communication protocol engineering 1590
1.3. Engineering interaction protocols 1591
1.3.1. One approach 
1591 
1.3.2. Some tools. 1593
1.4. Overview of the beghera application 1594
2. THE ANALYSIS STAGE 1596
3. THE FORMAL DESCRIPTION STAGE 1598
3.1. Towards a new interaction modeling language 1599
3.2. A component-based approach 1600
3.3. Definition of micro-protocols 1601
3.4. The CPDL language 1604
3.5 . Graphical modeling languages for protocol's representation 1605
3.6. UAML and UAMLe languages 1608
3.7. A tool for supporting agent interaction protocol design 1608
4. THE VALIDATION STEP 1610
4.1. The reachability analysis 1611
4.2. One example 1611
4.3. The model-checking approach 1613
5. THE PROTOCOL SYNTHESIS STAGE 1613
5.1. Phase role 1613
5.2. Two methods for the protocol synthesis 1614
5.2.1. Protocol synthesis approach 1614
5.2.2. Direct execution of protocols in CPDL language 
1615 
5.2.3. Comparison of these two approaches 1616
6. THE CONFORMANCE TESTING STAGE 1617
7. CONCLUSION AND PERSPECTIVES 1618
REFERENCES 1620
VOLUME V. NEURAL NETWORKS, FUZZY THEORY AND GENETIC ALGORITHMS 
1653 
NEURAL NETWORK SYSTEMS TECHNOLOGY AND APPLICATIONS IN CAD/CAM INTEGRATION 
1654 
1. INTRODUCTION 1654
2. ARTIFICIAL NEURAL NETWORKS 1655
3. ANN TECHNIQUES FOR FEATURE RECOGNITION 1655
3.1. The topology 1655
3.1.1. Feedforward networks 
1655 
3.1.2. Competitive networks 1656
3.1.3. Recurrent networks 1657
3.1.4. The three-Iayer feed-forward neural network 
1658 
3.1.5. The four-layer feed-forward neural network 
1659 
3.1.6. The five-layer, perceptrons quasi-neural network 
1659 
3.2. Input representation 1659
3.2.1. 2D feature representation 
1660 
3.2.2. Face adjacency matrix code 1660
3.2.3. Face score vector 1660
3.2.5. F-adjaccncy matrix and V-adjacency matrix 
1661 
3.2.6. 2D input patterns of 3D feature volume 1665
3.2.7. A vector based on the partitioned view-contours of a given object 
1665 
3.2.8. Simplified sheleton 
1666 
3.3. The output format 
1666 
3.3.1. Each neuron corresponding to a feature class 
1666 
3.3.2. Neurons representing the information of the recognised feature 1667
3.3.3. A matrix file containing the code for each recognised feature and its machining directions 
1667 
3.4. The training method 1667
3.4.1. Back propagation algorithms 
1667 
3.4.2. Conjugate gradient algorithm by the authors 
1668 
3.4.3. Training method by Prabhakar and Henderson 
1669 
3.5. Summary of ANN-based feature recognition 
1669 
4. ANN TECHNIQUES FOR CAPP 1669
4.1. The topology 1670
4.1.1. Feedforward network 
1670 
4.1.2. Hopfield network 
1670 
4.1.3. Brain-State-in-a-Box (BSB) 1672
4.1.4. MAXNET 1672
4.2. Input representation 1673
4.2.1. Standardised image data 
1673 
4.2.2. Input vector with value ranging from 0 to 1 
1674 
4.2.3. Input vector with integer value 1674
4.2.4. Input vector in binary form 
1674 
4.2.5. Input vector in mixed form 
1674 
4.3. Output format 1675
4.3.1. Output vector in ordered binary form 
1675 
4.3.2. Output vector with special values 1675
4.3.3. One-unit output in binary form 
1675 
4.4. Training method 1676
4.4. 1. Unsupervised learning algorithm 
1676 
4.4.2. Back-propagation 
1676 
4.5. Summary of ANN-based CAPP 
1678 
5. ANN-BASED HYBRID APPROACHES TO CAPP 1678
5.1. CAPP using expert system and ANN techniques 
1678 
5.1.1. Expert system control module 
1678 
5.1.2. Neural network control module 
1679 
5.2. CAPP using ANN, fuzzy logic and expert system techniques 
1680 
5.3. CAPP using GA, ANN and Fuzzy logic techniques 
1682 
5.3.1. Input representation 
1682 
5.3.2. Output format 
1685 
5.3.3. Topology and the training method of the proposed neural network 
1685 
5.4. Summary of ANN-based hybrid methods 1685
6. CONCLUSIONS 1685
7. REFERENCES 1686
NEURAL NETWORK SYSTEMS TECHNOLOGY AND APPLICATIONS IN PRODUCT LIFE-CYCLE COST ESTIMATES 
1689 
1. INTRODUCTION 1689
2. BACKGROUND 1690
2.1. General product development 1690
3. AN APPROXIMATE ESTIMATION METHOD FOR THE PRODUCT LIFE CYCLE COST USING ANNS 
1691 
3.1. The concepts 1691
3.2. Development of the life cycle cost factors 
1693 
3.3. Development of product attributes 
1694 
3.3.1. General product attributes 
1695 
3.3.2. Maintainability attributes 
1695 
3.3.3. Determining the final product attributes using statistical analysis 
1696 
4. A CASE STUDY 
1699 
4.1. Data collection 
1700 
4.2. Development of training algorithms 
1701 
4.2.1. Backpropagation algorithm 
1701 
4.2.2. Development of training algorithm with backpropagation 
1702 
4.3. Testing and the results 1703
4.4. Discussion 1705
5. CONCLUSIONS AND FUTURE WORKS 1707
REFERENCES 1709
NEURAL NETWORK SYSTEMS TECHNOLOGY IN THE ANALYSIS OF FINANCIAL TIME SERIES 
1710 
INTRODUCTION AND CONTEXT 1711
TIME SERIES AND THEIR TECHNIQUES 1714
Financial time series 1715
Components of financial time series: 
1715 
Switching time series 1726
NEURAL NETWORKS FUNDAMENTALS 1727
Artificial neural networks 1728
The single neuron element 1729
Training recurrent networks 1742
Validation 1744
DATA PREPARATION FOR NEURAL NETWORKS 1745
Detrending 1746
Smoothing 1746
Normalizing and scaling the data 1747
Structuring the data 1747
Time series and neural networks applications 1749
APPENDIX 1759
BIBLIOGRAPHY 1760
FUZZY RULE EXTRACTION USING RADIAL BASIS FUNCTION NEURAL NETWORKS IN HIGH-DIMENSIONAL DATA 
1762 
INTRODUCTION 1762
1. FUZZY SET THEORY: BASIC DEFINITIONS AND TERMINOLOGY 1765
1.1. Fuzzy sets 1765
1.2. Linguistic variables and linguistic values 1766
1.3. Membership function formulation and parametrization 1766
1.4. Fuzzy set operations 1769
2. FUZZY REASONING AND FUZZY INFERENCE SYSTEMS 1770
2.1. Fuzzy if-then rules 1771
2.2. Approximate reasoning 1771
2.3. Fuzzy inference systems 1772
3. RADIAL BASIS FUNCTION NEURAL NETWORKS 1775
3.1. Definition 1776
3.2. Training 1778
3.2.1. Hidden layer definition : 1778
3.2.2 . Output layer definition 1778
3.2.3. Functional Equivalence to FIS 1779
4. ANFIS ARCHITECTURE 1780
5. OPTIMAL DESIGN OF RBFN BASED ON FUZZY CLUSTERING 1781
5.1. Fuzzy Clustering and similarity measures 1781
5.2. Toeplitz covariance matrix estimator 1784
5.3. Optimal number of clusters 1786
5.4. Output layer supervised training and rule extraction 1787
5.5. Algorithm: elliptical radial basis function network design 1789
6. EXPERIMENTS 1789
6.1. Synthetic data experiments 1790
6. 1. 1. Experiment 1 1790
6.1.2. Experiment 2 1792
6.2. Benchmark data experiments 1796
7. SUMMARY 1798
REFERENCES 1799
FUZZY DECISION MODELING OF PRODUCT DEVELOPMENT PROCESSES 
1802 
1. INTRODUCTION 1802
2. MODELLING PRODUCT DEVELOPMENT INFORMATION WITH FUZZY SETS 1804
2.1. Introduction to fuzzy set theory 1804
2.2. Representing imprecision and preference information with fuzzy sets 1805
2.3 . Measures of possibility and necessity 
1806 
3. A FUZZY SET APPROACH FOR PRIORITIZATION OF DESIGN REQUIREMENTS 
1807 
3.1. Problem formulation 1807
3.2. A fuzzy outranking preference model to prioritize design requirements 
1809 
3.2.1. Representing imprecise information in QFD 1809
3.2.2. A fuzzy outranking preference model for prioritizing design requirements 1810
3.3. Illustrative example 1813
4. A FUZZY SET APPROACH FOR SELECTION OF DESIGN CONCEPTS 1815
4.1. Problem formulation 1815
4.2. Fuzzy outranking preference model for concept selection 1817
4.2.1. Construction offuzzy outranking relations 1817
4.2.2. Determination ofnon-dominated design concepts 1818
4.3. Illustrated example 1819
5. A FUZZY SET APPROACH FOR SCHEDULING OF PRODUCT DEVELOPMENT PROJECTS 
1823 
5.1. Problem formulation 1823
5.2. A fuzzy scheduling model to minimize schedule risk 1824
5.2.1. Comparison of two fuzzy temporal parameters 
1824 
5.2.2. Performance measure of fuzzy project scheduling 
1825 
5.2.3. Fuzzy scheduling with a genetic algorithm 
1825 
5.3. lllustrative example 1829
6. CONCLUSION 1831
REFERENCES 1831
EVALUATION AND SELECTION IN PRODUCT DESIGN FOR MASS CUSTOMIZATION 
1834 
1. INTRODUCTION 1834
2. CURRENT STATUS OF RESEARCH 1835
2.1. Design alternatives evaluation and selection 1836
2.2. Product family design evaluation and selection 1837
3. CUSTOMER-DRIVEN PRODUCT FAMILY DESIGN FOR MASS CUSTOMIZATION 
1838 
3.1. Strategies and technical challenges for mass customization 1838
3.2. Customer-driven design for mass customization 1839
3.3. Module-based product family design 1841
3.4. Knowledge support framework for CDFMC 1843
4. PRODUCT FAMILY DESIGN EVALUATION AND SELECTION 
1844 
4.1. Knowledge decision support scheme 1845
4.2. Customization/evaluation metrics 1846
4.3. Fuzzy clustering and design ranking methodology 1847
4.3.1. Fuzzy clustering analysis for design 
1847 
4.3.2. Fuzzy ranking for design 
1850 
4.3.3. Simplified fuzzy ranking for design 
1851 
4.4. Evaluation of product family design alternatives 1852
4.4.1. Heuristic evaluation function 
1852 
4.4.2. Evaluation index 
1852 
4.5. Neural network adjustment for membership functions 1853
5. CASE STUDY AND SYSTEM PROTOTYPE 1855
5.1. Case study 1855
5.2. System prototype 1855
6. DISCUSSION 1858
7. SUMMARY AND CONCLUSIONS 1859
REFERENCES 1860
GENETIC ALGORITHM TECHNIQUES AND APPLICATIONS IN MANAGEMENT SYSTEMS 
1864 
1. INTRODUCTION 1864
1.1. Resource-constrained scheduling problem 
1864 
1.2. Classes of the generalized problem for resource-constrained scheduling 1865
1.3. Structure of this chapter 1867
2. RELATED WORK ON SCHEDULING 1867
2.1. Exact solution methods 1867
2.2. Heuristic solution methods 1868
3. INTRODUCTION TO GENETIC ALGORITHM 1868
3.1. The concept of genetic algorithms 1868
3.2. A simple example of genetic algorithms 1869
3.3. Packages of genetic algorithm components 1869
3.4. Applications of GA and when not to use 1871
4. SURVEY OF GA TECHNIQUES ON SCHEDULING 1871
4.1. Representation issues 1872
4.1.1. Indirect representation 1872
4. 1.2. Direct representation 1874
4.2 . Operators 1874
4.3. Comparison between different approaches 1875
4.4 . Other methods and issues 
1876 
5. APPLY GENERIC ALGORITHMS TO SOFTWARE ENGINEERING 1877
5.1. Software project management 1877
5.2. Introduction to SPMNET 1878
5.3. GA for software project management 1878
5.3.1. Task-basedmodel 1878
1. Model 1879
2. A Test Problem and its Results 1880
5.3.2. Timeline-based model 1881
1. Model 1881
2. Numerical Experiments 1881
6. CONCLUSION AND FUTURE WORK 1881
REFERENCES 1882
ASSEMBLY SEQUENCE OPTIMIZATION USING GENETIC ALGORITHMS 
1885 
1. INTRODUCTION 1885
2. BACKGROUND TO ASSEMBLY PLANNING AND OPTIMISATION 1886
3. THE ASSEMBLY SEQUENCE PLANNING PROBLEM 1888
4. LITERATURE REVIEW ON ASP 1889
5. THE APPROACH USED TO S/O THE ASP PROBLEM 1890
6. A MODEL OF THE ASSEMBLY PROCESS 1891
6.1. The graph of liaisons and the table ofliaisons 1892
6.2. The wave model of the assembly process 1894
7. A MODEL FOR THE ASSEMBLY SEQUENCES 1895
7.1. Representation of SLMC assembly sequences 1895
7.2. Generalisation of the representation for non-SLMC assembly plans 1896
8. A MODEL OF THE PRODUCT FOR ASP AND THE AUTOMATIC GENERATION OF FEASIBLE ASSEMBLY SEQUENCES 
1898 
8.1. Background 1898
8.2. Intrinsic precedence relations 1901
8.3. Guided search algorithm for generation of assembly sequences considering only IPR 
1902 
8.4. Extrinsic precedence relations 1903
8.5. Implementation of EPR 
1904 
8.5.1. EPRfor individual liaisons 1905
8.5.2. Boolean relations 1905
8.5.3. EPR for groups of liaisons 
1906 
8.5.4. Definition of the assembly table 1908
9. QUALITY MEASURES FOR ASP AND THE FITNESS FUNCTION 1909
9.1. The fitness function for a single optimisation criterion 1911
9.2. The fitness function for multi-criteria optimisation 1912
10. THE GENETIC ALGORITHM FOR THE OPTIMISATION OF ASSEMBLY SEQUENCES 
1914 
10.1. Automatic generation of assembly sequences by guided search 1914
10.2. The crossover operator 1915
10.3. The fitness function 1916
10.4. The selection process 1916
11. A CASE STUDY 1916
12. CONCLUSIONS 1919
13. REFERENCES 1921
KERNEL-BASED SELF-ORGANIZED MAPS TRAINED WITH SUPERVISED BIAS FOR GENE EXPRESSION DATA MINING 
1923 
1. INTRODUCTION 1923
2. KERNEL-BASED SELF-ORGANIZED MAP ADAPTATION 1924
3. THE KSDG-SOM ALGORITHM 1925
3.1. Initialization phase 1927
3.2. Training run adaptation phase 1927
3.2.1. Map adaptation rules 1928
3.2.2. Evaluation of the map training run convergence condition 1929
3.3. Expansion phase 1929
3.4. Fine tuning adaptation phase 1929
3.5. Evaluation of classification performances 1930
3.6. Model selection step 1930
3.7. Node deletion 1931
4. THE EXPANSION PROCESS 1931
5. CRITERIA FOR CONTROLLING THE KSDG-SOM DYNAMIC GROWING 1934
6. APPLICATION 1936
7. CONCLUSIONS 1938
8. ACKNOWLEDGMENT 1938
REFERENCES 1939
COMPUTATIONAL INTELLIGENCE FOR FACILITY LOCATION ALLOCATION PROBLEMS 
1940 
1. INTRODUCTION 1940
1.1. Facility location problem 1940
1.2. Location-allocation problem 1943
1.3. Mathematical programming 1944
1.4. Organization 1945
2. BACKGROUND 1945
2.1. Branch-and-bound 1946
2.2. Lagrange relaxation and sub-gradient method 1949
2.3. Local search 1953
2.4 . Genetic algorithm 1954
2.5. Simulated annealing 1955
3. HYBRID METHOD FOR LOCATION-ALLOCATION PROBLEM 1957
3.1. Nested GA (GA + GA) 1958
3.2. GA + Branch and Bound 1958
3.3. GA + Lagrange 1959
4. MIXED TYPE CHROMOSOME AND ALTERNATE LOCATION ALLOCATION 1961
5. CURRENT OPTIMIZATION SOFTWARE 1962
6. EXPERIMENTS 1964
7. CONCLUSIONS 1969
REFERENCES 1970
INDEX 1972