Despite the large volume of publications devoted to neural networks, fuzzy logic, and evolutionary programming, few address the applications of computational intelligence in design and manufacturing. Computational Intelligence in Manufacturing Handbook fills this void as it covers the most recent advances in this area and state-of-the-art applications.This comprehensive handbook contains an excellent balance of tutorials and new results, that allows you to:obtain current informationunderstand technical detailsassess research potentials, anddefine future directions of the field Manufacturing applications play a leading role in progress, and this handbook gives you a ready reference to guide you easily through these developments.
Author(s): Jun Wang, Andrew Kusiak
Series: The Mechanical Engineering Handbook Series
Edition: 1
Publisher: CRC Press
Year: 2000
Language: English
Pages: 560
Computational Intelligence in Manufacturing Handbook......Page 1
Preface......Page 3
Editors......Page 5
Contributors......Page 6
Table of Contents......Page 8
1.2 Knowledge-Based Systems......Page 14
1.3 Fuzzy Logic......Page 17
1.4 Inductive Learning......Page 20
1.5 Neural Networks......Page 24
1.6 Genetic Algorithms......Page 28
1.6.3 Genetic Operators......Page 29
1.6.3.2 Crossover......Page 30
1.6.3.5 Control Parameters......Page 31
1.7.1 Expert Statistical Process Control......Page 32
1.7.3 Induction of Feature Recognition Rules in a Geometric Reasoning System for Analysing 3D Ass.........Page 34
1.7.4 Neural-Network-Based Automotive Product Inspection......Page 35
1.7.5 GA-Based Conceptual Design......Page 36
References......Page 38
2.1 Introduction......Page 43
2.2 Modeling and Design of Manufacturing Systems......Page 45
2.3 Modeling, Planning, and Scheduling of Manufacturing Processes......Page 52
2.4 Monitoring and Control of Manufacturing Processes......Page 56
2.5 Quality Control, Quality Assurance, and Fault Diagnosis......Page 60
References......Page 65
3.1 Introduction......Page 72
3.2 Agent-Oriented Manufacturing Systems......Page 73
3.3 The MetaMorph Project......Page 74
3.3.1 The MetaMorphic Architecture......Page 75
3.3.2 Agent Coalition (Clustering)......Page 76
3.3.3 Prototype Implementation......Page 77
3.3.4 MetaMorph II......Page 78
3.3.5 Clustering and Cloning in MetaMorph II......Page 79
3.4 Holonic Manufacturing Systems......Page 80
3.4.1 Origin of the Holonic Concept......Page 81
3.5.1 Holonic MetaMorphic Architecture......Page 82
3.5.3 Holonic Self-Organization......Page 83
3.5.4 Holonic Clustering......Page 84
3.6.1 Rationale for Fuzzy Modeling of Multiagent Systems......Page 85
3.6.2 Mathematical Statement of the Problem......Page 86
Example 1......Page 87
Example 2......Page 90
3.6.3.2.1 On Measures of Fuzziness......Page 92
3.6.3.2.3 Shannon Entropy as an Adequate Measure of Fuzziness......Page 93
3.6.3.4 Identifying Holonic Structures by Constructing the Least Uncertain Source-Plan......Page 96
3.7 MAS Self-Organization as a Holonic System: Simulation Results......Page 97
3.7.1 Case Study 1......Page 98
3.7.2 Case Study 2......Page 102
References......Page 107
4.1 Introduction......Page 119
4.2 Artificial Neural Networks......Page 121
4.3.1 Pattern Classification Based on Design and Manufacturing Features......Page 123
4.3.2 Pattern Classification Based on Part–Machine Incidence Matrices......Page 128
4.3.2.1Evolution of Neural Network Methods for Part–Machine Grouping......Page 129
4.3.2.4Kohonen’s Self-Organizing Feature Map Model......Page 130
4.3.2.5Adaptive Resonance Theory (ART) Model......Page 131
4.3.2.6Fuzzy ART Model......Page 132
4.3.3Sequence-Dependent Clustering......Page 134
4.3.4Capacitated Cell Formation......Page 136
References......Page 137
5.1 Introduction......Page 142
5.2 A Multi-Criterion Decision-Making Approach for Evaluation of Scheduling Rules......Page 143
5.5 A Hierarchical Structure for Evaluation of Scheduling Rules......Page 145
5.5.2 Consistency of the Decision-Maker’s Judgment......Page 148
5.5.3 Advantages and Disadvantages of Multi-Criterion Decision-Making Methods......Page 150
5.6.2 The Main Approach......Page 152
5.6.3 Operation-Selection Rules and Membership Functions......Page 153
5.7 Fuzzy-Based Part Dispatching Rules in FMSs......Page 156
5.8.2 Fuzzy Expert Systems......Page 158
5.8.3.1 Selection of a Routing......Page 160
5.8.3.6 AGVs Select Parts......Page 161
5.9.1 Routing Selection Using Fuzzy Set......Page 162
5.9.2 Fuzzy Expert System-based Rule Approach for Dispatching of Parts......Page 164
Defining Terms......Page 166
References......Page 167
6.1 Introduction......Page 171
6.2 The Design of Cellular Manufacturing Systems......Page 172
6.3.1 Mathematical Formulation of the MCS Coefficient......Page 174
6.3.2 Parts Similarity Coefficient......Page 176
6.4.1 Chromosome Representation for Different Routings......Page 177
6.4.1.4 The Algorithm......Page 179
6.5.2 The Crossover Operator......Page 180
6.5.6 The Algorithm......Page 181
6.6.5 The Replacement Strategy......Page 182
6.7.2 Methodology for Layout Design......Page 183
6.8.2 The Crossover Operator......Page 184
6.8.3 The Fitness Function......Page 185
6.9.1 Finding the Best Alternative Routings......Page 186
6.9.2 Machine Grouping......Page 187
Defining Terms......Page 189
References......Page 194
For Further Information......Page 195
7.1.1 Information Retrieval Systems vs. Design Retrieving Systems......Page 197
7.2.1 Retrieving “Similar” Designs Instead of Identical Designs......Page 198
7.3 Structure of an Intelligent System......Page 199
7.3.3 Using a Fuzzy ART Neural Network as Design Associative Memory......Page 200
7.5 Implementation Example......Page 201
7.5.1 Constructing Geometric Memory......Page 202
7.5.4 Similarity Parameter to Control the Similarity of References......Page 203
7.5.6 Comparison with GT-Based Methods......Page 204
Defining Terms......Page 205
For Further Information......Page 206
8.1 Introduction......Page 208
8.2 A Mixed Integer Program......Page 210
8.3.1 Applying GAs on the Parallel Operations Process......Page 212
8.3.1.3 Generating Initial Population......Page 213
8.3.3 A New Crossover Operator......Page 214
8.4.1 Tabu Search......Page 216
8.4.3 Move Mechanism......Page 217
8.4.6 Proposed Tabu Search Algorithm......Page 218
8.5 Two Reported Examples Solved by the Proposed GA......Page 219
8.6.1 The 18-Operation Example......Page 225
8.6.2 The 26-Operation Example......Page 227
8.7 Random Problem Generator and Further Tests......Page 229
8.7.1 Performance Comparison......Page 231
8.8 Conclusion......Page 233
References......Page 239
9.1 Introduction......Page 242
9.2.1 The Process Planning Domain......Page 244
9.2.2 The Process Planning Problem......Page 245
9.2.2.3 TAD Selection......Page 246
9.2.2.5 Precedence Relationships between Operations......Page 247
9.2.3 The Process Plan Evaluation Criterion......Page 248
9.2.4 An Optimization Process Planning Model......Page 250
9.2.5 Set-Up Plan Generation......Page 253
9.3.2 Initial Population......Page 254
9.3.5 Crossover and Mutation Operators......Page 255
9.3.6 GA Parameters......Page 256
9.3.7 A Process Planning Example Using the GA......Page 258
9.4 Applying Simulated Annealing to the Process Planning Problem......Page 259
9.4.3 A Criterion for Accepting or Rejecting a Change......Page 261
9.4.6 A Criterion for Stopping Further Changes under a Temperature......Page 262
9.5 Comparison between the GA and the SA Algorithm......Page 264
9.6 Conclusions......Page 265
References......Page 266
10.2 Resource-Constrained Project Scheduling Problem......Page 269
10.2.1 Priority-Based Encoding......Page 270
10.2.1.1 Procedure of Topological Sort......Page 272
10.2.2.2 Swap Mutation......Page 273
10.2.3 Evaluation and Selection......Page 274
10.2.4 Experimental Results......Page 276
10.3 Parallel Machine Scheduling Problem......Page 277
10.3.1 Dominance Condition......Page 278
10.3.2 Hybrid Genetic Algorithms......Page 281
10.3.2.1 Representation......Page 282
10.3.2.3 Job Sequence......Page 283
10.3.2.4 Evaluation and Selection......Page 284
10.4.1 Basic Approaches......Page 285
10.4.2.3 Topological Sort-Based Representation......Page 286
10.4.2.9 Disjunctive Graph-Based Representation......Page 287
10.4.3 Adapted Genetic Operators......Page 288
10.4.3.5 Cycle Crossover (CX)......Page 289
10.4.3.10 Substring Exchange Crossover......Page 290
10.4.5 Hybrid Genetic Algorithms......Page 291
10.4.6 Discussion......Page 292
10.5.1 Problem Description......Page 293
10.5.2.3 Evaluation......Page 295
10.6 Part Loading Scheduling Problem......Page 296
10.6.1 Problem Description......Page 297
10.6.3 Measure of Diversity......Page 298
10.6.4 Numerical Experiments......Page 299
References......Page 300
11.1 Introduction to Neural Network Predictive Process Models......Page 304
11.2 Ceramic Slip Casting Application......Page 305
11.2.1 Neural Network Modeling for Slip Casting Process......Page 306
11.3 Abrasive Flow Machining Application......Page 307
11.3.1 Engine Manifolds......Page 308
11.3.2 Process Variables......Page 309
11.3.3 Neural Network Modeling for Abrasive Flow Machining Process......Page 311
11.4 Chemical Oxidation Application......Page 312
11.4.1 Neural Network Modeling for Chemical Oxidation Process......Page 313
Acknowledgment......Page 314
References......Page 315
12.1 Introduction......Page 317
12.2.1 Manufacturing Process Monitoring......Page 318
12.2.2 Manufacturing Process Control......Page 321
12.3 Neural Network-Based Monitoring......Page 322
12.3.1 Feature Selection Method......Page 323
12.3.2 Classification Method......Page 325
12.4 Quality Monitoring Applications......Page 326
12.4.1 Tapping Process......Page 327
12.4.2 Solder Joint Monitoring......Page 329
12.4.3 Pipe Welding Process......Page 330
12.4.4 Laser Surface Hardening Process......Page 333
12.5 Neural Network-Based Control......Page 335
12.6.1 Machining Process......Page 338
12.6.2 Arc Welding Process......Page 341
12.6.4 Hydroforming Process......Page 344
12.7 Conclusions......Page 347
References......Page 348
13.1 Introduction......Page 351
13.2 The Role of Computational Intelligence......Page 352
13.2.1 Neural Networks......Page 353
13.2.2 Genetic Algorithms......Page 356
13.2.3 Expert Systems......Page 358
13.3.1 Modeling Using Backpropagation Neural Networks......Page 361
13.3.2.1 Neural Networks and Simulated Annealing in Plasma Etch Modeling......Page 363
13.3.2.2.1 The Hybrid Neural Network Approach......Page 365
13.3.2.2.2 The Model Transfer Approach......Page 366
13.3.2.3 Process Modeling Using Modular Neural Networks......Page 368
13.4 Optimization......Page 369
13.4.1.1.1 Individual Network Parameter Optimization......Page 370
13.4.1.2 Network Optimization Using Genetic Algorithms......Page 371
13.4.1.2.2 Optimization for Multiple PECVD Responses......Page 374
13.4.2 Process Optimization......Page 376
13.4.2.2 Results for Multiple Output Synthesis......Page 379
13.5 Process Monitoring and Control......Page 382
13.5.1.1 Time Series Modeling......Page 383
13.5.1.2 Malfunction Detection......Page 384
13.5.2.1 Run-by-Run Neurocontrol......Page 386
13.5.2.2 Real-Time Neurocontrol......Page 388
13.6.1.1 Hybrid Expert System Approach......Page 391
13.6.1.1.2 On-Line Diagnosis......Page 392
13.6.1.1.3 In-Line Diagnosis......Page 394
13.6.1.2 Time Series Modeling Approach......Page 396
13.6.1.3 Pattern Recognition Approach......Page 399
13.6.2 Circuit-Level Diagnosis......Page 400
13.7 Summary......Page 402
Defining Terms......Page 403
References......Page 404
Further Information......Page 406
14.1 Introduction......Page 408
14.2.1 The Basic Concept of Fuzzy Sets......Page 409
14.2.2 Fuzzy Sets and Probability Distribution......Page 411
14.2.3 Conditional Fuzzy Distribution......Page 413
14.3.1 Using Fuzzy Systems to Describe the State of a Manufacturing Process......Page 415
14.3.2 A Unified Model for Monitoring and Diagnosing Manufacturing Processes......Page 417
14.3.3 Linear Fuzzy Classification......Page 418
14.3.4 Nonlinear Fuzzy Classification......Page 421
14.3.5 Fuzzy Transition Probability......Page 423
14.4.1 Tool Condition Monitoring in Turning......Page 430
14.4.2 Tool Condition Monitoring in Boring......Page 432
Acknowledgments......Page 434
References......Page 435
15.1 Introduction......Page 437
15.2.1 Combination of Fuzzy System and Neural Network......Page 438
15.2.2 Fuzzy Neural Network......Page 439
15.3.1 Wavelet Transforms (WT)......Page 443
15.3.2 Wavelet Packet Transforms......Page 445
15.4 Tool Breakage Monitoring with Wavelet Transforms......Page 446
15.4.2 Wavelet Analysis of Tool Breakage Signals......Page 447
15.5 Identification of Tool Wear States Using Fuzzy Methods......Page 448
15.5.1 Experimental Setup and Results......Page 449
15.5.2.1 The Model......Page 451
15.5.2.2 Fuzzy Classification......Page 452
15.5.3 Multi-Parameter Fusion with Fuzzy Inference......Page 454
15.5.3.1 Fusion......Page 455
15.5.4 Results and Discussion......Page 457
15.6 Tool Wear Monitoring with Wavelet Transforms and Fuzzy Neural Network......Page 459
15.6.1 Acoustic Emission Signals and Tool Wear......Page 461
15.6.2 Signal Analysis and Features Extraction......Page 462
15.6.3 Experiments and Results......Page 466
References......Page 469
16.1 Introduction......Page 473
16.2 Methodologies......Page 474
16.2.1.1 The Learning Process......Page 475
16.2.2.1 Step 1: Divide the Input and Output Spaces into Fuzzy Regions......Page 476
16.2.2.2 Step 2. Generate Fuzzy Rules from Given Data Pairs through Experimentation......Page 478
16.2.2.3 Step 3: Assign a Degree to Each Rule and Resolve the Conflicting Rules......Page 479
16.3 Experimental Setup and Design......Page 480
16.3.1 Training and Testing Experiments......Page 482
16.4 The In-Process Surface Roughness Recognition Systems......Page 483
16.4.1 ISRR-ANN Model......Page 484
16.4.2 ISRR-FN System......Page 485
16.5.2 Conclusions......Page 486
References......Page 489
17.1 Introduction......Page 492
17.1.1 Classification of Parameters......Page 493
17.1.2 Limitations of Existing Off-Line Parameter Design Techniques......Page 495
17.1.4 Chapter Organization......Page 496
17.2.1.1 Multilayer Perceptron Networks......Page 497
17.2.1.2 Training MLP Networks Using Backpropagation Algorithm......Page 498
17.2.1.3 Iterative Inversion of Neural Networks......Page 499
17.3 Design of Quality Controllers for On-Line Parameter Design......Page 500
17.3.1 Identification Mode......Page 502
17.3.2 On-Line Parameter Design Mode......Page 503
17.3.2.2 Stochastic Search Method......Page 504
17.4.1 Experimental Technique......Page 505
17.4.3 Process Modeling Using Multilayer Perceptron Networks......Page 506
17.4.4.1 Establishing Target Process Outputs......Page 508
17.4.4.3 Comparison of Performance of IQCs and SPD......Page 510
17.5 Conclusion......Page 512
References......Page 515
For Further Information......Page 516
18.1 Statistical Process Control......Page 519
18.2 Neural Network Control Charts......Page 521
18.3.1 Data Input Module......Page 522
18.3.2 Data Processing Module......Page 523
18.3.2.1 Computing in a Neural Network......Page 524
18.3.2.2 Training of a Neural Network......Page 525
18.3.3 Decision-Making Module......Page 526
18.3.3.1.1 Fuzzy Sets and Fuzzy Variables......Page 527
18.3.3.1.3 Fuzzy Operators......Page 528
18.3.3.2 Decision Rules for M-NN......Page 530
18.3.4.1 Fuzzy Classifier for M-NN......Page 531
18.3.4.2.2 Neural Network NN-2 for Variance Change Magnitude Classification......Page 532
18.4.1 Example 1: Design a Hybrid Chart for Small Process Shifts......Page 534
18.5.2 Performance Comparison for Moving-Window Samples......Page 536
18.6 Final Remarks......Page 537
References......Page 538
19.1 Introduction......Page 541
19.2.1 Rough Set Theory......Page 542
19.2.2 Genetic Algorithms......Page 545
19.3.2.1 The Approach......Page 547
19.3.2.2 Framework of RClass*......Page 548
19.4 Validation of RClass*......Page 550
19.5 Application of RClass* to Manufacturing Diagnosis......Page 552
Defining Terms......Page 556
References......Page 557
Appendix......Page 558