Operations Research

An Introduction
Ninth Edition

Hamdy A. Taha
University of Arkansas, Fayettevilie

Boston Columbus Indianapolis New York San Francisco Upper Saddle River
Amsterdam Cape Town Dubai London Madrid Milan Munich Paris Montreal Toronto
Delhi Mexico City Sao Paulo Sydney Hong Kong Seoul Singapore Taipei Tokyo
Contents

What's Mew in t h e Ninth Edition 25

Acknowledgments 27
A b o u t t h e Author 29
Trademarks 31
Chapter 1 W h a t Is Operations Research? 35
1.1 Introduction 35
1.2 Operations Research Models 35
.3 Solving the OR Model 39
.4 Queuing and Simulation Models 40
.5 Art of Modeling 40
.6 More Than Just Mathematics 41
.7 Phases of an OR Study 43
.8 . About This Book 44
Bibliography 45
Chapter 2 Modeling w i t h Linear Programming 47
2.1 Two-Variable LP Model 47
2.2 Graphical LP Solution 50
2.2.1 Solution of a Maximization Model 50
2.2.2 Solution of a Minimization Model 58
2.3 Computer Solution with Solver and AMPL 61
2.3.1 LP Solution with Excel Solver 61
' 2.3.2 LP Solution with AMPL 65
2.4 Linear Programming Applications 69
2.4.1 Investment 69
2.4.2 Production Planning and Inventory Control 74
2.4.3 Manpower Planning 82
2.4.4 Urban Development Planning 86
2.4.5 Blending and Refining 91
2.4.6 Additional LP Applications 97
Bibliography 102
Chapter 3 T h e Simplex M e t h o d and Sensitivity Analysis 103
3.1 LP Model in Equation Form 103
3.2 Transition from Graphical to Algebraic Solution 706
8 Contents

3.3 The Simplex Method 110

3.3.1 Iterative Nature of the Simplex Method 111
3.3.2 Computational details of the Simplex algorithm 113
3.3.3 Summary of the Simplex Method 119
3.4 Artificial Starting Solution 123
3.4.1 M-Method 123
3.4.2 Two-Phase Method 128
3.5 Special Cases in the Simplex Method 133
3.5.1 Degeneracy 133
3.5.2 Alternative Optima 136
3.5.3 Unbounded Solution 138
3.5.4 Infeasible Solution 740
3.6 Sensitivity Analysis 742
3.6.1 Graphical Sensitivity Analysis 142
3.6.2 Algebraic Sensitivity Analysis—Changes in the
Right-hand Side 148
3.6.3 Algebraic Sensitivity Analysis—Objective
function 757
3.6.4 Sensitivity Analysis with Tora, Solver,
and Ampl 163.
3.7 Computational Issues in Linear Programming 765
Bibliography 770

Chapter 4 Duality and Post-Optimal Analysis 171

4.1 Definition of the Dual Problem 777
4.2 Primal-Dual Relationships 775
4.2.1 Review of Simple Matrix Operations 775
4.2.2 Simplex Tableau Layout 776
4.2.3 } Optimal Dual Solution 777
4.2.4 SimplexTableau Computations 784
4.3 Economic Interpretation of Duality 7S7
4.3.1 Economic Interpretation of Dual Variables 788
4.3.2 Economic Interpretation of Dual Constraints 790
4.4 Additional Simplex Algorithms 792
4.4.1 Dual Simplex Algorithm 793
4.4.2 Generalized Simplex Algorithm 798
4.5 Post-optimal Analysis 799
4.5.1 Changes Affecting Feasibility 200
4.5.2 Changes Affecting Optimally 205
Bibliography 208
 Contents 9

Chapter 5 Transportation M o d e l and Its Variants 209

5.1 Definition of the Transportation Model 209
5.2 Nontraditional Transportation Models 276
5.3 The Transportation Algorithm 227
5.3.1 Determination of the Starting Solution 222
5.3.2 Iterative Computations of the Transportation
Algorithm 225
5.3.3 Simplex Method Explanation of the Method of
Multipliers 233
5.4 The Assignment Model 234
5.4.1 The Hungarian Method 235
5.4.2 Simplex Explanation of the Hungarian Method 240
Bibliography 242

Chapter 6 Network Models 243

6.1 Scope and Definition of Network
Models 243
6.2 Minimal Spanning Tree Algorithm 246
6.3 Shortest-Route Problem 257
6.3.1 Examples of the Shortest-Route Applications 251
6.3.2 Shortest-Route Algorithms 255
6.3.3 Linear Programming Formulation of the Shortest-Route
Problem 264
6.4 Maximal Flow Model 268
6.4.1 Enumeration of Cuts 269
6.4.2 Maximal Flow Algorithm 270
6.4.3 Linear Programming Formulation of Maximal Flow
Mode 278
6.5 CPM and PERT 287
6.5.1, Network Representation 287
6.5.2 Critical Path Method (CPM) Computations 286
6.5.3 Construction of the Time Schedule 289
6.5.4 Linear Programming Formulation of CPM 295
6.5.5 PERT Networks 296
Bibliography 299

Chapter 7 Advanced Linear Programming 3O1

7.1 Simplex Method Fundamentals 307
7.1.1 From Extreme Points to Basic Solutions 303
7.1.2 Generalized Simplex Tableau in Matrix Form 306
10 Contents

7.2 Revised Simplex Method 309

7.2.1 Development of the Optimality and Feasibility
conditions 309
7.2.2 Revised Simplex Algorithm 372
7.3 Bounded-Variables Algorithm 377
7.4 Duality 324
7.4.1 Matrix Definition of the Dual Problem 324
7.4.2 Optimal Dual Solution 324
7.5 Parametric Linear Programming 328
7.5.1 Parametric Changes in C 329
7.5.2 Parametric Changes in b 337
7.6 More Linear Programming Topics 334
Bibliography 334

Chapter 8 Goal Programming 335

8.1 A Goal Programming Formulation 335
8.2 Goal Programming Algorithms 340
8.2.1 The Weights Method 340
8.2.2 The Preemptive Method 342
Bibliography 348

Chapter 9 Integer Linear Programming 349

9.1 Illustrative Applications 349
9.1.1 Capital Budgeting 350
9.1.2 Set-Covering Problem 354
9.1.3 Fixed-Charge Problem 359
9.1.4 Either-Or and If-Then Constraints 364
9.2 Integer Programming Algorithms 369
9.2.1 Branch-and-Bound (B&B) Algorithm 370
9.2.2 Cutting-Plane Algorithm 378
Bibliography 383

Chapter 1O Heuristic Programming 385

10.1 Introduction 385
10.2 Greedy (Local Search) Heuristics 386
10.2.1 Discrete Variable Heuristic 386
10.2.2 Continuous Variable Heuristic 388
10.3 Metaheuristic 397
10.3.1 Tabu Search Algorithm 392
10.3.2 Simulated Annealing Algorithm 399
10.3.3 Genetic Algorithm 405
 Contents 11

10.4 Application of Metaheuristics to Integer Linear

Programs 470
10.4.1 ILP Tabu Algorithm 472
10.4.2 ILP Simulated Annealing Algorithm 476
10.4.3 ILP Genetic Algorithm 420
10.5 Introduction to Constraint Programming (CP) 425
Bibliography 425

Chapter 11 Traveling Salesperson Problem (TSP) 429

11.1 Example Applications of TSP 429
11.2 TSP Mathematical Model 437
11.3 Exact TSP Algorithms 447
11.3.1 B&B Algorithm 447
11.3.2 Cutting-Plane Algorithm 444
11.4 Local Search Heuristics 446
11.4.1 Nearest-Neighbor Heuristic 447
11.4.2 Reversal Heuristic 447
11.5 Metaheuristic 450
11.5.1 TSP Tabu Algorithm 450
11.5.2 TSP Simulated Annealing Algorithm 454
11.5.3 TSP Genetic Algorithm 457
Bibliography 467

Chapter 12 Deterministic Dynamic Programming 463

12.1 Recursive Nature of Dynamic Programming (DP)
, Computations 463
12.2 Forward and Backward Recursion 467
12.3 Selected DP Applications 468
, 12.3.1 Knapsack/Fly-Away Kit/Cargo-Loading Model 469
12.3.2 Workforce Size Model 477
12.3.3 Equipment Replacement Model 480
12.3.4 Investment Model 483
12.3.5 Inventory Models 487
12.4 Problem of Dimensionality 487
Bibliography 490

Chapter 13 Deterministic Inventory Models 491

13.1 General Inventory Model 497
13.2 Role of Demand in the Development of Inventory
Models 492
12 Contents

13.3 Static Economic-Order-Quantity (EOQ) Models 494

13.3.1 Classical EOQ Model 494
13.3.2 EOQ with Price Breaks 499
13.3.3 Multi-Item EOQ with Storage Limitation 503
13.4 Dynamic EOQ Models 505
13.4.1 No-Setup EOQ Model 507
13.4.2 Setup EOQ Model 570
Bibliography 527

Chapter 14 Review of Basic Probability 523

14.1 Laws of Probability 523
14.1.1 Addition Law of Probability 524
14.1.2 Conditional Law of Probability 525
14.2 Random Variables and Probability Distributions 526
14.3 Expectation of a Random Variable 529
14.3.1 Mean and Variance (Standard Deviation)
of a Random Variable 530
14.3.2 Joint Random Variables 537
14.4 Four Common Probability Distributions 534
14.4.1 Binomial Distribution 535
14.4.2 Poisson Distribution 535
14.4.3 Negative Exponential Distribution 537
14.4.4 Normal Distribution 538
14.5 Empirical Distributions 540
Bibliography 546

Chapter 15 Decision Analysis and Games 547

15.1 Decision Making under Certainty—Analytic Hierarchy
Process (AHP) 547
15.2 Decision Making under Risk 557
15.2.1 Decision Tree-Based Expected Value
Criterion 557
15.2.2 Variants of the Expected Value
Criterion 563
15.3 Decision under Uncertainty 577
15.4 Game Theory 575
15.4.1 Optimal Solution of Two-Person
Zero-Sum Games 576
15.4.2 Solution of Mixed Strategy Games 579
Bibliography 585
 Contents 13

Chapter 16 Probabilistic Inventory Models 587

16.1 Continuous Review Models 587
16.1.1 "Probabilitized" EOQ Model 587
16.1.2 Probabilistic EOQ Model 590
16.2 Single-Period Models 594
16.2.1 No-Setup Model (Newsvendor Model) 594
16.2.2 Setup Model (s-S Policy) 598
16.3 Multiperiod Model 607
Bibliography 603

Chapter 17 M a r k o v Chains 605

17.1 Definition of a Markov Chain 605
17.2 Absolute and n-Step Transition
Probabilities 608
17.3 Classification of the States in a Markov Chain 670
17.4 Steady-State Probabilities and Mean Return Times
of Ergodic Chains 672
17.5 First Passage Time 677
17.6 Analysis of Absorbing States 627
Bibliography 626

Chapter 18 Queuing Systems 627

18.1 Why Study Queues? 627
18.2 Elements of a Queuing Model 629
18.3 Role of Exponential Distribution 630
/ 18.4 Pure Birth and Death Models (Relationship Between
the Exponential and Poisspn Distributions) 634
18.4.1 Pure Birth Model 634
' 18.4.2 Pure Death Model 638
18.5 General Poisson Queuing Model 640
18.6 Specialized Poisson Queues 645
18.6.1 Steady-State Measures of Performance 646
18.6.2 Single-Server Models 650
18.6.3 Multiple-Server Models 657
18.6.4 Machine Servicing Model—(M/M/R):
(GD/K/K), R < K 667
18.7 (M/G/1):(GD/oo/oo)—Pollaczek-Khintchine
(P-K) Formula 670
18.8 Other Queuing Models 672
14 Contents

18.9 Queuing Decision Models 672

18.9.1 Cost Models 673
18.9.2 Aspiration Level Model 677
Bibliography 679

Chapter 19 Simulation Modeling 681

19.1 Monte Carlo Simulation 687
19.2 Types of Simulation 686
19.3 Elements of Discrete Event Simulation 687
19.3.1 Generic Definition of Events 687
19.3.2 Sampling from Probability Distributions 688
19.4 Generation of Random Numbers 695
19.5 Mechanics of Discrete Simulation 697
19.5.1 Manual Simulation of a Single-Server Model 697
19.5.2 Spreadsheet-Based Simulation of the Single-Server
Model 703
19.6 Methods for Gathering Statistical Observations 704
19.6.1 Subinterval Method 705
19.6.2 Replication Method 707
19.7 Simulation Languages 708
Bibliography 770

Chapter 2 0 Classical Optimization Theory 711

20.1 Unconstrained Problems 777
20.1.1 Necessary and Sufficient Conditions 772
20.1.2 The Newton-Raphson Method 775
20.2 Constrained Problems 777
20.2.1 Equality Constraints 777
20.2.2 Inequality Constraints—Karush-Kuhn-Tucker (KKT)
Conditions 727
Bibliography 732

Chapter 21 Nonlinear Programming Algorithms 733

21.1 Unconstrained Algorithms 733
21.1.1 Direct Search Method 733
21.1.2 Gradient Method 737
21.2 Constrained Algorithms 740
21.2.1 Separable Programming 747
21.2.2 Quadratic Programming 749
21.2.3 Chance-Constrained Programming 754
 Contents 15

21.2.4 Linear Combinations Method 758

21.2.5 SUMT Algorithm 760
Bibliography 767

Appendix A Statistical Tables 763

Appendix B Partial Answers to Selected Problems 767

Index 813