Discrete-Event System Simulation

Transcription

1 Discrete-Event System Simulation FIFTH EDITION Jerry Banks Technolögico de Monterrey, Campus Monterrey John S. Carson II Independent Simulation Consultant Barry L. Nelson Northwestern University David M. Nicol University of Illinois, Urbana-Champaign Upper Saddle River Boston Columbus San Francisco New York Amsterdam Cape Town Dubai London Madrid Milan Munich Paris Montreal Toronto Delhi Mexico City Sao Paulo Sydney Hong Kong Seoul Singapore Taipei Tokyo

2 Contents Preface 11 What's New in the Fifth Edition 15 List of Materials Available on 16 About the Authors 17 I Introduction to Discrete-Event System Simulation 19 1 Introduction to Simulation When Simulation Is the Appropriate Tool When Simulation Is Not Appropriate Advantages and Disadvantages of Simulation Areas of Application Some Recent Applications of Simulation Systems and System Environment Components of a System Discrete and Continuous Systems Model of a System Types of Models Discrete-Event System Simulation Steps in a Simulation Study 34 References 39 Exercises 40 2 Simulation Examples in a Spreadsheet The Basics of Spreadsheet Simulation How to Simulate Randomness 43

3 4 Contents The Random Generators Used in the Examples How to Use the Spreadsheets How to Simulate a Coin Toss How to Simulate a Random Service Time How to Simulate a Random Arrival Time A Framework for Spreadsheet Simulation A Coin Tossing Game Queueing Simulation in a Spreadsheet Waiting Line Models Simulating a Single-Server Queue Simulating a Queue with Two Servers Inventory Simulation in a Spreadsheet Simulating the News Dealer's Problem Simulating an (M,N) Inventory Policy Other Examples of Simulation Simulation of a Reliability Problem Simulation of Hitting a Target Estimating the Distribution of Lead-Time Demand Simulating an Activity Network Summary 94 References 95 Exercises 96 3 General Principles Concepts in Discrete-Event Simulation The Event Scheduling/Time Advance Algorithm WorldViews Manual Simulation Using Event Scheduling List Processing Basic Properties and Operations Performed on Lists Using Arrays for List Processing Using Dynamic Allocation and Linked Lists Advanced Techniques Summary 132 References 133 Exercises Simulation Software History of Simulation Software The Period of Search ( ) The Advent ( ) The Formative Period ( ) The Expansion Period ( ) 138

4 Contents The Period of Consolidation and Regeneration ( ) The Period of Integrated Environments ( ) The Future ( ) Selection of Simulation Software An Example Simulation Simulation in Java Simulation in GPSS Simulation in SSF Simulation Environments AnyLogic Arena AutoMod Enterprise Dynamics ExtendSim Flexsim ProModel SIMUL Experimentation and Statistical-Analysis Tools Common Features Products 170 References 173 Exercises 174 II Mathematical and Statistical Models Statistical Models in Simulation Review of Terminology and Concepts Di screte random variables Continuous random variables Cumulative distribution function Expectation The mode Useful Statistical Models Queueing systems Inventory and supply-chain systems Reliability and maintainability Limited data Other distributions Discrete Distributions Bernoulli trials and the Bernoulli distribution Binomial distribution Geometric and Negative Binomial distributions Poisson distribution 205

5 6 Contents 5.4 Continuous Distributions Uniform distribution Exponential distribution Gamma distribution Erlang distribution Normal distribution Weibull distribution Triangular distribution Lognormal distribution Beta distribution Poisson Process Properties of a Poisson Process Nonstationary Poisson Process Empirical Distributions Summary 236 References 237 Exercises Queueing Models Characteristics of Queueing Systems The Calling Population System Capacity The Arrival Process Queue Behavior and Queue Discipline Service Times and the Service Mechanism Queueing Notation Long-Run Measures of Performance of Queueing Systems Time-Average Number in System L Average Time Spent in System Per Customer w The Conservation Equation: L = Xw Server Utilization Costs in Queueing Problems Steady-State Behavior of Infinite-Population Markovian Models Single-Server Queues with Poisson Arrivals and Unlimited Capacity: M/G/l Multiserver Queue: M/M/c/oo/oo Multiserver Queues with Poisson Arrivals and Limited Capacity: M/M/c/N /00П6 6.5 Steady-State Behavior of Finite-Population Models {M/M/c/K/K) Networks of Queues Rough-cut Modeling: An Illustration Summary 285 References 286 Exercises 286

6 Contents 7 III Random Numbers Random-Number Generation Properties of Random Numbers Generation of Pseudo-Random Numbers Techniques for Generating Random Numbers Linear Congruential Method Combined Linear Congruential Generators Random-Number Streams Tests for Random Numbers Frequency Tests Tests for Autocorrelation Summary 312 References 312 Exercises Random-Variate Generation Inverse-Transform Technique Exponential Distribution Uniform Distribution WeibuII Distribution Triangular Distribution Empirical Continuous Distributions Continuous Distributions without a Closed-Form Inverse Discrete Distributions Acceptance-Rejection Technique Poisson Distribution Nonstationary Poisson Process Gamma Distribution Special Properties Direct Transformation for the Normal and Lognormal Distributions Convolution Method More Special Properties Summary 345 References 345 Exercises 346 IV Analysis of Simulation Data Input Modeling Data Collection Identifying the Distribution with Data Histograms 359

7 Contents Selecting the Family of Distributions Quantile-Quantile Plots Parameter Estimation Preliminary Statistics: Sample Mean and Sample Variance Suggested Estimators Goodness-of-Fit Tests Chi-Square Test Chi-Square Test with Equal Probabilities Kolmogorov-Smirnov Goodness-of-Fit Test p-values and "Best Fits" Fitting a Nonstationary Poisson Process Selecting Input Models without Data Multivariate and Time-Series Input Models Covariance and Correlation Multivariate Input Models Time-Series Input Models The Normal-to-Anything Transformation Summary 394 References 396 Exercises Verification, Calibration, and Validation of Simulation Models Model Building, Verification, and Validation Verification of Simulation Models Calibration and Validation of Models Face Validity Validation of Model Assumptions Validating Input-Output Transformations Input-Output Validation: Using Historical Input Data Input-Output Validation: Using a Turing Test Summary 431 References 432 Exercises Estimation of Absolute Performance Types of Simulations with Respect to Output Analysis Stochastic Nature of Output Data Absolute Measures of Performance and Their Estimation Point Estimation Confidence-Interval Estimation Output Analysis for Terminating Simulations Statistical Background Confidence Intervals with Specified Precision Quantiles 451

8 Contents Estimating Probabilities and Quantiles from Summary Data Output Analysis for Steady-State Simulations Initialization Bias in Steady-State Simulations Error Estimation for Steady-State Simulation Replication Method for Steady-State Simulations Sample Size in Steady-State Simulations Batch Means Method for Steady-State Simulations Steady-State Quantiles Summary 471 References 472 Exercises Estimation of Relative Performance Comparison of Two System Designs Independent Sampling Common Random Numbers (CRN) Confidence Intervals with Specified Precision Comparison of Several System Designs Bonferroni Approach to Multiple Comparisons Selection of the Best Metamodeling Simple Linear Regression Metamodeling and Computer Simulation Optimization via Simulation What Does "Optimization via Simulation" Mean? Why is Optimization via Simulation Difficult? Using Robust Heuristics An Illustration: Random Search Summary 518 References 519 Exercises 520 V Applications Simulation of Manufacturing and Material-Handling Systems Manufacturing and Material-Handling Simulations Models of Manufacturing Systems Models of Material Handling Systems Some Common Material-Handling Equipment Goals and Performance Measures Issues in Manufacturing and Material-Handling Simulations Modeling Downtimes and Failures Trace-Driven Models 537

9 10 Contents 13.4 Case Studies of the Simulation of Manufacturing and Material Handling Manufacturing Example: An Assembly-line Simulation System Description and Model Assumptions Presimulation Analysis Simulation Model and Analysis of the Designed System Analysis of Station Utilization Analysis of Potential System Improvements Concluding Words: The Gizmo Assembly-Line Simulation Summary 548 References 548 Exercises Simulation of Networked Computer Systems Introduction Simulation Tools Process Orientation Event Orientation Model Input Modulated Poisson Process Poisson-Pareto Process Pareto-length Phase Time WWW Traffic Mobility Models in Wireless Systems The OSI Stack Model Physical Layer in Wireless Systems Propagation Models Determining the Receivers Media Access Control Token-Passing Protocols Ethernet Data Link Layer TCP Model Construction Construction DML Example Summary 605 References 606 Exercises 607 Appendix 609 Index 625