Abstract

We review techniques for optimizing stochastic discrete-event systems via simulation. We discuss both the discrete parameter case and the continuous parameter case, but concentrate on the latter which has dominated most of the recent research in the area. For the discrete parameter case, we focus on the techniques for optimization from a finite set: multiple-comparison procedures and ranking-and-selection procedures. For the continuous parameter case, we focus on gradient-based methods, including perturbation analysis, the likelihood ratio method, and frequency domain experimentation. For illustrative purposes, we compare and contrast the implementation of the techniques for some simple discrete-event systems such as the (s, S) inventory system and the GI/G/1 queue. Finally, we speculate on future directions for the field, particularly in the context of the rapid advances being made in parallel computing.

Citations

632	Parallel discrete event simulation - Fujimoto - 1990
420	A stochastic approximation method - Robbins, Monro - 1951
295	Empirical Model-Building and Response Surfaces - Box, Draper - 1987
269	Adaptive algorithms and stochastic approximations - Benveniste, M¶etivier, et al. - 1990
223	Stochastic approximation methods for constrained and unconstrained systems - Kushner, Clark - 1978
124	Discrete Event Systems: Sensitivity Analysis and Stochastic Optimization by the Score Function Method - Rubinstein, Shapiro - 1993
115	Gradient estimation via perturbation analysis - Glasserman - 1991
102	Discrete-Event System Simulation - Banks, Carson, et al. - 2009
102	Stochastic estimation of the maximum of a regression function - Kiefer, Wolfowitz - 1952
95	Analysis of recursive stochastic algorithms - Ljung - 1977
85	Statistical Tools for Simulation Practitioners - Kleijnen - 1987
58	Simulation Modeling and analysis (2nd edition - Law, Kelton - 1991
51	M A Single-sample multiple Decision Procedure for Selecting the Multinomial Event which has the Highest Probability - Elmaghraby, Morse - 1959
49	Likelihood ratio gradient estimation for stochastic systems - Glynn - 1990
41	Sensitivity analysis for simulations via likelihood ratios - Reiman, Weiss - 1989
40	Likelihood ratio gradient estimation: an overview - Glynn - 1987
30	Perturbation Analysis: The State of the Art and Research Issues Explained via the GI/G/l Queue - Suri - 1989
28	Stochastic approximation for Monte Carlo optimization - Glynn - 1986
28	Stochastic Approximation - Wasan - 1969
25	An overview of derivative estimation - L’Ecuyer
25	Sensitivity analysis via likelihood ratios - Reiman, Weiss - 1986
24	Schruben, “Techniques for simulation response optimization - Jacobson, W - 1989
23	A gradient technique for general buffer storage design in a production line - Ho, Eyler, et al. - 1979
23	Stochastic optimization by simulation: Convergence proofs for the GI/G/1 queue in steady state - L'Ecuyer, Glynn - 1994
22	Monte Carlo Optimization, Simulation and Sensitivity of Queueing Networks - Rubinstein - 1986
22	Massively parallel and distributed simulation of a class of discrete event systems: A different perspective - Vakili - 1992
21	Accelerated stochastic approximation - Kesten - 1958
20	Asymptotic distribution of stochastic approximation ~ procedures - Sacks - 1958
19	Optimization of stochastic systems via simulation - Glynn - 1989
18	Convergence of a stochastic approximation algorithm for the GI/G/1 queue using infinitesimal perturbation analysis - Fu - 1990
18	Some guidelines and guarantees for common random numbers - Glasserman, Yao - 1992
17	On a Stochastic Approximation Method - CHUNG - 1954
17	Discrete Event Dynamic Systems and Perturbation Analysis - Ho, Cao - 1991
17	A new method of stochastic approximation type - Polyak - 1990
17	Perturbation analysis gives strongly consistent sensitivity estimates for the M/G/1 queues - Suri, Zazanis - 1988
17	Using a standard clock technique for efficient simulation - Vakili - 1991
16	On the pathwise computation of derivatives with respect to the rate of a point process: The phantom RPA method Queueing Systems Theory Appl. 10, 249-270. Charles Bordenave and Giovanni Luca Torrisi - Brémaud, Vazquez-Abad - 1992
16	Perturbation analysis for the design of flexible manufacturing system flow controllers - Caramanis, Liberopoulos - 1992
16	Optimization of stochastic systems - Glynn - 1986
16	Analysis of Parallel, Replicated Simulations under a Completion Time Constraint - Glynn, Heidelberger - 1991
15	Stochastic approximations and nonlinear regression - Albert, Gardner - 1967
15	Stochastic optimization by simulation: Numerical experiments with the M/M/1 queue in steady state - L'Ecuyer, Giroux, et al. - 1994
15	Response surface methodology: 1966–1988 - Myers, Khuri, et al. - 1989
14	Sensitivity analysis and performance extrapolation for computer simulation models - Rubinstein - 1989
13	Almost sure approximations to the Robbins-Monro and Kiefer-Wolfowitz processes with dependent noise - Ruppert - 1982
13	An experimental procedure for simulation response surface model identification - Schruben, Cogliano - 1987
12	A unified view of the - L’Ecuyer - 1990
11	Extensions and generalizations of smoothed perturbation analysis in a generalized semi-markov process framework - Fu, Hu - 1992
11	Special issue on Dynamics of Discrete Event Systems - Ho, editor - 1989
11	Response Surfaces - Khuri, Cornell - 1987