Management system simulation, optimization, and application

Author: Chen Zhimin
Publisher:
Publish Date: 2005-06-01
Features: This book is divided into eight chapters. The first chapter elaborates on the scope and steps of research on management system simulation technology, as well as related concepts. It then defines the specific form of the simulation environment optimization problem, and finally provides a comprehensive review of the current development status of simulation optimization problems, along with two examples of simulation optimization. The second chapter primarily discusses the effective implementation of the response surface method in a simulation optimization environment to improve the inefficiency of the existing response surface method in estimating surface parameters through repeated runs. The third chapter discusses the application of stochastic approximation algorithms in simulation environments for optimization. It presents stochastic approximation-based optimization algorithms for unconstrained and constrained simulation optimization problems and provides specific discussions on the selection of relevant parameters in the algorithms, along with our recommendations. The fourth chapter primarily discusses the infinitesimal perturbation analysis (IPA) method in gradient estimation methods. It first elaborates on the basic ideas of infinitesimal perturbation analysis for discrete event dynamic systems (DEDS) and the gradient estimation problem in steady-state simulation using IPA. Building on this, it extends infinitesimal perturbation analysis to the gradient estimation problem in terminal-state simulation and proves the unbiasedness and strong consistency of IPA gradient estimation. The fifth chapter discusses the implementation issues of optimization in simulation environments based on the optimization theories in the second, third, and fourth chapters. The sixth chapter conducts experimental comparisons of three simulation optimization methods: the response surface method for single-run surface parameter estimation, the stochastic approximation process under symmetric difference gradient estimation, and the stochastic approximation process under IPA gradient estimation. The seventh chapter applies system simulation technology to study a concrete supply system with one plant and multiple construction sites. It establishes an SLAM simulation model for the system, approximates the probability distribution of input variables in the simulation model, and provides behavioral metrics for the concrete supply system based on simulation output results. It also offers recommendations for the rational scheduling of concrete mixer trucks. The eighth chapter is a summary of the entire book.