【摘要】 介绍了仿真优化的特点与难点,阐述了遗传算法作为仿真优化算法的优点。为了解决随机性系统的仿真优化问题,提出一种改进的遗传算法并基于它设计了一个仿真优化包。针对系统的随机特性,改进的遗传算法提出了一种新的K精英选择算子并设计了一个候选解收集器,上述改进能有效保护遗传进化过程中出现的优良个体并加快遗传算法的收敛速度。使用统一建模语言(Un ified M ode ling Language,UML)活动图来说明仿真优化软件包的原理并给出了软件包的UML类图,面向对象方法和设计模式技术的应用使仿真优化包具有较好的通用性和可扩展性。库存仿真优化实例表明所设计的仿真优化包是可行且有效的。更多还原

【Abstract】 The features and hardness of simulation optimization are introduced and the advantages of genetic algorithm are expounded when it is used as simulation optimization algorithm in the paper.An improved genetic algorithm is proposed to solve the simulation optimization problem of the stochastic system and a simulation optimization package is designed based on the improved genetic algorithm.A new K elites selection operation and a candidate solutions collector are proposed due to the stochastic nature of the system.Above improvement can protect elite individuals emerging in the process of evolution and then get better performance.UML activity diagram is used to explain the principle of the simulation optimization package and the UML class diagram of package is presented.Simulation optimization package becames more universal and easier to extend since Object-Oriented and design pattern techniques are adopted.A simulation optimization case of inventory system shows that the simulation optimization package designed in this paper is feasible and efficient.更多还原