仿真优化技术在一类CAS问题不确定性建模中的应用

An Application of Simulation-based Optimization Technique in Modeling of a Kind of CAS Uncertainty

供应链是一类CAS(复杂自适应系统),减少或者消除其中的不确定性是提高其性能的重要策略之一。首先分析不确定环境下供应链模型的特点,然后对其仿真优化技术进行剖析,最后在FuzzySCSim(基于Arena软件开发的专用仿真平台)的基础上,应用该技术对不确定性环境下供应链配置进行优化分析。在试验中,提出采用综合SS(散点搜索算法)与TS(禁忌搜索算法)两者特点的混合搜索算法,并以一个复合指标GMTICSL来衡量供应链的整体性能。结果表明该模型能够较好的描述这些不确定性的产生机理,并且为减少或者消除这些不确定性提供了有效的分析平台。

Reduction or even elimination of supply chain uncertainties is one of important strategies to improve the (performance) of the chain. In order to understand supply chain behavior and improve supply chain performance, one (simulation-based) optimization technique is developed in this paper. Under such uncertain environment, the simulation-based optimization technique is used to find optimal configuration of supply chain structure. In the experiment based on (FuzzySCSim) simulation platform(one special simulation tools developed by Arena simulation package), a hybrid approach (combing) scatter search and taboo search is used and a new performance measure(GMTICSL) is proposed. The result shows the method could better describe and analyze supply chain uncertainty and support relative decisions.