基于模型预测控制的供应链操作优化仿真方案  

Simulation Scheme for Centralized Supply Chain Operation Based on Model Predictive Control Strategy

作　　者：雷蓓蓓[1] 杨春节[1] 曹柬[1] 

机构地区：[1]浙江大学工业控制技术国家实验室,浙江杭州310027

出　　处：《系统仿真学报》2008年第16期4221-4226,共6页Journal of System Simulation

基　　金：国家自然科学基金(60674086)

摘　　要：基于模型预测控制算法为供应链中的操作决策问题建立动态定量的供应链整体系统模型进行研究,结合供应链特点改进目标函数,使得利润最大化同时保证了客户服务最大可能地接近期望水平,提出了一整套供应链滚动操作优化决策方案,为供应链决策者提供一体化的动态决策支持;并进一步针对不确定性市场需求的现实,引入风险值库存管理策略,定量测量和控制需求预测误差带来的库存风险,设计出应用于需求不确定市场环境下供应链滚动操作优化决策方案。基于GAMS的仿真结果证明了算法的有效性。A dynamic quantitative model for the supply chain operation process was established to obtain an integrative decision-making solution schema based on modelpredictive control strategy, and a revision to the optimization objective was made to ensure that the optimal solution minimizes overall costs, while it satisfies a certain customer demand satisfactory （CDS） level; Furthermore, the Value-at-Risk （VaR） inventory management strategy was inducted to measure and control the inventory risk quantitatively, and a dynamic rolling decision-making schema for supply chain operation was proposed under uncertainty of demands. The GAMS simulation results prove the efficiency of the algorithm.

关 键 词：GAMS 模型预测控制 供应链管理 风险值估计库存管理策略(VaR) 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]