MTO再制造系统最优生产计划提前期研究——基于最小相对熵的方法  

Optimal Planned Lead Time In A MTO Remanufacturing System Based On The Minimum Relative Entropy Method

作　　者：包兴[1] 季建华[1] Ou Tang 

机构地区：[1]上海交通大学经济与管理学院,上海200052 [2]瑞典大学生产经济系

出　　处：《工业工程与管理》2007年第2期51-56,64,共7页Industrial Engineering and Management

基　　金：教育部社会科学基金重点项目(02JAZJ79007)

摘　　要：研究了MTO再制造系统关于单一核心部件在分拆和采购提前期均不确定下的生产运作;由于回收部件并非100%可用,使得以最小化再制造成本为目标得到的生产计划提前期呈现双峰分布的形式;本文将最小相对熵(MRE)方法应用于生产计划提前期分布的拟合,仿真算例表明:与传统的正态分布和Pearson近似方法相比,MRE近似的结果更为接近目标服务水平和真实的成本曲线,所得结论对再制造企业安排生产计划、设备投资以及旧产品回收策略等有现实的指导意义。The production process of single-component-reuse remanufacturing system, in which both disassembly and procuring lead time are uncertain, is considered. Due to the component yield rate is not 100%, the distribution of remanufacturing planned lead time, which is obtained via minimizing the remanufacturing cost, takes a bimodality form. The Minimum Relative Entropy （MRE） method is applied to approximate the distribution of planned lead time, which provides better results in approximating target service level and actual cost curve when compared to the traditional Normal and Pearson approximation through numerical examples. And the results are helpful to the production planning, investment and product return policy determination in practice.

关 键 词：MTO再制造系统 生产计划提前期 双峰分布 最小相对熵方法 

分 类 号：TH16[机械工程—机械制造及自动化] O21[理学—概率论与数理统计]