并行的嵌套式遗传算法在库存路径问题中的应用  

Application of parallel nested genetic algorithm in centralized inventory routing problem

作　　者：张凯清[1] 周永务[2] 

机构地区：[1]合肥工业大学数学学院,安徽合肥230009 [2]合肥工业大学管理学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2009年第7期1020-1024,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(70771034);新世纪优秀人才支持计划资助项目(NCET-05-0557);高等学校全国优秀博士学位论文作者2005年专项资金资助项目(200565);高等学校博士学科点专项科研基金资助项目(20060359007)

摘　　要：文章考虑了一个供应商和其多个分销子公司组成的2级供应链分配系统,商品通过有容量限制单一型号车辆从供应商到各个分销子公司,利用约数策略和遗传算法,达到"智能分区"的效果;在路径处理上,采取分担运输和直运相结合的策略,解决了订货量大于车载容量时面临的难题;在需求不变(供求不变)的情况下,给出几个成本因素对算法进行有效性分析,得出相应结论,并对并行和串行算法进行比较和优劣分析。This paper considers a two-echelon supply chain distribution system consisting of a supplier and its multiple retailers. The goods are transported to each affiliate by capacity-limited single-style vehicle. An intelligent-partition effect is reached by the divisor-policy and genetic algorithm When searching the routings, a combination policy of direct transport and split transport is used to solve the problem of the order quantity more than the loading capacity of the vehicle. In the circumstance of a fixed demand, several cost factors are given to analyze the effectiveness of the algorithm, and then corresponding results are obtained. The parallel algorithm and the serial algorithm are compared and their performance are analyzed.

关 键 词：库存路径问题 约数策略 并行的嵌套式遗传算法 C-W节约算法 

分 类 号：TP301.5[自动化与计算机技术—计算机系统结构] F502[自动化与计算机技术—计算机科学与技术]