遗传算法在逆向物流网络规划设计中的应用  

Application of Genetic Algorithm in the Plan and Design of Reverse Logistics Network

作　　者：谢如鹤[1] 邱祝强[1] 

机构地区：[1]中南大学物流工程研究所,长沙410075

出　　处：《武汉理工大学学报（交通科学与工程版）》2007年第5期902-905,共4页Journal of Wuhan University of Technology（Transportation Science & Engineering）

基　　金：国家社科基金项目(批准号:04BJY067);中南大学博士创新选题基金项目(批准号:CSU20030126)资助

摘　　要：考虑到客户、初始回收点和回收中心3个层次的逆向物流网络,选择合适的回收中心,确定初始回收点个数和回收周期,基于一些基本假设,以最小化总的相关成本建立了1个内嵌两个指派模型的非线性混合整数规划模型,对符号体系和求解思路进行了分析.借助遗传算法,结合具体实例,给出了求解步骤和求解结果分析.A simple description of the structure and characteristics of a general multi-level reverse logistics network is presented firstly in this paper. Considering tree basic nodes in reverse logistics network in- cluding customer, initial collecting point and return center, a planning problem is proposed to locate return center, determining the number and periods of the required initial levels collecting points with the objective of minimizing the aggregate cost. Then, to reduce the solving complexity of the problem to some extent, some necessary presumptions are presented, based on which a non linear mixed inte- ger programming model is developed. Simultaneously, two assignment models are embedded. Denota- tion system and existing solving methods are described and discussed respectively. Furthermore, com- bined with a specific instance, a genetic algorithm （GA） is chosen to solve the problem and detailed solving process is listed as well. At last, by Matlab programming, the optimal solution to the instance is obtained and corresponding numerical analysis is carried out.

关 键 词：遗传算法 逆向物流 物流网络 规划设计 非线性整数规划 

分 类 号：F253[经济管理—国民经济]