面向第4方物流的多代理人作业整合优化算法  

Optimization algorithm for multi-agent job integration for fourth-party-oriented logistics

作　　者：王勇[1] 吴志勇[1] 陈修素[2] 吴金奇[1] 

机构地区：[1]重庆大学经济与工商管理学院,重庆400044 [2]重庆工商大学数学与统计学院,重庆400067

出　　处：《管理科学学报》2009年第2期105-116,共12页Journal of Management Sciences in China

基　　金：国家自然科学基金资助项目(70572028,70872123)

摘　　要：物流作业整合是现代物流方法中减少物流成本的主要手段.在综合考虑代理商选择和线路优化两问题的基础上,建立了基于图状结构的面向第4方物流的多代理人作业整合优化模型.为了避免将代理商选择和线路优化作为两个互相分离的子问题来处理,提出了求解它的两层邻域搜索算法.第1层采用了转移、交换和环形移动3种移动策略求解作业在代理商之间的分配,而在第2层提出了路合并、路生成替换策略,形成了适于该问题的优化算法.随机产生20例算例,将两层邻域搜索算法的计算结果与基于k-最短路的枚举算法的计算结果进行比较,说明了该算法的可行性和有效性.The integration of logistics job is a main method to reduces the logistics cost. In this paper, we con- sider the agent and road choice problems comprehensively, and then set up an multi-agent optimization model for job integration based on graph structure and fourth-party-oriented logistics. We propose a two layers neigh- borhood search algorithm to avoid the separation between agent choices and road optimization. In the first lay- er, shift move and swap move and loop move are used to assign jobs to agents; in the second layer, the line merging strategy, line generation and replacing strategy are used to form a suitable optimization algorithm for this problem. Twenty examples are generated randomly; to examine the feasibility and validity of the two-lay- ers search algorithm, whose results are compared to that of enumerating algorithm based on the k-shortest road.

关 键 词：第4方物流 多代理人 作业整合 邻域搜索算法 

分 类 号：F224.3[经济管理—国民经济] O22[理学—运筹学与控制论]