一种基于Voronoi图求解车辆路径问题混合启发式算法.docVIP

一种基于Voronoi图求解车辆路径问题的混合启发式算法 　　摘 要:针对由多个配送中心和多个客户点组成的物流网络中的车辆路径问题,提出了一种基于“集群第一,路线第二”的路径优化策略,即首先使用Voronoi分割对配送区域进行划分,然后引入综合插入算法和变邻域搜索算法的混合启发式算法求解配送区域内车辆路径问题。通过算例和应用系统的分析与验证表明,该混合算法既能获取质量较优解,同时也具有较好的实时性,能较好地满足实际应用需求。 　　关键词:Voronoi分割; 混合启发式算法; 插入式算法; 变邻域搜索; 邻接信息 　　中图分类号:TP301.6 　　文献标志码:A 　　 　　文章编号:1001-3695(2010)02-0515-04 　　doi:10.3969/j.issn.1001-3695.2010.02.031 　　 　　Hybrid heuristic algorithm using Voronoi diagram forvehicle routing problem 　　 　　ZHANG Zhi-jun??1??, LI-Feng??1 ??, CAO Bu-yang??2?? 　　 　　(1.School of Computer Science Telecommunication Engineering, Jiangsu University, Zhenjiang Jiangsu 212013,China; 2.School of Software Engineering, Tongji University, Shanghai 201804, China) 　　 　　Abstract:This paper proposed the optimization strategy of vehicle route based on the strategy of “cluster first, route second” which aimed at the vehicle route problems made up of multiple dispatching centers and sale-points in the logistics network. Firstly, used Voronoi tessellation to divide the dispatching regions, and then introduced a hybrid heuristic algorithm which combined the plug-in algorithm and the variable neighbor search(VNS) algorithm in order to solve the optimization problems of vehicle route in dispatching regions. This hybrid algorithm can achieve a better solution to vehicle route optimization problems through the experiment results and the analysis and verification of the application system. At the same time, it also has good real-time which can better meet the needs of practical application. 　　Key words:Voronoi tessellation; hybrid heuristic algorithm; CI; VNS; adjacency information 　　 　　0 引言 　　车辆路径问题(vehicle routing problem,VRP)一直是运筹学、管理学、计算机科学等领域的研究热点问题,在现实生活中有着广泛的应用,是一个具有重要理论意义和实际应用价值的研究课题。车辆路径问题一般定义为:对一系列客户,组织适当的行车路线,使车辆有序地通过,在满足一定的约束条件(如货物需求量、发货量、交货时间、车辆容量限制、行驶里程限制、时间限制等)下,在客户可以接受的演算时间内,给出尽可能优化的路线规划方案,最大限度地降低客户的运营成本(运营时间、距离或其他费用)。该问题1959年由Dantzig等人[1]提出后,几十年来已取得大量研究成果。由于VRP是NP-hard问题,难以用精确算法求解,对此类问题求解方法的研究主要集中于能在较短时间内给出较优解的启发式算法