• 论文 •    

蚁群算法及其在有硬时间窗的车辆路径问题中的应用

刘志硕,柴跃廷,申金升   

  1. 1.清华大学 自动化系,北京  100084;2.北京交通大学 交通运输学院,北京  100044
  • 收稿日期:2005-02-27 修回日期:2005-03-23 出版日期:2006-04-15 发布日期:2006-04-25
  • 基金资助:
    国家863/CIMS主题资助项目(2001AA414230)。

Ant colony algorithm for vehicle routing problem with hard time window

LIU Zhi-shuo,CHAI Yue-ting,SHEN Jin-sheng   

  1. 1.Dep. of Automation, Tsinghua Univ., Beijing  100084, China; 2.Sch. of Traffic & Transportation, Beijing Jiaotong Univ., Beijing  100044, China
  • Received:2005-02-27 Revised:2005-03-23 Online:2006-04-15 Published:2006-04-25
  • Supported by:
    Project supported by the National High-Tech.R&D Program for CIMS, China (No.2001AA414230).

摘要: 为求解有硬时间窗的车辆路径问题,提出了一种基于可行解两阶段构造策略的自适应混合蚁群算法。在第一阶段,用蚂蚁的局部遍历代替传统的全局遍历,每个蚂蚁采用蚁群算法进行局部遍历,构造一个回路,蚂蚁转移采用回路两阶段构路策略;在第二阶段,由前一阶段所构造的回路通过采用近似解可行化策略来组合形成可行解。此外,为提高算法的寻优能力,在转移规则中引入了基于时间窗的紧迫性因子和匹配度因子,并与节约算法和爬山法有机结合。实验结果表明,自适应混和蚁群算法性能优良,能够有效地求解有硬时间窗的车辆路径问题。

关键词: 有硬时间窗的车辆路径问题, 自适应混合蚁群算法, 可行解两阶段构造策略, 回路两阶段构造策略, 紧迫性因子, 匹配度因子

Abstract: An Adaptive Hybrid Ant Colony Algorithm (AHACA) based on the two-phase construction procedure of feasible solution was established to solve the Vehicle Routing Problem with Hard Time Window (VRPHTW). There were two phases in the construction of feasible solutions of VRPHTW. In the first phase, ant colony algorithm was applied to construct these sub-tours and the two-phase sub-tour construction procedure was used in its state transition rule. In the second phase, the sub-tours were composed of feasible solutions by means of the feasibility process of approximate solution which could increase the number of optimal feasible solutions in larger scale. Moreover, to increase the ability of optimization and avoid the stagnation phenomenon, some approaches were introduced into the algorithm such as time-window-based urgency degree factor and match degree factor, C-W saving algorithm and 2-opt approach. Experimental results showed that the performance of AHACA was comparable with other meta-heuristic algorithms and could be used in practical VRPHTW with sufficient reliability and speed.

Key words: vehicle routing problem with hard time window, adaptive hybrid ant colony algorithm, two-phase feasible solution construction procedure, two-phase sub-tour construction procedure, urgency degree factor, match degree factor

中图分类号: