运筹与管理 ›› 2017, Vol. 26 ›› Issue (4): 70-75.DOI: 10.12005/orms.2017.0084

• 理论分析与方法探讨 • 上一篇    下一篇

基于两阶段启发式算法的物流配送选址-路径问题研究

王道平, 徐 展, 杨 岑   

  1. 北京科技大学 东凌经济管理学院, 北京 100083
  • 收稿日期:2013-12-04 出版日期:2017-04-25
  • 作者简介:王道平(1964-),男,北京人,教授,博士生导师,研究方向为供应链管理、知识管理等;徐展(1991-),男,北京人,硕士研究生,研究方向为供应链管理;杨岑(1985-),女,山西人,博士研究生,研究方向为供应链管理、知识管理、系统工程。
  • 基金资助:
    国家自然科学基金资助项目(71172169);中央高校基本科研业务经费资助项目(FRF-BR-16-002B)

Study on Location-routing Problem of Logistics Distribution Based on Two-stage Heuristic Algorithm

WANG Dao-ping, XU Zhan, YANG Cen   

  1. Donglinks School of Economics and Management, Beijing University of Science and Technology, Beijing 100083,China
  • Received:2013-12-04 Online:2017-04-25

摘要: 为了解决配送中心选址与带时间窗的多中心车辆路径优化组合决策问题,利用双层规划法建立了配送中心选址与车辆路径安排的多目标整数规划模型,针对该模型的特点,采用两阶段启发式算法进行了求解。首先,通过基于聚集度的启发式算法对客户进行分类,确定了备选配送中心的服务范围;然后,基于双层规划法,以配送中心选址成本最小作为上层规划目标,以车辆配送成本最小作为下层规划目标,建立了多目标整数规划模型;最后,利用改进的蚁群算法进行了求解。通过分析实例数据和Barreto Benchmark算例的实验结果,验证了该模型的有效性和可行性。

关键词: 运筹学, 选址-路径优化, 双层规划法, 蚁群算法, 客户重要度

Abstract: To solve the combined problem of distribution center location and multi-depot vehicle routing with time windows, using bi-level programming, the multi-objective integer planning model of distribution center location and vehicle routing is formulated. Taking into account the special feature of the model, we propose a suitable solving method. By utilizing two-stage heuristic algorithm, aggregation-based algorithm is firstly used to classify customers, and then, the memberships between customers and distributions center as well as the service area of alternative distribution center are determined. Secondly, based on bi-level programming, the multi-objective integer programming model is established. Its upper planning targets at the minimum cost of distribution center location, while lower planning at the minimum cost of vehicle transportation. Finally, by means of improved ant colony algorithm which introduces the crossover operator, objective values are obtained. Through the analysis of experimental data and Barreto Benchmark instances, we verify the feasibility and validity of the proposed model.

Key words: operational research, location-routing optimization, bi-level programming, ant colony algorithm, customer importance rate

中图分类号: