地球信息科学学报

地球信息科学学报 ›› 2015, Vol. 17 ›› Issue (2): 172-177.doi: 10.3724/SP.J.1047.2015.00172

• • 上一篇    下一篇

一种基于改进蚁群算法与GIS的多约束配送中心选址方法

赵仁辉1,2(), 杨丽娜1,*(), 邵静1,2   

  1. 1. 中国科学院遥感与数字地球研究所,北京 100101
    2. 中国科学院大学,北京 100049
  • 收稿日期:2014-05-05 修回日期:2014-10-30 出版日期:2015-02-10 发布日期:2015-02-10
  • 通讯作者: 杨丽娜 E-mail:zhaorenhui2008@126.com;yangln@radi.ac.cn
  • 作者简介:

    作者简介：赵仁辉(1988-),男,安徽六安人,硕士,主要从事智能空间优化研究。E-mail:zhaorenhui2008@126.com

  • 基金资助:
    国家自然科学基金项目(41201397);国家科技部政策引导项目(2011FU125Z24)

A Method for Multi-constraint Location Decision of Distribution Center Based on Refined Ant Colony Algorithm and GIS

ZHAO Renhui1,2(), YANG Lina1,*(), SHAO Jing1,2   

  1. 1. Institute of Remote Sensing and Digital Earth, CAS, Beijing 100101, China
    2. University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2014-05-05 Revised:2014-10-30 Online:2015-02-10 Published:2015-02-10
  • Contact: YANG Lina E-mail:zhaorenhui2008@126.com;yangln@radi.ac.cn
  • About author:

    *The author: SHEN Jingwei, E-mail:jingweigis@163.com

RichHTML

PDF (PC)

赞

被引次数

5

摘要：

针对单一指派约束和容量约束的设施选址问题（Single Source Capacitated Facility Location Problem, SSCFLP）,建立了一种基于改进蚁群算法与GIS的配送中心选址方法。构建了以总成本费用最小为目标的配送中心选址模型;提出了适合求解SSCFLP问题的改进双层蚁群算法,将求解过程划分为彼此关联的设施选择层和需求指派层2层蚁群,采用改进的全局信息素更新策略加强双层蚁群交流,并对迭代最优解的指派关系进行局部优化;将方法应用于汽车配送中心的选址,利用GIS工具构建选址空间。实验结果表明,该选址方法能找到质量较好的选址及指派结果,对于求解同类问题具有较强的借鉴意义。

关键词: 改进双层蚁群算法, SSCFLP, GIS, 容量约束

Abstract:

Location decision of any logistics distribution center meets multiple constraints, such as the specific spatial environment, the single assignment constraint, the capacity of warehouses and the minimum cost of capital. This paper proposed a model based on refined ant colony algorithm and GIS tools to solve Single Source Capacitated Facility Location Problem (SSCFLP). Firstly, a location selection model was established, which met the target of minimizing the total cost. Secondly, by combining ant colony algorithm and local search, the refined bi-level ant colony optimization to solve the SSCFLP problem was proposed. The solving process was divided into two layers: the layer of choosing facilities and the layer of assigning demands. These two layers were associated with each other. In each iteration, the ants would generate solutions by selecting new sets of facility locations from the candidate sites according to the capacity constraint, and establish the assignment of each customer to a selected facility location using pseudorandom search. The iteration-best solution was optimized and memorized using local search. Then the global optimal solution could be attained through conducting multiple iterations. Finally, a location decision case of the car logistics distribution center in Binhai district was constructed. Site selection space was constructed based on GIS tools, considering demands, candidate sites and shipping cost, and other spatial factors, such as land use, hydrology and terrain. The experimental results revealed that the method was efficient and could find reasonable scheme for determining location and allocation. It had certain academic significance to other similar problems.

Key words: refined bi-level ant colony optimization, SSCFLP, GIS, capacity constraint