单亲遗传算法的冷链物流车辆路径问题（VRP）优化研究

发布时间：2018-03-11 12:01

本文选题：冷链物流　切入点：多配送中心　出处：《燕山大学》2016年硕士论文　论文类型：学位论文

【摘要】：在互联网的带动下,农产品冷链物流需求越来越大,而成本一直是制约冷链物流发展的关键因素,如何有效降低冷链物流成本成为国家、企业和消费者共同关注的热点话题。与此同时,冷链物流市场的扩大以及消费者对服务品质的要求提升,促使企业不得不提升自身服务质量,提高满意度来增加顾客黏性,在竞争激烈的市场中占据一席之地。因此,本文的研究旨在通过合理的车辆调度和路径优化,在保证满意度处于高水平的情况下,使综合成本最低,为冷链物流企业的日常调度工作提供依据。论文突破了仅考虑运输成本及单配送中心来优化冷链物流路径的局限性,克服了传统遗传算法在求解VRP(Vehicle Routing Problem)问题的不足,提出了基于时间及品质因素的满意度函数来满足企业的现实需求,因而具有重要的理论和实际意义。首先,综述了国内外在冷链物流路径优化方面的研究成果,指出了当前研究中的不足,同时阐述了冷链物流路径问题相关的基础理论,为论文的后续研究奠定了坚实的理论基础。其次,从VRP基本模型出发,结合冷链物流特点,在考虑货损成本和能耗成本的多成本目标优化前提下,分别构建了满意度约束的冷链VRP模型和具有多配送中心的冷链VRP模型。再次,为了避免传统遗传算法对VRP问题染色体的破坏性以及“早熟收敛”问题,提出了单亲遗传算法的局部精英选择策略,并设计了基于染色体基因段的三种变异算子和交叉算子,其中在交叉算子中提出概率融合操作以优化调度车辆数。设计自适应遗传算法作为对比实验,运用MATLAB软件实现编程。最后,以S食品公司数据为例,验证了本文构建的两个模型和单亲遗传算法的有效性和优越性。
[Abstract]:Driven by the Internet, the demand for cold chain logistics of agricultural products is increasing, and cost has been the key factor restricting the development of cold chain logistics. How to effectively reduce the cost of cold chain logistics has become a country. At the same time, the expansion of the cold-chain logistics market and the improvement of consumers' demand for service quality have prompted enterprises to enhance their own service quality and satisfaction to increase customer stickiness. Therefore, through reasonable vehicle scheduling and route optimization, the purpose of this paper is to ensure that the overall cost is the lowest under the condition that satisfaction is at a high level. This paper breaks through the limitation of only considering transportation cost and single distribution center to optimize cold chain logistics path, and overcomes the deficiency of traditional genetic algorithm in solving VRP(Vehicle Routing problem. This paper puts forward the satisfaction function based on time and quality factors to meet the actual needs of enterprises, so it has important theoretical and practical significance. Firstly, this paper summarizes the research results of cold chain logistics path optimization at home and abroad. The deficiency of current research is pointed out, and the basic theory of cold chain logistics routing problem is expounded, which lays a solid theoretical foundation for further research. Secondly, starting from the basic model of VRP, combined with the characteristics of cold chain logistics, On the premise of multi-cost optimization, the cold-chain VRP model with satisfaction constraint and the cold-chain VRP model with multi-distribution center are constructed respectively. In order to avoid the damage of the traditional genetic algorithm to the chromosome of VRP problem and the problem of "premature convergence", the local elite selection strategy of the single parent genetic algorithm is proposed, and three mutation operators and crossover operators based on chromosome gene segment are designed. In the crossover operator, the probability fusion operation is put forward to optimize the number of vehicles. The adaptive genetic algorithm is designed as a contrast experiment, and the program is implemented by using MATLAB software. Finally, taking the data of S Food Company as an example, The validity and superiority of the two models and the parthenogenic genetic algorithm are verified.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U116.2

【相似文献】