基于改进K-means聚类方法的新零售物流配送路径优化  

Optimization of Logistics Distribution Route in Context of New Retail Based on Improved K-means Clustering Method

作　　者：陈婵丽 钟映竑[1] Chen Chanli;Zhong Yinghong(School of Management, Guangdong University of Technology, Guangzhou 510520, China)

机构地区：[1]广东工业大学管理学院

出　　处：《物流技术》2019年第5期73-78,126,共7页Logistics Technology

摘　　要：在新零售背景下,根据新零售的要求,从末端物流配送路径的角度出发,建立了车辆配送路径最优化数学模型,将遗传算法和K-means聚类算法进行结合与改进,对末端实体店配送方法进行优化。并以广州市天河区实际数据为例,通过Tensorflow软件仿真实验进行验证。实验结果证明,和传统遗传算法相比,基于遗传算法改进的K-means聚类方法在复杂的区域物流内可配送路程,解决了重复配送路径问题,并且优化了物流配送路径,提高了配送效率,从而改善了服务质量和用户体验度,为新零售时代的物流配送提供路径优化方法。In the context of new retail, according to its requirements, and from the point of view of distribution routing in endpoint logistics, we established the optimization mathematical model of the vehicle routing problem, combined and improved the genetic algorithm and K- means clustering algorithm to optimize the means of distribution for the terminal stores. Then based on the actual data of Tianhe district, Guangzhou, we demonstrated the validity of the process through a simulation experiment using the software TensorFlow. The result of the simulation showed that, compared to the traditional genetic algorithm, the GA-improved K-means clustering method could work better in complex regional logistics situation, solve the repeated routing problem, optimize the logistics distribution routes and improve the distribution efficiency, thus improving service quality and user experience in the era of new retail.

关 键 词：新零售 配送路径 遗传算法 K-MEANS聚类算法 路径优化 

分 类 号：F252.1[经济管理—国民经济]