Research On Logistics Distribution Route Optimization Under The Background Of New Retail

Since 2016,the concept of new retail has become more and more familiar.With the development of the times,enterprises have achieved common development through online and offline integration through logistics.In order to adapt to the new retail era,enterprises carry out a series of strategic layout and cooperation.With the advent of the new retail era,consumers can shop through various channels,and consumers' shopping experience will become the most important.Logistics distribution mode with high frequency and small quantity has become a common mode of transportation in modern terminal logistics.How to optimize the terminal logistics distribution method and integrate online and offline,so as to improve the efficiency of distribution and enhance consumers' shopping experience,is a research focus worthy of further study.This paper study the optimization method of physical stores' logistics distribution path under new retail from the point of view of logistics distribution path of terminal physical stores,under the condition that each physical stores cooperate with each other under new retail and each physical stores mask has a certain amount of inventory which can be used as a distribution center.The main research contents are as follows:Starting from the background of the new retail era,this paper introduces the basic concepts of new retail,terminal logistics,joint distribution,vehicle routing problem and Traveling Salesman Problem(TSP),and summarizes the research status of new retail mode and characteristics,new retail market pattern,joint distribution of terminal logistics and TSP.Secondly,from the point of view of logistics distribution path of end-store,genetic algorithm and K-means clustering algorithm are studied,and the advantages and disadvantages of traditional genetic algorithm and K-means clustering algorithm are put forward,which are insufficient to solve the current distribution path of end-store logistics under new retail.The genetic algorithm and K-means clustering algorithm are improved and combined.The clustering center is fixed by modifying the clustering rules,and the TSP path distance is obtained by genetic algorithm as the clustering condition.At the same time,according to the Euclidean distance of the cluster center as the convergence rule,the better solution is obtained by continuous re-clustering and according to the convergence condition.The distribution strategy after re-clustering can be new.Logistics distribution route optimization under retail background.In addition,because physical stores with the same or similar inventory and partnership can be used as distribution centers for collaborative distribution,the hybrid algorithm proposed in this paper also aims at merging and optimizing orders belonging to different distribution centers but at the same distribution target point,thus reducing duplicate distribution routes,adapting to the current new retail model,and reducing logistics generation.In order to improve user experience,a new distribution route optimization method is proposed.Finally,using the actual store data and orders in Tianhe District of Guangzhou as real experimental data,the algorithm is programmed and simulated through Tensorflow platform,and the two improved algorithms are compared and validated.The experimental results show that the proposed algorithm is effective.The experimental results show that compared with the traditional genetic algorithm,the improved K-means clustering algorithm with genetic algorithm can make the logistics distribution path to be greatly optimized.