Study On Integration Optimization Problem Of Distribution Network Under B2C Mode

With the increasing popularity of Internet,E-commerce has entered a period of rapid development in China,and consumers have become accustomed to ordering various commodities through Internet.The amount of online retail trade(B2C)in China is growing at a rate of nearly 40% per year.In order to seize this rapid development opportunity to occupy the market and expand business,express delivery enterprises or E-commerce enterprises,such as SF,Rookie Network and Jingdong,are committed to building or optimizing their logistics distribution network based on B2 C mode.However,in terms of distribution range of customer point,customers quantity,customer satisfaction and delivery method,the logistics distribution network under B2 C mode is different from traditional mode,which make the construction or optimization are diverse.Therefore,it is of great practical significance to find a suitable network integration optimization method by studying the characteristics of logistics distribution network under B2 C mode: it is beneficial to enterprises to establish an efficient logistics distribution system,improve distribution efficiency,reduce distribution cost,and improve Customer satisfaction,increases the market share.Based on this,this paper takes the facilities in the logistics distribution network under the B2 C mode as the research object.According to the different characteristics of each facility in the network,the algorithms and models suitable for the location of the corresponding facilities are designed and proposed.Bottom-up implementation of the integration optimization of logistics distribution network under B2 C mode,provides a better solution for express delivery companies or e-commerce companies to build or optimize their own logistics distribution network.The main contents and contributions are as follows:(1)According to the pickup point as a place directly serving customers in B2 C mode,the problem of integrated location based on the number and location of the pickup points based on customer points mixed address is studied.Firstly,according to the longitude and latitude coordinates and the administrative address,two kinds of customer point addresses algorithms are designed respectively to realize the location selection of the pickup point.According to the longitude and latitude coordinate address of the first customer point,a grid dynamic density clustering algorithm is designed to determine the number and the alternative location of the pickup point.According to the administrative address of the customer point,the customer point administrative address clustering algorithm is designed,and index points representing customer points are clustered as the alternative location of the pickup point.Then,a pickup point gradually covering location model with customer satisfaction and maximum coverage as two objectives is established to solve the self-service point location problem.Finally,the model is solved in IBM ILOG CPLEX.The effectiveness of the algorithm and model is verified by an example.Two kinds of pickup point location strategies for network retail distribution are proposed based on the calculation results.(2)In the considering of the characteristics of circular distribution and timeliness of distribution from transit station to pickup points in the B2 C mode,the distribution area partition algorithm based on the minimum spanning tree and the balanced distribution algorithm based on the volume and time are presented,and also,the distribution mode and distribution area partition on the basis of delivery timeliness of self-lifting point under the B2 C mode is implemented.Then,a two-layer path planning strategy is developed for customers who need to deliver goods to the door.An upper-layer path planning model aimed at acquiring the shortest distribution path is employed,which can be solved by an improved ant colony algorithm,to determine the path planning between the transfer station and pickup points in the distribution area.And the lower-layer door number sorting algorithm is used to finish the path planning between customer points belong to the pickup point respectively.In addition,the location problem of the transfer station is studied according to the planned distribution path,and the location of the transfer station is decided by the linear programming model and heuristic algorithm.Finally,the validity and practicability of the algorithms and models are verified by the calculation results of distribution area division,transit station location and distribution path planning,taking the customer points in Yudong Street,Ba'nan District,Chongqing as an example.(3)The location of distribution center based on transit station distribution in distribution network under the B2 C mode is studied.Owing to the hub-and-spoke distribution mode of the distribution from the distribution center to the transfer station,the improved K-means clustering algorithm and distribution volume balancing algorithm are designed to determine the alternative location of distribution center.Then,considering the location characteristics of multiple coverage of distribution center to transit station under B2 C mode,the distribution center with both shortest distribution distance and minimum construction cost is established.Set coverage location model is solved in CPLEX.Finally,an example is given to verify the validity of the location algorithm and model.(4)The previous research only concentrated on the optimization of single-level facility location in the current distribution network.In order to implement the integration optimization of the two-echelon logistics distribution network facilities under the B2 C mode,the research on location and route planning of the two-level facilities under the B2 C mode was carried out.The technical and economic indicators such as facility capacity,transportation model,facility construction cost and transportation cost were considered to establish the two-level facility integration optimization model,and the improved simulated annealing algorithm of the optimization model was designed.Finally,the validity of the algorithm for solving the problem of two-stage facility location path is verified by a case study,which can provide theoretical and methodological guidance for the logistics distribution network design of e-commerce enterprises.