Research On Distribution Center Layout And Heterogeneous Vehicle Routing Problem Optimization Of T Company

With the rapid progress of technology and information technology,e-commerce is booming.As a branch of e-commerce,fresh e-commerce has always been concerned.As a necessity in people’s daily life,fresh agricultural products have the characteristics of wide target customers,high purchase frequency,vulnerable consumption,etc.,but due to the small and scattered orders of C-end customers,it is more difficult to distribute,Some enterprises began to turn their attention to B-end customers such as community convenience stores,catering enterprises,canteens,etc.with large demand for fresh agricultural products and relatively stable demand.At present,enterprises engaged in one-stop distribution of fresh agricultural products on the B-end are springing up.For such enterprises,the key to be invincible in the fierce market competition is to achieve high business operation efficiency,low operation cost and good customer service experience.This study takes the agricultural products e-commerce project of T company as the research object.Based on the in-depth investigation and analysis of the project,it finds that there are problems such as low sorting efficiency,high distribution cost and serious customer loss.In view of the existing problems of T company,SLP method is used to optimize the distribution center layout and improve the sorting efficiency of the distribution center.Then,aiming at the situation that multiple distribution models coexist in the distribution process of the project,and the customer demand can be split according to the goods category,aiming at minimizing the distribution cost,minimizing the cost of goods damage and maximizing customer satisfaction,a heterogeneous vehicle routing optimization model with split demand is established.A genetic algorithm is designed to solve the problem,split the customer demand by adding virtual customer nodes.Compared with the current distribution scheme of T company,the optimized result reduces the total cost by 25.04%.Then,based on the distance between the customer node and the distribution center,this paper explores the changes of the algorithm’s running time and the solution results when the customer demand is split within different radius.It is found that in the process of taking the distribution center as the center and increasing the radius of the split range,the quality of the solution increases,but due to the limitation of the algorithm’s optimization ability,the quality of the final solution obtained by the algorithm begins to decline when the total customer node exceeds a certain scale.Considering the quality of solution time and reconciliation,it is suggested that the project should make distribution plan on the premise of only splitting the customer demand within 5km from the distribution center.