Research On Optimizayion Of Out-of-stock Allocation Problem Under Multi-item Combination

This paper studies goods delivery problem faced by the fast fashion retail industry.In the fast fashion industry,retailers will forecast the sales volume for each item in next week,and arrange the vehicles to pick up or deliver the goods with exact number in each store at the beginning of each week.If the current number of a specific item is more than its prediction,the extra goods need to be picked up and transported to the warehouse.Otherwise,the insufficient quantities should be replenished.This situation derives the pick-up and delivery vehicle routing problem with multi-item requirement.Such frequent transportation result in relatively high logistics costs and warehouse processing costs.In addition,this problem is generically also applied to the international crude oil raw material transportation,raw material transportation in the tobacco manufacturing industry and the sharing of bicycle station scheduling problems,which has universal practical significance and research value.This article will study from the following aspects:(1)This article firstly summarizes the existing literatures on vehicle routing problems with pick-up and delivery,derives the actual vehicle routing problems faced by fast fashion retail industry,compare with existing research results,and find similarities and differences.Finally,this article determines the research direction and optimization objectives;(2)In the studied problem,it is recommended to transport excess goods from one store to another,and the goods that cannot be satisfied by stores are picked up from the warehouse or shipped back to the warehouse,and the warehouse handling costs occurs.A mixed integer programming model is formulated by minimizing the total traveling distance while minimizing the total handling cost of the warehouse;(3)For pick-up demand and delivery demand,the store is allowed to be visited by different batches of vehicles,and the split unit idea is used to determine the upper limit of the number of visits under multiple visits.Based on this idea,a mathematical programming model is established.Compared with the previous model,the results show that multiple accesses will make the problem more complicated,but the optimization objective is better;(4)In order to solve the large-scale problem in reality,an efficient heuristic algorithm based on genetic algorithm is proposed to solve the pick-up and delivery routing problem with multi-item requirement under multiple visits.The proposed algorithm uses local search to generate the initial population and adopts the 2-opt operator to refine the population.The numerical results verify the performance of the proposed heuristic algorithm.