Application of Genetic Local Search Algorithm to Location-Allocation Problem in Reverse Logistics Network

Abstract

Location-allocation problem is one of the important decision problems in designing logistics network. Reverse logistics plays a key role in current management strategy of many foreign and domestic enterprises. The complexity of reverse logistics makes the operation of logistics system more dependent on logistics network, so the location-allocation problem in reverse logistics network has been the most important problem in the logistics system management. The Location-allocation problem in designing reverse logistics network is analyzed in this paper at first, and then the mathematical model of the problem is formulated. In the model the logistics network is shared by forward logistics and reverse logistics, and the demanded quantities is considered as stochastic variable. The objective of the model is to optimize the total cost of the logistics system. Because the formulated model is NP-hard, genetic local search algorithm is designed to solve the problem. The local search is used in the algorithm to improve the search effectiveness. Simulations based on several numerical examples shows that the proposed mathematical model and the algorithm are effective.