단일 및 복수단계 역물류 네트워크 비교

Comparison of the Single and Multiple Stage Reverse Logistics Networks: Genetic Algorithm Approach

In this paper we develop two types of reverse logistics (RL) networks: single stage reverse logistics network (ssRL) and multiple stage reverse logistics network (msRL). The ssRL considers the RL network flow of customer -> integration center -> secondary market and the msRL considers the RL network flow of customer -> collection center -> remanufacturing center -> redistribution center -> secondary market. In the ssRL, the integration center simultaneously takes the functions of the collection center, remanufacturing center, and redistribution center considered in the msRL. The ssRL and the msRL are formulated using a mixed integer programming (MIP) model and the MIP model is implemented in genetic algorithm (GA) approach. In a numerical experiment, types I and 2 with various flows of RL network are presented and they are programmed using the GA approach. Various measures of performance such as CPU time, optimal solution, and optimal setting are considered to compare the efficiency of the ssRL and the msRL. Finally, the performance of the ssRL is more efficient in terms of the CPU time and optimal solution than that of the msRL.