Comparing heuristic algorithms for the logistics network design model considering demand fill rate constraint

The design of logistics system satisfying demand of each retailer has been one of the important issues in Supply Chain Management. Therefore, various models including the fixed charge facility location model (FCFLM) have been proposed so as to present logistics network. As more researchers are interested in FCFLM, some practical issues are incorporated into the traditional FCFLM. For example, Shen et al.(2003) considered new location model combining FCFLM with inventory management model. Another extension of FCFLMis to consider demand fill rate which is measured by the ratio of demand fulfilled within the given time or fulfilled by a DC located within given distance.(Melo et al., 2009) This paper deals with the combined model of FCFLM and inventory management model considering demand fill rate constraint. The considered model is formulated as a non-linear integer programming and two heuristic algorithms based on Tabu Search and GRASP are proposed. To compare the performance of the proposed algorithms, 1620 data sets considering 5 design factors are randomly generated. The performances of the algorithms are measured in terms of the average objective function value and the average elapsed time. Test results show that Tabu Search based heuristic algorithm outperforms GRASP based heuristic algorithm in terms of the effectiveness and GRASP based heuristic algorithm outperforms in terms of Tabu Search based heuristic algorithm in terms of the efficiency. The contributions of this paper are two-fold. First, it developed a new heuristic algorithm based on Tabu Search to solve the combined model of FCFLM and inventory management model considering demand fill rate constraint. Second, it compared the performance of the proposed algorithm with the existing GRASP algorithm.