Solving the Location - allocation Problem of Distribution Center by Tabu Search Approach

• Main Authors: LI-JUNG SHI, 史立榮
• Other Authors: SHIH-WEI LIN
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/71123432976712240124

碩士 === 華梵大學 === 資訊管理學系碩士班 === 96 === It is evidence that the decision of the distribution center location problem (includes location and the number of location distribution center) is a troublesome and hard problem to be faced for logistics industry. Due to the fact that the number and the location of distribution center is heavily related to the setup cost and the transportation cost. The purchase (lend) cost not only influences the setup cost of distribution center, but also affects the transportation cost and the service level provided. The more distribution centers used the less transportation cost it has and the higher service level it has. However, the setup cost is also increased significantly. On the other hand, the fewer distribution centers used the higher transportation cost it has, while the setup cost can be reduced. Therefore, to take consideration of cost control and the required service level, how to apply information technology and minimize the related cost is an import issue in the logistics industry. The problem to be solved in this proposal is the variants of the set covering problems (SCPs). In addition to determine the location of distribution center, the number of distribution center, and the demand points is served by which distribution center is also required, the complexity of the proposed problem is more than that of SCPs. This proposal not only provides the mathematical model of the argued problem, but also applies the ILOG optimization software to obtain the real optimal solutions for problems. However, due to the constraint of the optimization software, it can be expected that the large-scaled problems cannot be solved by optimization software. Therefore, this proposal is plan to adopt the tabu search approach to solve this problem. In addition to obtain the real optimal solution of small-scaled problems using less computational time, the large-scaled problems will be used to verify the effectiveness and availability of the proposed approach.