Using a Hybrid Particle Swarm Optimization Method for the Inventory Routing Problem

• Main Authors: Li, Hong-Shuo, 李泓碩
• Other Authors: Shu-Chu Liu
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/57378621925109869690

碩士 === 國立屏東科技大學 === 資訊管理系所 === 101 === The inventory routing problem belongs to NP-hard problems, it is not easy to get the optimal solution. Therefore, in order to get the optimal solution. That is necessary to use heuristic algorithms to calculate in medium or large problem size. On the basis of reference documents and past studies, the investigators found out that using particle swarm optimization algorithm(PSO) in Travelling Salesman Problem(TSP),Vehicle Routing Problem(VRP) and Open Location Routing Problem(OLRP) to solve the path combination is the best way. In consequence, I used Particle Swarm Optimization Algorithm as a foundation to develop the structure of algorithms and we combined the hybrid Large Neighborhood Search(LNS) to ameliorate the disadvantage: (1) fast convergence and (2) fall into regional solutions in Particle Swarm Optimization Algorithm. Then, we developed a hybrid algorithm to plan optimal path combination and transport cycle to make cost minimization in inventory routing problem. The results of this study shows the hybrid particle swarm optimization algorithm’s cost is 2.41% better than particle swarm optimization algorithm in the small size problem, and in the big size problem the cost is 1.84%. The results showed a good performance to use hybrid algorithm in inventory routing problem in my study.