Using Simulation Method to Solve a Two-Echelon Repairable Inventory System Problem Subject to Service Constraints

• Main Authors: Fei-FangKuo, 郭斐芳
• Other Authors: Shing-Chih Tsai
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/52213229047985218518

碩士 === 國立成功大學 === 工業與資訊管理學系碩博士班 === 100 === In the field of production management,the study of multi-echelon inventory system has been a hot topic. We study a two-echelon repairable inventory system, which contains a central warehouse and multiple field depots that stock spare parts. When a failure occurs,the field depots serve the customers with part replacement and replenish their inventory from the central warehouse, following a base stock policy. The central warehouse also acts as a repair facility, and replenishes its inventory by repairing the defective parts passed by the field depots. The goal of our research is to find the best stocking levels in the central warehouse and field depots to minimize the system-wide inventory investment while maintaining an acceptable level of the expected response time over multiple field depots. A two-echelon inventory system problem consists a deterministic objective function and stochastic constraints in our study. The problem we formulated has relatively large solution space; however,ranking and selection procedure (R＆S) is mainly applied for the problem with a small number of solutions. Therefore, we propose a simulation optimization algorithm that combines regression models and R＆S to solve the problem. We integrate regression models and R＆S in our approach, which has never been proposed in solving simulation optimization problems to our knowledge and provide numerical results to show the efficiency of our proposed method.