Fuzzy Inference – based FMS Dynamic Dispatching

• Main Authors: ying-jie Liu, 劉英傑
• Other Authors: Ming-Shan Lu
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/73283204045510786770

碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 93 === Due to the variety of product and rapid demand, manufacturer must have capability and flexibility to product high quality and variety of product to satisfy the customer needs. Therefore, the Flexible Manufacturing System suited for small to medium size and batch production had been developed. Under the uncertain and complex manufacturing environment, there are some variable that affect the production performance of different objectives. If the scheduling and dispatching for work order are planned with considering only one objective, the flexibility and agility for flexible manufacturing system can not be achieved, at the same time, the performance of multiple objectives can not be also achieved. In this research, a dynamic dispatching rules for multiple objectives based on fuzzy inference in proposed. First, the variables affecting the system performance are routed out. Then the fuzzy membership function and fuzzy rule database for dispatching rule are established from simulation data. The priority of each dispatching rule is decided based on the degree of influence on objective by the statues of variables. The highest priority of dispatching rules in selected to improve system performance and stability. From the simulation results the performance of the proposed fuzzy-based dynamic dispatching rules is better than fixed dispatching rules.