Semi-dynamic Capacity Planning for Semiconductor Manufacturing

• Main Authors: Chi-Rei Weng, 翁啟瑞
• Other Authors: Yon-Chun Chou
• Format: Others
• Language: zh-TW
• Published: 1999
• Online Access: http://ndltd.ncl.edu.tw/handle/79561773586283862593

碩士 === 國立臺灣大學 === 工業工程學研究所 === 87 === Queueing and simulation methodologies have been applied to capacity modeling in semiconductor manufacturing. The advantage of queueing capacity model is the rapid computation time. The software that implements a queueing model can usually obtain estimates under two minutes for fabs that take hours to simulate. Simulation model could provide detail information but consume numerous computation load. However, product mix is changed rapid since the life cycle of the product is shorter than before and market demand is fluctuated with time. The manufacturing system would not be the steady state because of short product mix cycles and the long process flow production. The queueing capacity model based on steady state is invalid for the semiconductor manufacturing. This study developed semi-dynamic capacity model for non-steady state manufacturing system using queueing capacity model. This research contains queueing and semi-dynamic capacity models. First, the characteristic of semiconductor manufacturing is described. Then, a queueing capacity model and WIP tracking system were developed. The system evaluates non-steady state manufacturing system using agile simulation-based model.