Integrating Dynamic System and Overall Performance Methods to Solve Dynamic Multiple Response Problems: A Case Study for IC Industries

• Main Authors: Hung Li-Ping, 洪麗萍
• Other Authors: 劉浚明
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/84amw2

碩士 === 逢甲大學 === 工業工程與系統管理學系 === 103 === Due to the rapid development and progress of the semiconductor industry, the global economy has become more prosperous. How to improve the semiconductor manufacturing quality and increase the yield in production are the challenges that most semiconductor companies must face. In this study, the dynamic systems and overall performance index methods are proposed to solve the dynamic multiple response quality problem raised in the semiconductor manufacturing companies in order to improve the manufacturing quality. One semiconductor company in Taiwan is selected as an case study. Six control parameters including deposition temperature, deposition pressure, nitrogen flow, silane flow rate, setting time and cleaning methods and three response factors including defect rate, layer density and deposition rate in the manufacturing process are considered in the experiment. Using dynamic regression analysis and overall performance index we solve for the best combination of control parameters, achieving the desired state of response variables. Results of this study suggest that by finding the best combination of parameters the semiconductor company can improve productivity in manufacturing processes, achieve higher product quality and increase efficiency to reach the enterprise sustainable development goals. Keywords: Dynamic multiple response quality problems, overall performance index, dynamic systems.