Constructing a Issue Equipment Diagnosis AnalysisUsing Data Mining Technology In TFT-LCD Manufacturing Factory

• Main Authors: Chih-Wei Chuang, 莊智偉
• Other Authors: Ban-Chang Shia
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/10615407416482322165

碩士 === 輔仁大學 === 應用統計學研究所 === 96 === Yield improvement is a critical for TFT-LCD manufacturing factory to remain competitive in the market. In order to keep high yield, a quick response diagnostic tool to find root cause about a quality accident is more important in today’s manufacturing fab. The focus of this research is on the development of a quality diagnostic tool using statistics and data mining method. K-W Test, Box-Plot, and Decision Tree will be applied to build a diagnostic tool base to find the root causes of defective products. The integration engineers will trace root cause using the diagnostic tool and reference past experience or related process, product knowledge and then keep them into knowledge base in the corporation Data from a specific TFT-LCD manufacturer demonstrate that the proposed approach is a useful tool in tracking root cause for defective panels.