應用資料探勘技術控制PI製程之拉力值

Abstract

在製造polyimide film過程中會發生拉力值(Peel)的良率問題。對於製造業而言，製程良率之高低會影響產品的成本以及企業之競爭力，因此良率的改善已成為各競爭廠商的重要課題。對於複雜的製造程序所產生的大量製程資料，回顧過去製程良率改善文獻，發現資料探勘技術能有效且迅速的找出影響製程的關鍵參數，提升產品製程良率。因此本研究利用資料探勘技術1R與C4.5並結合統計迴歸分析，建構一個改善製程良率的資料探勘新架構，實地應用於台灣某製造polyimide film公司的製程上，結果有效的找出關鍵因子幫助公司將製程良率從43%提升到84%。

Peel is a quality attribute in polyimide film (PI) production. A piece of PI film that is easy to be peeled off is unfavorable. In practice, it is hard to reduce the peel issues because the PI production process includes a few hundreds of process parameters. This research combined two data mining algorithms, 1R and C4.5, for identifying the critical process parameters for reducing the peel issues. Two critical process parameters as well as their appropriate ranges of values were identified by the proposed approach. An empirical experiment reveals that the process yield for reducing the peel issues could be improved from 43% to 84%.