應用雙反應曲面法最佳化多品質實驗設計

Abstract

工業界常使用實驗設計(Design of Experiments, D.O.E.)與田口方法（Taguchi Methods）來找尋製程之最佳因子水準組合以改善製程及提升產品品質。然而實驗設計與田口方法僅能最佳化單一品質特性，但現今消費者的需求不再侷限於單一的品質特性，因此發展一套有效率的多品質特性同時最佳化之演算法，成為現今工業界提升品質的重要議題。目前有關於多品質特性同時最佳化之中外文獻，大多是針對田口方法規劃實驗所提出的多品質同時最佳化演算法，少有文獻是針對以實驗設計所規劃實驗提出多品質同時最佳化演算法。因此，本研究之主要目的是發展一套適用於實驗設計的多品質同時最佳化之演算法，本論文利用傳統實驗設計法規劃實驗，結合灰關聯分析法(Grey Relation Analysis, GRA)及雙反應曲面法(Dual Response Surface Methodology, DRSM)提出一套多品質特性同時最佳化之演算法。本研究最後以新竹科學園區某半導體公司之蝕刻製程及薄膜製程的實際案例，驗證本論文所提出的演算法確實有效可行。

Design of experiments（DOE）and Taguchi methods are often applied in industry to develop new product or improve the process. Due to the product design becomes increasingly complicated, more than one quality characteristic must be simultaneously optimized to improve the product quality. However, DOE and Taguchi method can only deal with optimizing single-response problem. Consequently, developing a procedure to simultaneously optimize multiple quality characteristics becomes an important issue in. Most of the literatures of optimizing multi-response problems are for Taguchi’s experiments. Furthermore, some studies indicate that optimization of DOE is more effective than Taguchi’s method. The purpose of this study is to devolop a method of finding the optimal parameter-setting of a process or product with multi-response using DOE, Grey Relation Analysis(GRA) and Dual Response Surface Methodology（DRSM）. Finally, two real cases from a semiconductor factory in Taiwan are utilized to verify the effectiveness of the proposed procedure.