Full metadata record
DC FieldValueLanguage
dc.contributor.author林敬凱en_US
dc.contributor.authorChing-Kai Linen_US
dc.contributor.author唐麗英en_US
dc.contributor.author洪瑞雲en_US
dc.contributor.authorLee-Ing Tongen_US
dc.contributor.authorRuey-Yun Horngen_US
dc.date.accessioned2014-12-12T01:17:53Z-
dc.date.available2014-12-12T01:17:53Z-
dc.date.issued2007en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT009533546en_US
dc.identifier.urihttp://hdl.handle.net/11536/39174-
dc.description.abstract工業界常使用實驗設計(Design of Experiments, D.O.E.)與田口方法(Taguchi Methods)來設定製程參數最佳水準組合以降低研發成本或改善產品品質。然而D.O.E.及田口方法僅適用於單一品質特性之最佳化,但隨著消費者對產品品質的要求日趨複雜,產品品質之好壞已非單一品質特性最佳化可以解決,因此發展一套有效率同時最佳化多個品質特性之演算法,就成為現今工業界提升品質的重要技術。目前中外文獻所提出之多品質特性同時最佳化的方法,大多是針對田口方法所得之實驗數據而設計,因此不能應用到由D.O.E.所得之實驗數據上。此外,由於田口方法的一些限制,D.O.E.比田口方法更容易達成多個品質特性同時最佳化的目的。因此,本論文針對由D.O.E.所規劃之具多品質特性的實驗數據,結合資料包絡分析法(Data Envelopment Analysis, DEA)之交叉效率(Cross Efficiency)模式與自組性演算法(Group Method of Data Handling, GMDH),提出一套多品質特性同時最佳化演算法。本論文最後引用新竹科學園區某半導體公司之蝕刻製程的實際案例,以說明本演算法確實有效可行。zh_TW
dc.description.abstractDesign of experiments (D.O.E.) and Taguchi methods are often applied in industry to reduce the cost of finding the optimal parameter-setting of a process or to improve the product quality. Both of D.O.E. and Taguchi methods can only be used for optimizing one quality characteristic. However, recent consumers have more concerns about the product quality, optimizing just one quality characteristic can not satisfy consumers’ demands. Consequently, developing a procedure to simultaneously optimize multiple quality characteristics becomes an important issue in recent years. Most of the previous studies on optimizing multi-response problems are focused on experimental data from Taguchi methods. Because of the limitations of Taguchi methods, an optimal parameter-setting can be obtained more easily via D.O.E. Therefore, this study develops a procedure to simultaneous optimizing the multi-response problem with experimental data from D.O.E utilizing the cross efficiency model of Data Envelopment Analysis (DEA) and Group Method of Data Handling (GMDH). A real case from a wafer manufacturing company in Taiwan is used to demonstrate the effectiveness of the proposed procedure.en_US
dc.language.isozh_TWen_US
dc.subject實驗設計zh_TW
dc.subject多品質特性同時最佳化zh_TW
dc.subject資料包絡分析法zh_TW
dc.subject交叉效率模式zh_TW
dc.subject自組性演算法zh_TW
dc.subjectDesign of Experimentsen_US
dc.subjectOptimization, Multiple Responsesen_US
dc.subjectData Envelopment Analysisen_US
dc.subjectCross Efficiency Modelen_US
dc.subjectGroup Method of Data Handlingen_US
dc.title具多反應實驗設計最佳化演算法之研究zh_TW
dc.titleOptimizing Multiple Responses in Designed Experimentsen_US
dc.typeThesisen_US
dc.contributor.department工業工程與管理學系zh_TW
Appears in Collections:Thesis