傳統的田口方法主要是針對單一品質特性的最佳化，實際上，產品或製程常常具有兩個或兩個以上的品質特性。而一般處理多重品質特性問題，只能依靠工程人員的經驗來選擇因子水準，而由於工程人員的認定標準不一，容易產生不確定性與模糊性，且各品質特性的相關性愈強或所考量的品質特性愈多時，工程師要決定最佳因子水準組合就愈困難，同時帶給工程人員解決多重品質特性上衝突的難度。因此，產品多重品質特性最佳化的決定對提昇企業產品品質是相當重要的議題。

針對這項議題，本研究先利用田口方法中處理單品質特性的方法，分別計算出各品質特性的S/N比(signal to noise ratio，信號雜訊比)，再運用灰關聯分析將S/N比轉為灰關聯係數，最後結合模糊推論系統導出代表多重品質特性之衡量指標(MPCI)，利用其數值愈高愈好的特性，找出最佳參數組合，再由變異數分析找出同時影響兩種品質特性的顯著因子，以進行包含兩種以上品質的最佳參數組合之模擬驗證，俾驗證最佳參數組合之正確性。

最後，利用案例-液晶顯示器製程來闡述所提最佳化程序與模式之數值分析，驗證本研究之有效性與實用性，幫助公司節省實驗成本、縮短新產品由實驗階段導入生產階段時程。研究結果顯示，本研究不僅可以應用於不同品質特性組合，且都有提升原先之產品品質。

The traditional Taguchi method mainly aims at the optimizing of the single quality characteristic. In fact, the product or the process frequently has two or two above quality characteristics. And generally, deals with the problem of multiple quality characteristic, only can depend on the engineer's experience to choose the factor level. Because the engineer’s recognizes are different, it’s easy to have the uncertainty and the fuzziness, when each qualities characteristic relevance is stronger or which considers the quality characteristic are more, engineer must decide the best factor level combination is more difficult, simultaneously takes to the engineer to solve the difficulty which in the multiple quality characteristic conflicts. Therefore, the product multiple quality characteristic optimization decision to promotes the enterprise product quality is the quite important subject.

In view of this subject, the research first uses Taguchi method by only considering the single quality characteristic ,then separately calculating the S/N ration(signal to noise ratio) according to the quality characteristic, then transferring the S/N ration by using the Grey Relation Analysis to the Grey Relational Coefficient, finally unifies the Fuzzy Inference System to find the Multiple Performance Characteristics Index(MPCI) , uses the characteristic which the value is the higher the better ,to find the best parameter combination, then to find the main factor which affects two quality characteristic simultaneously by Analysis of Variance, demonstrating the best parameter combination of two above quality characteristics. And demonstrating the correctness of the best parameter combination.

Finally, using the case – LCD process elaborated proposes which is the optimization procedure and the pattern of numerical analysis. Demonstrating the effectiveness and the usability of the research, helps the company to save the experimental cost and to reduce the new product to induct the production base time interval by the experimental stage. The findings showed, not only this research may apply in the different quality characteristic combination, also all has promoted originally product quality.

摘要............................................................................................................I
Abstract.................................................................................................III
誌謝............................................................................................................V
目錄..........................................................................................................VI
圖目錄......................................................................................................IX
表目錄......................................................................................................XI
第壹章 緒論..............................................................................................1
1.1 研究背景.........................................................................................2
1.2 研究動機.........................................................................................3
1.3 研究目的.........................................................................................4
1.4 研究限制與範圍.............................................................................4
1.5 研究流程與內容.............................................................................5
第貳章 文獻探討......................................................................................7
2.1 田口品質工程.................................................................................7
2.1.1 損失函數..............................................................................9
2.1.2 田口品質特性....................................................................10
2.1.3 影響產品或製程績效的因子............................................12
2.1.4 穩健設計............................................................................14
2.2 多重品質特性製程.......................................................................15
2.2.1 以工程人員之專業知識做判斷........................................15
2.2.2 主成份分析法....................................................................16
2.2.3 多屬性決策法....................................................................17
2.2.4 統計相關方法....................................................................18
2.2.5 類神經網路........................................................................19
2.2.6 模糊集合理論....................................................................19
2.2.7 灰關聯分析........................................................................21
2.3 LCD顯示器....................................................................................23
2.4 本章小結.......................................................................................25
第參章 研究方法....................................................................................27
3.1 灰色系統理論...............................................................................27
3.1.1 灰關聯分析........................................................................28
3.2 模糊理論.......................................................................................31
3.2.1 模糊集合............................................................................32
3.2.2 模糊推論系統....................................................................34
3.3 研究方法小結...............................................................................36
第肆章 建構算則與實證........................................................................39
4.1 模型的建構與求解.......................................................................39
4.2 實例驗證.......................................................................................44
4.2.1 膠框製程............................................................................44
4.2.2 光片製作製程....................................................................59
4.2.3 金屬鍍膜製程....................................................................73
第伍章 結論與建議................................................................................88
5.1 結論...............................................................................................88
5.2 建議...............................................................................................89
5.3 未來研究.......................................................................................89
參考文獻..................................................................................................90

中文部份：
1.王宗富，「多重品質特性製程參數最佳化研究-以高分子有機電激發光顯示器為例」，國立台灣科技大學工業管理系碩士論文，民國九十年六月。

2.王春和，「田口方法於線外品管多種品質特性製程最佳化之應用研究」，國立交通大學工業工程研究所碩士論文，民國八十二年六月。

3.方國富，「多品質特性產品之穩健參數設計」，國立成功大學工業管理科研究所碩士論文，民國八十九年六月。

4.王耀南(1996)，「智能控制系統-模糊邏輯、專家系統、神經網路控制」，湖南大學出版社，長沙。

5.吳祥輝(2003)，「應用模糊田口方法於架空式起重機桁架穩健多目標最佳化設計」，國立高雄第一科技大學機械與自動化工程研究所碩士論文。

6.吳漢雄、鄧聚龍、溫坤禮(1996)，灰色分析入門，高力圖書公司。

7.林江龍(1999)，放電加工電極消耗可靠度與製程參數最佳化研究，國立中央大學機械工程研究所博士論文。

8.紀勝財和徐立章（2000），「模糊多重品質特性田口實驗設計法之建立與電漿電弧銲接之應用」，Journal of the Chinese Institute of Industrial, Vol. 18, No 4, pp.97-110。

9.柯瑞芬，「多重品質特性之產品穩健設計決策模式」，國立成功大學工業管理科研究所碩士論文，民國八十四年六月。

10.洪維宗(2002),灰色關聯分析於整合田口方法多重品質特性最佳化-在食品工業製程實證，東海大學食品科學研究所碩士論文。

11.徐世輝(1998)，「品質管理」，三民書局。

12.孫宗瀛.楊英魁,1999,FUZZY控制理論、實作與應用, 全華科技圖書公司,台北。

13.夏郭賢、吳漢雄(1998)，“灰關聯分析之線性數據前處理探討”，灰色系統學刊，第一卷第一期，47-53。

14.翁義哲(2007)，應用模糊語意於多目標數控車削之最佳化研究，大同大學機械工程研究所碩士論文。

15.翁慶昌、陳嘉爜、賴宏仁(2001)，灰色系統基本方法及其應用，高立圖書公司。

16.陳志堅，「應用模糊理論於田口線外品管多重品質特性最佳化之問題研究」，國立交通大學工業工程研究所碩士論文，民國八十三年六月。

17.陳明佑(2002),「利用模糊目標規劃法求解田口式多品質特性最佳化問題」，國立成功大學工業 管理研究所碩士論文

18.張哲維，「模糊多屬性決策方法應用於田口方法多重品質特性問題-以IC 製造為例，中華大學工業工程與管理研究所碩士論文，民國八十八年六月。

19.莊達人，（2003）， VLSI製造技術，高立圖書有限公司，P236。

20.陳耀茂(1998)，「田口實驗計劃法」，滄海書局。

21.黃冠維(2006)，以灰關聯分析方法求解多屬性決策問題，國立成功大學製造工程研究所碩士論文。

22.黃得誌(2007)，結合灰關聯分析法與田口法於多目標數控車削參數最佳化研究，大同大學機械工程研究所碩士論文。

23.楊玉如，「應用主成份分析方法於多重品質製程最佳化之研究」，國立交通大學工業工程研究所碩士論文，民國八十五年六月。

24.黎正中譯(1993),穩健設計之品質工程,台北圖書有限公司。

25.鄧世輝，「漫談傳統型品質工程與田口型品質工程」，品質管制月刊，第二十三卷第一期，第34-44 頁，民國七十六年。

26.鄧聚龍(2000)，灰色系統理論與應用，高力圖書公司。

27.詹雅嵐，「應用多屬性決策法於動態系統多重品質特性最佳化之研究」，國立交通大學工業工程研究所碩士論文，民國八十七年六月。

28.葉馨雅，「動態系統多重品質特性最佳化之研究」，國立交通大學工業工程研究所碩士論文，民國八十六年六月。

29.劉冠鷨(2006)，應用灰色關聯分析與模糊推論於多重品質特性製程之研究，東海大學工業工程與經營資訊研究所碩士論文。

30.羅中育(2000)田口品質工程應用於模擬退火法參數組合之研究-以旅行推銷員問題(TSP)為例，國立雲林科技大學工業工程與管理研究所碩士論文。


英文部份：
1..Antony,J.(2000)"Multi-response optimization in industrial experiments using Taguchi's quality loss function and principal component analysis,"Quality and Reliability Engineering International,Vol.16,pp.3-8.

2.Deng, J.(1982), "Control Problems of Grey System", Systems and Control Letters,Vol.5,No.3,pp.288-294.

3.Deng, J.(1989), "Introduction to Grey System Theory", The Journal of Grey System.,Vol.1,No 1,pp.1-24.

4.G. Taguchi, Introduction to Quality Engineering, Asian Productivity Organization, Tokyo, 1990.

5.Hirano, K., Okamura, J., Taira, T., Sano, K., Toyoda, A., Ikeda, M., （2001）, " An efficient treatment technique for TMAH wastewater by catalytic oxidation ", Semiconductor Manufacturing, vol. 14, pp.202 - 206.

6.Hsu,Chih-Ming(2001),"Solving Multi-response Problems Through Neural Networks and principle Component Analyze,"Journal of the Chinese Institute of Industrial Engineers, Vol.18,No.5,pp.47-54

7.Jewell, T.E., White, D.L., （1989）, " Spatial frequency doubling lithography (SFDL) of periodic structures for integrated optical circuit technology ", Lightwave Technology, vol. 7, pp.1386 - 1393.

8.Jeyapaul, R., P. Shahabudeen, and K. Krishnaiah(2005), “Quality management research by considering multi-response problems in the Taguchi method –a review,” International Journal of Advanced Manufacturing Technology, Vol. 26, pp. 1331-1337.

9.Juan,H.S.,F.Yu, and B.Y.Lee(2003),"The optimal cutting-parameter selection of Production cost in HSM for SKD61 tool steels,"International Journal of Machine Tools and Manufacture,Vol.43,pp.679-686.
10.Koji, S., Koji, S., Yoshimitsu, O., Hiroshi, O., Yoshiaki, K., and Georg, P., （1999）, "Effect of photoacid generators on the formation of residues in an organic BARC process," Proc. of SPIE, vol. 3678, pp.380-387.

11.L.A.Zadeh,(1965),Fuzzy Sets,information and Control,Vol.8,p338-353.

12.Lee,B.Y.and Y.S.Tarng(2001),"Surface roughness inspection by compute vision in turning operations,"International Journal of Machine Tools and Manufacture,Vol.41,pp.1251-1263.

13.Lin, C. H., Wang , L. A. and Chen, H. L.,（2000）, "Optimized bilayer hexamethyl -disiloxane film as bottom antireflective coating both KrF and ArF lithographies," J. Vac. Sci. Technol. B, vol. 18, no. 6, pp.3323-3327.

14.Lin, C. H., Wang, L. A.,（2001）,"Feasibility of utilizing hexamethyldisiloxane film as a bottom antireflective coating for 157 nm lithography", J. Vac. Sci. Technol. B, vol. 19, no. 6 , pp.2357-2361.

15.Lin C.L.,W.D.Chou,and J.L.Lin(2001),"Optimization of the Electrical Discharge Machining Process based on the Taguchi Method with Fuzzy Logics,"Journal of Science and Technology,Vol.10,No.2,pp.119-127.

16.Lin,C.L.(2004),"Use of the Taguchi Method and Grey Relational Analysis to Optimize Operations with Multiple Performance Characteristics,"Materials and Manufacturing Processes,Vol.19,No.2,pp.209-220.

17.Lin, C. P., Yang, H. and Chao, C. K., （2003）, "A New Microlens Array Fabrication Method Using UV Proximity Printing," Journal of Micromechanics and Microengineering, vol. 3, no. 5, pp.748-757.

18.Lu,Dawei,and J. Antony(2002),"Optimization of multiple responses using a fuzzy-rule based inference System,"Taylor & Francis,Vol. 40,No.7,pp.1613-1625.

19.Phadke, M. S. (1989), Quality engineering using robust design, AT& T Laboratories.

20.P. J. Ross, Taguchi Techniques for Quality Engineering, McGraw-Hill, New York, 1988.

21.Provin, C., Monneret, S.,（2002）, " Complex ceramic-polymer composite microparts made by micro stereo lithography ", Electronics Packaging Manufacturing, vol. 25, pp.59 - 63.

22.Shiau,G.H.(1990),"A study of the sintering properties of iron ores using the Taguchi's parameter design,"Hournal of the Chinese Statistical AssociationmVol.28,No.2,pp.253-275.

23.Sung, M. G., Lee, Y. M., Lee, E. M., Sohn, Y.S., An, I. and Oh, H.K., （2000）, "Soft bake effect in 193 nm chemically amplified resist," Proc. of SPIE, vol.3999, pp.1062-1069.

24.Suzuki , K., Matsui , S., and Ochiai , Y., （2000）, "Sub-Half-Micron Lithography for ULSIs," chapter1, Cambridge University Press, Cambridge.

18.Tong,L.I.and C.T.Su(1997),"Multi-response robust design by principal component analysis,"Total Quality Management,Vol.8,No.6,pp.409-416.

19.Tong,L.I.and C.T.Su(1997),"Multi-response robust problems in Taguchi Method by fuzzy Multiple Attribute Decision Making,"Quality and Reliability Engineering International,Vol.13,pp.25-34.

20.Tong,L.I.,C.H.Wang,and H.C.Chen(2005),"Optimization of multiple responses using principal component analysis and technique for order preference by similarity to ideal solution,"International Journal of Advanced Manufacturing Technology Vol.27,pp.407-414

21.Wu,F.C.(2004),"Optimizing Robust Design for Correlated Quality Characteristics,"International Journal of Advanced Manufacturing Technology,Vol.24 No.1,pp.1-8.

22.Y. S. Tarng, W. H. Yang and S. C. Juang(2000),"The Use of fuzzy logic in the Taguchi Method for the optimization of the Submerged Arc Welding Process,"International Journal of Advanced Manufacturing Technology,Vol.16,pp.688-694.