在工業製程上，設定最佳製程參數是提升產品品質且可以有效節省成本的方法之一。本文中，我們將探討在多特徵目標值的情形下，利用Taguchi的損失函數，建立相關模型，並求此模型下的最佳參數。為方便討論，最佳參數仍以平均數為主。而在變數型態上，我們的產品輸出特徵將同時考慮量化及質性特徵，針對質性特徵，我們利用潛在變數模型進行建模，並以牛頓法求解，最後舉數值範例加以說明，提供一個思考方向。

Abstract

In the industrial manufacturing process, setting the optimum process parameter is a cost-saving method to improve product quality effectively. In this article, we will use Taguchi's loss function to generalize the model and explore the optimal parameters of the multiple input variables under linear assumption. To facilitate our discussion, the best average is still the mean parameter. We will consider the quantitative and qualitative characteristics in the output variable. For the qualitative characteristics, we use latent variable model and consider Newton's Method to find the optimal process parameter.
Finally, we illustrate the paper with some examples and provide a direction of thinking.

[1]Taguchi, G., Chowdhury, S., Wu, Y. ( 2004).Taguchi’s Quality Engineering Handbook, Wiley-Interscience.
[2]Taguchi, G. ( 1986). Introduction to Quality Engineering , Asia Productivity Organization, Tokyo.
[3]Huang,Y.-F. ( 2001). “trade-off between quality and cost,” Quality& Quantity , 35 , pp.265-276.
[4]Yang,S.F. (1997). “An optimal design of joint & control charts using quadratic loss function, ” Inter. J. of Quality and Reliability Management, 11(9) , pp.948-966.
[5]Taguchi, G.( 1987). Introduction to Quality Engineering , Course Manual, American Supplier Institute Inc.
[6]Golhar, D.Y and Pollock, S.M. ( 1988). “Determination of the optimal process mean and the upper limit of the canning problem,” Journal of Quality Technology, 20 , pp.188-192.
[7]Dodson , B.L.( 1993). “Determining the optimum target value for a process with upper and lower specifications, ” Quality Engineering, 5(3) , pp.393-402.
[8]Spring,F.A. and Yeung A.S. ( 1998). “A general class of loss functions with industrial applications, ” Journal of Quality Technology , 30 , pp.152-162.
[9]Huang, Y.-F.( 2000). “Determination of the Best Mean in the Taguchi：Quality Selection Model, ”International Journal of Production Planning & Control, 11(6) , pp.565-571.
[10] Elsayed, E.A., Chen, A.( 1993). “Optimal levels of process parameters with multiple characteristics, ”Inter. J. of Prod. Res. , 31(5) , pp.1117-1132.
[11]Lee, M.K., Kwon, H.M., Hong, S.H., and Ha, J.W. (2006). “Economic selection of sub-process mean values for a mixture roduction process, ” Inter. J. of Prod. Res. , 44(20) , pp.4367-4376.
[12]Varian, H. R. (1975). A Bayesian approach to real estate assessment. In：S. E. Fienberg and A. Zellner (Eds.), studies in Bayesian Econometrics and Statistics in Honor of Leonard J. Savag, Amsterdam：North-Holland, pp. 195-208.
[13]Zellner,A.( 1986). “Bayesian estimation and prediction using asymmetric loss functions. ”
Journal of the American Statistica Association , 81 , pp.446-451.
[14]Chang, Y.-C. and Hung, W.-L.( 2007). “LINEX loss functions with applications to determining the optimum process parameters, ” Quality& Quantity , 41 , pp.291-301.
[15] Chang,Y.-C., Huang, W.-L., and Hung, W.-L. ( 2009). “Optimization of process parameters using weighted convex loss functions , ” European Journal of Operational Research, 196 , pp.752-763.
[16]李輝煌(2008)，田口方法：品質設計的原理與實務，高立圖書有限公司。
[17]陳蓉貞(2006)，考量品質損失函數於多階段連續生產系統之最佳製程平均數設定，南台科技大學工業管理研究所碩士論文。
[18]徐業良(1995)，工程最佳化設計，宏明圖書印製，華泰圖書總經銷。