近年來，軟性電子顯示器受到越來越多的關注。由於光學膜具備許多重要的特性，例如可撓曲性和可折疊性、輕量、薄型化及大面積製造的可行性，因此使用光學膜來製造這種軟性電子顯示器。卷對卷光學膜製程有若干優點，它能夠長時間大面積進行塗佈作業，提高產量而降低成本。但長時間連續式的生產，製程參數設定可能會隨生產時間對反應變數產生影響、使整卷成品的厚度可能因此發生變化。傳統的實驗設計需要龐大的實驗次數，在有限的實驗資源下並不適用；使用部份因子實驗與田口直交表能夠在較少的實驗組合下取得有用的資訊，本研究先採用D-optimal建構實驗矩陣；將實驗結果透過顯著性檢定來瞭解控制因子與反應變數（厚度）間的相關性進行模型建立，再依據田口穩健設計設定因子水準。另一方面，因為均值與變異數跟實驗變數間可能會有複雜的關係，透過操作簡單、可行性高的雙反應曲面法來估計反應變數均值及標準偏差，再透過最陡路徑法沿著最接近目標反應變數的方向逐次實驗，將其複雜關係拆解，直到厚度均值最接近目標值為止；最終找出使整體厚度最接近目標值且變異最小的一組最佳參數組合。最後並透過比較個案公司現有的製程數據，證明了本研究方法的適用性和有效性，進而實現光學膜的目標厚度均勻性並降低品質變異，且結果可以作為後續製程參數調整之參考。

In recent years, wearable electronics have received increasing attention. Such wearable electronic products are often fabricated using optical films due to valuable properties exhibited by optical films such as very good flexibility, foldability, lightweight, and, thinness. Roll-to-roll coating process for manufacturing the optical films yields several benefits, processing a large-area at one time, resulting in high throughput and cost reduction. However, the process setting may cause the response change with the long and continuous production time. The Taguchi method can obtain useful information with fewer experimental combinations, and can reduce the impact of the cause of variation, and to make the system less sensitive to noise or variation on the functional characteristics of product or process. First, we generate the experimental design matrix to understand the correlation between the control factor and the response (thickness) through the significance test of regression analysis of the experimental results to build a model. The levels of factors were determined by Taguchi method, and the mean and variance of models were estimated through dual response surface methodology by using Path of Steepest Ascent to find the optimal setting of parameters for achieving the thickness target value and minimize variation of thickness. Finally, in the case study, we compare the uniformity performance using the optimal setting and the setting under the current operating conditions, the result finds that the variance reduced and thickness uniformity improved.

中文文獻
王柏齡，(2005)，反應六標準差之多品質特性混合實驗最佳化演算法，碩士論文，國立交通大學
朱君瀚，(2012)，以精密塗佈技術製備高分子發光二極體，碩士論文，國立清華大學
吳宏達，(2015)，結合田口與多屬性決策法求解多重品質特性之穩健設計-以鋰電池銲接製程為例，碩士論文，國立成功大學
李達源，莊愷瑋，(2003)，應用地理統計界定污染場址中之污染範圍，台灣土壤及地下水環境保護協會，7，2-13。
楊玫萍，(2008)，運用克利金與SpaceSyntax於建立監視器設置決策支援系統-以台中市水湳派出所轄區為例，碩士論文，逢甲大學
劉大佼，(2007)，塗佈加工技術，科學發展，418，62-67。
英文文獻
Baba, I., Midi, H., Rana, S., & Ibragimov, G. (2015). An alternative approach of dual response surface optimization based on penalty function method. Mathematical Problems in Engineering.
Box, G. E., & Draper, N. R. (1987). Empirical model-building and response surfaces. John Wiley & Sons.
Cressie, N.A. (1990). The origins of kriging. Mathematical Geology, 22(3) 239-252.
Ding, X., Liu, J., & Harris, T. L. (2016). A review of the operating limits in slot die coating processes. AIChE Journal, 62(7) 2508-2524.
Heredia-Langner, A., Carlyle, W. M., Montgomery, D. C., Borror, C. M., & Runger, G. C. (2003). Genetic algorithms for the construction of D-optimal designs. Journal of Quality Technology, 35(1), 28-46.
Hill, T., Lewicki, P., & Lewicki, P. (2006). Statistics:Methods and applications:a comprehensive reference for science, industry, and data mining. StatSoft.
He, X. (2010). Laplacian regularized D-optimal design for active learning and its application to image retrieval. IEEE Transactions on Image Processing, 19(1), 254-263.
Joseph, V. R., Gu, L., Ba, S., & Myers, W.R. (2018). Space-filling designs for robustness experiments. Technometrics, 61(1), 24-37.
Kackar, R.N., & Shoemaker, A.C. (1986). Robust design: A cost-effective method for improving manufacturing processes. AT&T Technical Journal, 65(2), 39-50.
Logothetidis, S., Georgiou, D., Laskarakis, A., Koidis,C., & Kalfagiannis,N. (2013). Solar Energy Materials and Solar Cells. 112(5), 144-156.
Li, Y. (2017). A Kriging-based global optimization method using multi-points infill search criterion. Journal of Algorithms & Computational Technology, 11(4), 366-377.
Mateu, J., Crujeiras, R. M., Menezes, R., & Montes, F. (2017). Spatio-temporal statistical methods in environmental and biometrical problems. Spatial Statistics, 22(2), 219-224.
Myers, R.H., & Carter, W.H. (1973). Response surface techniques for dual response systems. Technometrics, 15(2), 301-317.
Myers, R. H., & Anderson-Cook, C. M. (2009). Response surface methodology: process and product optimization using designed experiments. John Wiley & Sons.
Ozdemir, A. (2017). Development of the D-optimality-based coordinate-exchange algorithm for an irregular design space and the mixed-integer nonlinear robust parameter design optimization. PhD dissertation, 82-83.
Palavesam, N., Marin, S., Hemmetzberger, D., Landesberger, C., Bock, K., & Kutter, C. (2018). Roll-to-roll processing of film substrates for hybrid integrated flexible electronics. Flexible and Printed Electronics, 3(1).
Pander, A., Hatta, A.,& Furuta, H. (2016). Optimization of catalyst formation conditions for synthesis of carbon nanotubes using Taguchi method. Applied Surface Science, 371, 425-435.
Park, J., Shin. K. H., & Lee, C. (2015). Improvement of cross-machine directional thickness deviation for uniform pressure-sensitive adhesive layer in roll-to-roll slot-die coating process. International Journal of Precision Engineering and Manufacturing, 16(5), 937-943.
Rady, E. A., El-monsef, M. M. E. Abd., & Seyam, M. M. (2009). Relationships among several optimality criteria. InterStat, 15(6), 1-11.
Shin, D., Lee, S. M., & Yun, I. (2016). Effect of process parameters on hard coating film characteristics in roll-to-roll printing process system. International Conference on Electronics Packaging, 141-144.
Sutanto, E., & Alleyne, A. (2015). A semi-continuous Roll-to-Roll (R2R) electrohydrodynamic jet printing system. Mechatronics, 31(10), 243-254.
Tang, L. C., & K. Xu,K. (2002). A unified approach for dual response surface optimization. Journal of Quality Technology, 34(4), 437-447.
Wang, B., Fu, X., Song, S., Chu, H. O., Gibson, D., Li,C., Shi,Y., & Wu, Z. (2018). Simulation and optimization of film thickness uniformity in physical vapor deposition. coatings, 8(9), 325.
Weber, S., Klier, J., Ellrich, F., Paustian, S., Güttler, N., Tiedje, O., Jonuscheit, J., & Freymann,G. (2017). Thickness determination of wet coatings using self-calibration method. 42nd International Conference on Infrared, Millimeter, and Terahertz Waves (IRMMW-THz).
Zhou, X., Ma, Y., Tu, Y., & Feng, Y. (2013). Ensemble of Surrogates for dual response surface modeling in robust parameter design. Quality and Reliability Engineering International, 29(2), 173-197.