薄膜電晶體液晶顯示器(Thin Film Transistor-Liquid Crystal Display, TFT-LCD)是電子產品中關鍵的零組件，也是臺灣重要的電子代工業，在中國大陸的產能擴展與韓國大廠新科技發展的夾擊之下，臺灣的TFT-LCD產業在品質控制以及產能提升的能力需要加快腳步。近年來高畫質的螢幕產品已經成為消費者的首選條件，而高畫質的產品對於濕蝕刻製程中關鍵尺寸損失量(Critical Dimension Loss，CD Loss)的掌控能力是一大考驗，如何將CD Loss控制在一個合理的規格內是目前濕蝕刻製程的重要挑戰。
本研究收集濕蝕刻製程時的生產數據，使用部分最小平方迴歸法來建立CD Loss量測值的預測模型，以提供製程工程師在進行製程改善時的工具，以期減少找尋製程因子的時間成本和實驗成本，並且透過預測模型可以得到CD Loss量測值的估計值，作為在得到實際量測值之前進行初步檢驗。部分最小平方迴歸法是找到同時能解釋自變數與應變數的潛在變數(Latent Variable)，並利用潛在變數建立由自變數預測應變數的迴歸模型，部分最小平方迴歸法是比多元線性迴歸和主成分迴歸的更佳分析方法，因為它提供了更加穩定的迴歸模型。本研究是使用TFT製程中濕蝕刻製程的生產數據，來預測銅製程中蝕刻後關鍵尺寸損失量的量測值，由已知的影響因子中找到可供預測CD Loss量測值的迴歸方程式，並由迴歸方程式的迴歸係數得知蝕刻液中的銅離子濃度以及整體的蝕刻時間是影響蝕刻後關鍵尺寸損失量的重要影響因子。

Thin Film Transistor-Liquid Crystal Display (TFT-LCD) is the key component in electronic products. In recent years, high-resolution screen products have become the preferred condition for consumers, and high- resolution products are hard to control the Critical Dimension Loss (CD Loss) in the wet etching process. How to control the CD Loss at reasonable specification is currently a crucial challenge in the wet etching process. Using Partial Least Squares Regression to set up a predictive model of CD Loss measurements, and the estimated value of the CD Loss measurement can be obtained through the predictive model. Partial Least Squares Regression is to research the latent variables that would be explained the independent variables and predict dependent variables at the same time, and that would be using the latent variables to set up a regression model for forecasting the responses from the independent variables.
In this study, the wet etching production data in the TFT process was used to predict the measurement of the CD Loss in the copper process, and therefore we knew the copper ion concentration in the etching solution and overall etching time are the important factors of CD Loss, that were obtained from the regression coefficient of the regression equation.

朱柏豪 (2017)。什麼是蝕刻(Etching)。國家奈米元件實驗室奈米通訊，24(2)，29-31。
呂仁雄 (2005)。銅/鉬酸性蝕刻液之研究。國立臺灣大學化學工程學研究所碩士論文，臺北市。
邱琬雯 (2017)。全球面板產業投資概況及發展趨勢。機械工業技術與產業資訊專輯，406，45-56。
侯傑泰 (1994 )。為何需要結構方程模式及如何建立潛伏變項。教育研究舉報，9(1)，87-92。
袁鶴齡 (2006)。我國 TFT-LCD 產業發展模式之探討。科技發展政策報導，2，119-135。
張丘坤 (2010)。半導體廠附屬設備管路區環境監測改善研究-以CVD區為例。國立交通大學工學院碩士在職專班產業安全與防災組碩士論文，新竹市。
郭俊賢 (2004)。銅在酸性溶液之蝕刻研究。國立臺灣大學化學工程學研究所碩士論文，臺北市。
陳泓翰. (2010)。液晶顯示器陣列/彩色濾光片製程大面積光阻塗佈狹縫式噴嘴國產化之測試研究。國立交通大學工學院精密與自動化工程碩士論文，新竹市。
陳順宇 (2005)。多變量分析(四版)。臺北市：華泰書局。
陳順宇 (2009)。迴歸分析: SPSS、STATISTICA軟體操作(四版)。臺北市：三民書局。
湯家偉 (2016)。結構方程模式：偏最小平方法PLS-SEM(初版)。臺北市：高等教育文化事業有限公司。
楊雲凱 (2015)。物理氣相沉積(PVD)介紹。國家奈米元件實驗室奈米通訊，22(4)，33-35。
龍柏華 (2003)。濕蝕刻製程介紹暨機台原理簡介。光連雙月刊：光電產業與技術情報，48，37-41。
龍柏華&何教恆 (2004)。乾蝕刻製程暨機台原理簡介。光連雙月刊：光電產業與技術情報，53，34-41。
謝雨生&鄭宜仲 (1993)。多元迴歸分析的假定與實例檢討－多元線性重合現象的診斷。農業推廣學報，10，189-213。
Barclay , D., Higgins , C., & Thompson , R. (1995). The Partial Least Squares (PLS) approach to causal modeling: personal computer use as an illustration. Technology Studies, 2, 285 – 309 .
Bottle, A., Aylin, P., & Majeed, A. (2006). Identifying patients at high risk of emergency hospital admissions: a logistic regression analysis. Journal of the Royal Society of Medicine, 99(8), 406-414.
Bro, R., & Smilde, A. K. (2014). Principal component analysis. Analytical Methods, 6(9), 2812-2831.
de Souza, A., Ikefut, P. V., Garcia, A. P., Santos, D. A. S., & de Oliveira, S. S. (2018). Analysis of the relationship between O3, NO and NO2 using multiple linear regression techniques. Geographia-Uff, 20(43), 124-135.
Gardner, R. M., Bieker, J., & Elwell, S. (2000, 12-14 Sept. 2000). Solving tough semiconductor manufacturing problems using data mining. Paper presented at the 2000 IEEE/SEMI Advanced Semiconductor Manufacturing Conference and Workshop. ASMC 2000 (Cat. No.00CH37072).
Geladi , P. (1988). Notes on the history and nature of partial least squares (PLS) modeling. Journal of Chemometrics, 2(4), 231-246.
Graham, M. H. (2003). Confronting multicollinearity in ecological multiple regression. Ecology, 84(11), 2809-2815.
Hsieh, K.-L. (2010). Incorporating ANNs and statistical techniques into achieving process analysis in TFT-LCD manufacturing industry. Robotics and Computer-Integrated Manufacturing, 26(1), 92-99.
Jakes, W., Gerdova, A., Defernez, M., Watson, A. D., McCallum, C., Limer, E., . . . Kemsley, E. K. (2015). Authentication of beef versus horse meat using 60 MHz H-1 NMR spectroscopy. Food Chemistry, 175, 1-9.
Kao, W.-H., Chang, Y.-C., & Lin, C.-C. (2011). The module design for critical dimension bias correction of slanted line on high level large area mask. Paper presented at the 2011 Fourth International Conference on Ubi-Media Computing.
Knofczynski, G. T., & Mundfrom, D. (2008). Sample sizes when using multiple linear regression for prediction. Educational and Psychological Measurement, 68(3), 431-442.
Li, D.-C., Chen, W.-C., Liu, C.-W., Chang, C.-J., & Chen, C.-C. (2012). Determining manufacturing parameters to suppress system variance using linear and non-linear models. Expert Systems with Applications, 39(4), 4020-4025.
Lim, J., Kim, G., Mo, C., Kim, M. S., Chao, K., Qin, J., . . . Cho, B. K. (2016). Detection of melamine in milk powders using near-infrared hyperspectral imaging combined with regression coefficient of partial least square regression model. Talanta, 151, 183-191.
Lin, C.-S., Shih, S.-J., Lu, A.-T., Hung, S.-S., & Chiu, C.-C. (2012). The quality improvement of PI coating process of TFT-LCD panels with Taguchi methods. Optik, 123(8), 703-710.
Liu, D., Qin, G., Cai, W. C., Wang, B. Q., Zhou, Y., Rao, Y., . . . Lv, J. J. (2017). Profile and critical dimension bias of Mo/Al/Mo electrode. Chinese Journal of Liquid Crystals and Displays, 32(11), 877-885.
Lynn, S. A., Ringwood, J., & MacGearailt, N. (2012). Global and local virtual metrology models for a plasma etch process. IEEE Transactions on Semiconductor Manufacturing, 25(1), 94-103.
Mateos-Aparicio, G. (2011). Partial Least Squares (PLS) methods: Origins, evolution, and application to social sciences. Communications in Statistics-Theory and Methods, 40(13), 2305-2317.
Nunes, C. A., Alvarenga, V. O., de Souza Sant'Ana, A., Santos, J. S., & Granato, D. (2015). The use of statistical software in food science and technology: Advantages, limitations and misuses. Food Res Int, 75, 270-280.
Pirouz, D. M. (2006). An Overview of Partial Least Squares. SSRN Electronic Journal.
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational and Behavioral Statistics, 31(4), 437-448.
Purwins, H., Barak, B., Nagi, A., Engel, R., Höckele, U., Kyek, A., . . . Weinzierl, K. (2014). Regression Methods for Virtual Metrology of Layer Thickness in Chemical Vapor Deposition. IEEE/ASME Transactions on Mechatronics, 19(1), 1-8.
Sainani, K. L. (2014). Logistic regression. PM R, 6(12), 1157-1162.
Saurina, J. (2010). Characterization of wines using compositional profiles and chemometrics. Trac-Trends in Analytical Chemistry, 29(3), 234-245.
Stott, A. E., Kanna, S., Mandic, D. P., & Pike, W. T. (2017, 5-9 March 2017). An online NIPALS algorithm for Partial Least Squares. Paper presented at the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).
Svante Wold, Michael Sjostrom, & Eriksson, L. (2001). PLS-regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratory Systems, 58(2), 109-130.
Tripepi, G., Jager, K. J., Dekker, F. W., & Zoccali, C. (2008). Linear and logistic regression analysis. Kidney Int, 73(7), 806-810.
Wang, G. F., Guo, Z. W., & Qian, L. (2014). Tool wear prediction considering uncovered data based on partial least square regression. Journal of Mechanical Science and Technology, 28(1), 317-322.
Whitman, C., Gilbert, T. M., Rahn, A. M., & Antonell, J. A. (2006). Determining factors affecting ESD failure voltage using DOE. Microelectronics Reliability, 46(8), 1228-1237.
Wold, H. (1974). Causal flows with latent variables - Partings of ways in light of NIPALS modelling. European Economic Review, 5(1), 67-86.
Yousefian-Jazi, A., Ryu, J. H., Yoon, S., & Liu, J. J. (2014). Decision support in machine vision system for monitoring of TFT-LCD glass substrates manufacturing. Journal of Process Control, 24(6), 1015-1023.