電子級異丙醇在半導體晶圓清洗製程中扮演關鍵角色，但其中殘留的丙酮可能導致污染，影響產品品質。為提升丙酮去除率並確保製程穩定性，本研究聚焦蒸餾製程參數（如蒸餾塔壓力、溫度等），整合實驗設計之田口方法、反應曲面法與資料探勘之M5P模式樹，建立系統性實驗框架，實現參數篩選、最佳化與預測。首先，透過變異數分析篩選歷史數據，識別顯著影響因子。接著，採用田口方法以直交表設計實驗，初步篩選穩健參數組合；隨後利用反應曲面法的中央合成設計建構二階模型，精細優化參數至丙酮濃度目標值。此外，採用M5P模式樹分析歷史數據，生成分段線性迴歸方程式，預測丙酮濃度數值，驗證優化結果的準確性。最終，透過實際代工廠的實驗，驗證所提出最佳參數設定的可行性與有效性，確保研究結果對實際製程具備應用價值。

Electronic-grade Isopropyl Alcohol plays a critical role in semiconductor wafer cleaning processes, but residual acetone can cause contamination, affecting product quality. To enhance acetone removal efficiency and ensure process stability, this research focuses on distillation process parameters. A systematic experimental framework is established, integrating the Taguchi method, Response Surface Methodology, and M5P model tree to achieve parameter screening, optimization, and prediction.
First, Analysis of Variance is employed to screen historical data and identify significant influencing factors. Next, the Taguchi method, utilizing orthogonal arrays, is applied for experimental design to initially screen for robust parameter combinations. Subsequently, RSM, specifically Central Composite Design, is used to construct a second-order model for fine-tuning parameters towards the target acetone content. M5P model tree is then applied to analyze historical data, generating piecewise linear regression equations to predict acetone content values and verify the accuracy of the optimization results. Finally, experimental validation is conducted at a contract manufacturing facility to confirm the feasibility and effectiveness of the proposed optimal parameter settings, ensuring the practical application value of the research findings.

李炘叡(2023)。射出成型能耗分析與優化策略。國立高雄科技大學模具工程系碩士班。
馬瑞菊, 林佩璇, 林俊男, 鄭婉如, 李佳欣, 蕭嘉瑩, & 蘇珉一(2020)。運用決策樹演算法於肝硬化重症病人死亡預測。Journal of Data Analysis，15(4)，1-14。https://doi.org/10.6338/jda.202008_15(4).0001
莊偉程(2023)。以田口方法改善顯影製程缺陷之研究。國立成功大學工程科學研究所在職專班。
莊憶萱(2024)。應用模式樹於提升標準成本訂價效率-以C公司為例。國立成功大學工程科學研究所在職專班。
陳昌鴻(2021)。以反應曲面法對學生方程式賽車前翼尾翼進行角度調整之最適化研究。國立臺北科技大學車輛工程系碩士班。
黃漢強(2022)。應用機器學習法預測旋轉電機發電量-機車發電機為例。國立陽明交通大學管理學院工業工程與管理學程碩士班。
鄭智中(2024)。應用實驗設計進行鹼性燃料電池膜電極組(MEA)製作參數最佳化---使用反應曲面法。國立高雄科技大學化學工程與材料工程系研究所碩士班。
Al-Dawalibi, A., Al-Dali, I. H., & Alkhayyal, B. A. (2020). Best marketing strategy selection using fractional factorial design with analytic hierarchy process. MethodsX, 7, 100927. https://doi.org/10.1016/j.mex.2020.100927
Biglete, E. R., Manuel, M. C. E., Cruz, J. C. D., Verdadero, M. S., Diesta, J. M. B., Miralpez, D. N. G., Javier, R. A. C., & Picato, J. I. C. (2020). Surface Roughness Analysis of 3D Printed Parts Using Response Surface Modeling. 2020 11th IEEE Control and System Graduate Research Colloquium (ICSGRC).
Bouyer, D., Moumouh, J., Benjelloun, S., Latifi, A., Khamar, L., & Aimar, P. (2023). Data-driven modeling and optimization of an industrial phosphoric acid production unit. MATEC Web of Conferences, 379. https://doi.org/10.1051/matecconf/202337907008
Box, G. E. P., & Hunter, J. S. (1957). Multi-Factor Experimental Designs for Exploring Response Surfaces. The Annals of Mathematical Statistics, 28(1), 195-241. http://www.jstor.org/stable/2237033
Cavazzuti, M., & Cavazzuti, M. (2013). Design of experiments. Optimization methods: from theory to design scientific and technological aspects in mechanics, 13-42.
Czitrom, V. (1999). One-factor-at-a-time versus designed experiments. The American Statistician, 53(2), 126-131.
Davis, R., & John, P. (2018). Application of Taguchi-Based Design of Experiments for Industrial Chemical Processes. In Statistical Approaches With Emphasis on Design of Experiments Applied to Chemical Processes. https://doi.org/10.5772/intechopen.69501
Delgarm, N., Sajadi, B., Azarbad, K., & Delgarm, S. (2018). Sensitivity analysis of building energy performance: A simulation-based approach using OFAT and variance-based sensitivity analysis methods. Journal of Building Engineering, 15, 181-193.
Fayyad, U., PiatetskyShapiro, G., & Smyth, P. (1996). From data mining to knowledge discovery in databases. Ai Magazine, 17(3), 37-54. <Go to ISI>://WOS:A1996VJ67100006
Frawley, W. J., Piatetskyshapiro, G., & Matheus, C. J. (1992). Knowledge Discovery in Databases - an Overview. Ai Magazine, 13(3), 57-70. <Go to ISI>://WOS:A1992JP64000015
Grömping, U. (2018). R package DoE. base for factorial experiments. Journal of Statistical Software, 85, 1-41.
Kechagias, J. D., Aslani, K.-E., Fountas, N. A., Vaxevanidis, N. M., & Manolakos, D. E. (2020). A comparative investigation of Taguchi and full factorial design for machinability prediction in turning of a titanium alloy. Measurement, 151. https://doi.org/10.1016/j.measurement.2019.107213
Khuri, A. I., & Mukhopadhyay, S. (2010). Response surface methodology. Wiley interdisciplinary reviews: Computational statistics, 2(2), 128-149.
Kiss, A. A. (2013). Distillation technology – still young and full of breakthrough opportunities. Journal of Chemical Technology & Biotechnology, 89(4), 479-498. https://doi.org/10.1002/jctb.4262
Kittur, A. A., Kulkarni, S. S., Aralaguppi, M. I., & Kariduraganavar, M. Y. (2005). Preparation and characterization of novel pervaporation membranes for the separation of water–isopropanol mixtures using chitosan and NaY zeolite. Journal of Membrane Science, 247(1), 75-86. https://doi.org/https://doi.org/10.1016/j.memsci.2004.09.010
Li, X.-X., Sun, Y.-F., & Zhao, G.-Y. (2019). Thermal Fatigue Life of Copper-Filled Laminated Micropore Based on Response Surface Methodology. 2019 International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering (QR2MSE).
Manmai, N., Unpaprom, Y., & Ramaraj, R. (2020). Bioethanol production from sunflower stalk: application of chemical and biological pretreatments by response surface methodology (RSM). Biomass Conversion and Biorefinery, 11(5), 1759-1773. https://doi.org/10.1007/s13399-020-00602-7
Mojaddam, M., & Pullen, K. R. (2019). Optimization of a centrifugal compressor using the design of experiment technique. Applied Sciences, 9(2), 291.
Montgomery, D. C. (2017). Design and analysis of experiments. John wiley & sons.
Quinlan, J. R. (1992). Learning With Continuous Classes.
Saha, S. P., & Ghosh, S. (2014). Optimization of xylanase production by Penicillium citrinum xym2 and application in saccharification of agro-residues. Biocatalysis and Agricultural Biotechnology, 3(4), 188-196.
Tang, Z., Xia, W., Li, F., Zhou, Z., & Zhao, J. (2010). Application of response surface methodology in the optimization of burnishing parameters for surface integrity. 2010 International Conference on Mechanic Automation and Control Engineering.
Vaziri, H., Hedayati Moghaddam, A., & Mirmohammadi, S. A. (2020). Optimization of distillation column in phenol production process for increasing the isopropyl benzene concentration using response surface methodology and radial basis function (RBF) coupled with leave-one-out validation method. Chemical Papers, 74(10), 3311-3324. https://doi.org/10.1007/s11696-020-01162-w