在實際生產線上，最在乎的就是生產良率，因為產品良率會直接影響生產成本。因此生產者在產品研發階段以及量產階段遇到良率不佳時，期望透過生產過程中的資訊來進行良率分析，找出影響良率之根本原因並加以改善，達到快速提升產品良率之目的，避免不必要的成本浪費。
傳統良率分析之方法是透過長時間的歷史資料以及專家經驗，來找出影響良率的原因。隨著工廠規模以及製程複雜度日益上升，所收集的各式資料勢必為一個高維度且巨量之資料，對於傳統良率分析方法勢必會耗費相當多的時間。因此，本研究團隊提出了關鍵參數搜尋演算法 (Key-variable Search Algorithm, KSA)，用以解決上述所提及之問題。
在整個生產線上，會收集到各式各樣的資料 (如生產路徑資料、生產製程資料…等)。鑒於全自動虛擬量測系統 (Automatic Virtual Metrology, AVM) 在進行分析前，針對不同的資料進行資料品質之把關。基於相同理念，本論文針對進入KSA架構中的各式資料，提出適用於關鍵參數搜尋架構之各式資料品質評估機制來進行資料品質的把關，使KSA架構更為完善，避免KSA之分析結果受到資料品質影響。

Yield is the most important issue on actual production lines because it will directly affect the production cost. Therefore, when encountering a poor yield in the R&D stage and the mass production stage, the manufacturers expect to do the yield analysis with various manufacturing-related data for finding the root causes of the yield loss and improving the yield to avoid unnecessary cost.
The traditional approach of yield analysis is to find the causes of the yield loss by historical data and expert knowledge. As the expansion of the factory scales and the complexity of the manufacturing processes increase, all kinds of the collected data would certainly become big and high-dimensional data, which means that it would take a lot of time to analyze data with the traditional yield analysis approach. Therefore, our research team proposes Key-variable Search Algorithms (KSA) to solve the problems mentioned above.
Various kinds of data such as route data, process data, etc., would be collected throughout the production lines. Based on Automatic Virtual Metrology (AVM), which includes data-quality-evaluation module that can perform data quality check before analyzing/processing, various data-quality-evaluation algorithms are proposed for all kinds of data analyzed by the KSA scheme in this paper. These evaluation algorithms enable the KSA scheme to work better and avoid its performance being affected by bad data quality.

[1] F.-T. Cheng, Y.-S. Hsieh, J.-W. Zheng, S.-M. Chen, R.-X. Xiao, and C.-Y. Lin, “A Scheme of High-dimensional Key-variable Search Algorithms for Yield Improvement,” IEEE Robotics and Automation Letters, vol. 2, no. 1, pp. 179-186, Jan. 2017
[2] John S. Oakland, Statistical Process Control, 6th ed., Oxford, UK : Elsevier, 2007
[3] G. Spitzlsperger, C. Schmidt, G. Ernst, H. Strasser, and M. Speil, “Fault Detection for a Via Etch Process Using Adaptive Multivariate Methods,” IEEE Transactions on Semiconductor Manufacturing, vol. 18, no. 4, pp.528-533, Nov. 2005
[4] C. J. Spanos, H. F. Guo, A. Miller, and J. Levine-Parrill, “Real-Time Statistical Process Control Using Tool Data,” IEEE Transactions on Semiconductor Manufacturing, col. 5, no. 4, pp. 308-318, Nov. 1992
[5] Carter, Russell, and Jay Lubinsky. Rehabilitation research: principles and applications. Elsevier Health Sciences, 2015.
[6] F.-T. Cheng, H.-C. Huang, and C.-A. Kao, “Developing an Automatic Virtual Metrology System,” IEEE Trans. on Automation Science and Engineering, vol. 9, no. 1, pp. 181-188, Jan. 2012.
[7] Y.-T. Huang and F.-T. Cheng, “Automatic data quality evaluation for the AVM system,” IEEE Transactions on Semiconductor Manufacturing, vol. 24, no. 3, pp.445-454, August 2011.
[8] Y.-T. Huang, F.-T. Cheng, Y.-H. Shih and Y.-L. Chen, “Advanced ART2 scheme for enhancing metrology-data-quality evaluation,” J. Chin. Inst. Eng., vol. 37, no. 8, pp. 1064-1079, 2014.
[9] F.-T. Cheng, C.-A. Kao, C.-F. Chen; W.-H. Tsai, “ Tutorial on Applying the VM Technology for TFT-LCD Manufacturing,” IEEE Transactions on Semiconductor Manufacturing, vol. 28, no. 1, pp. 55-69, 2015.
[10] J. Dattorro, Convex Optimization and Euclidean Distance Geometry, Palo Alto, CA: Meboo, 2005.
[11] I. H. Witten and E. Frank, Data Mining: Practical Machine Learning Tools and Techniques, San Francisco, CA: Morgan Kaufman, 2005.
[12] Robert McGill, John W. Tukey and Wayne A. Larsen, “Variations of Box Plots,” The American Statistician, Vol. 32, No. 1, pp. 12-16, 1978.
[13] J. A. Hartigan and M. A. Wong, “Algorithm AS 136: A K-Means Clustering Algorithm,” Journal of the Royal Statistical Society. Series C (Applied Statistics), vol. 28, no. 1, pp. 100-108, 1979.
[14] André Fujita, Daniel Y. Takahashi, Alexandre G. Patriota, “A non-parametric method to estimate the number of clusters”, Computational Statistics and Data Analysis, vol. 73, pp. 27-39, 2014.
[15] Peter J. Rousseeuw, “Silhouettes: A graphical aid to the interpretation and validation of cluster analysis,” Journal of Computational and Applied Mathematics, vol. 20, pp. 53-65, 1987.
[16] 田民波，林怡欣校訂，TFT LCD面板設計與構裝技術，五南圖書出版社，2010，第七章。
[17] 蕭宏，半導體製程技術與導論，第三版，全華出版社，2014。