隨著面板製造技術日新月異與智慧製造，有效控制生產過程及時發現缺陷原因，以提升良率成為一大關鍵議題。在面板製造過程中，若機台或製程參數異常，會造成面板缺陷以致良率下降，因此在製程中均會進行線上檢測以辨識各缺陷，藉由觀察其缺陷分布圖上缺陷的群聚現象，工程師可憑自身經驗去判斷製程異常的原因。實務上憑工程師以人工目視的方式辨別缺陷分布圖群聚特徵，不僅需花費許多時間及人力成本，人為主觀因素亦會造成判斷結果之差異，以致無法有效率排除異常原因，進而造成良率損失。然而過去文獻少有針對面板缺陷分布圖之群聚特徵進行探討，亦尚未有文獻探討此特性對於異常成因搜尋的助益。
本研究針對線缺陷發展一套面板缺陷分布圖之資料挖礦分析架構，以有效辨識缺陷分布圖之群聚特徵，先利用連檢定(Run Test) (Wald and Wolfowitz, 1943)將缺陷分布圖分為隨機、群聚兩種類別，將有群聚現象發生的缺陷分布圖透過變化點分析(Change Point Analysis)(Gottman, 1981)判斷缺陷群聚之位置，最後利用最小絕對壓縮挑選機制（Least Absolute Shrinkage and Selection Operator，Lasso）(Tibshirani, 1996)尋找異常成因。本研究以台灣某指標性面板廠為實證研究對象，藉由本研究方法判斷其缺陷分布圖特性以及搜尋異常成因，透過訂定效度檢驗指標以驗證本研究結果之有效性，驗證結果指出本研究所提出之面板缺陷分布圖之資料挖礦分析架構，可以有效辨識缺陷分布圖之群聚特徵以及造成異常之原因，以協助公司導入智慧製造。

With the rapid development of the TFT-LCD manufacturing, in order to maintain competitiveness, effectively control of the production process to enhance the yield has become the key issue for panel factories. To ensure the assignable root cause of the yeild loss, inline inspection system has applied to identify defects on the panel. Engineers rely on the information from the inline inspection system for trouble shooting, one of most effective way is to analyze the clustering pattern of defect map. Defect map can provide important rules for engineers to find the root cause by identifying patterns correctly. Nowadays, most companies still rely on engineers’ experiences of visual inspections and personal judgments in the map patterns. This manual approach is not only subjective, lack of justice and consistent standard, but also very time consuming and inefficient.
This study proposes a data mining framework of line defects for identifying clustering pattern of the defect map and the root cause of the clustering pattern of defect map. First, critical defect map which possesses potential clustering pattern is testing for run test to classified into Random and Clustered. For the clustered map, applying change point analysis to determine the position of the defect cluster, and find the root cause of each clustering pattern by LASSO. To examine the validity of this approach, an empirical study was conducted in a TFT-LCD company in Taiwan. The results show that the framework can identify the clustering pattern of the defect map and the root cause effectively.

李佶玟(2012)，「應用空間相關圖於半導體晶圓圖樣型分類」，清華大學工業工程與工程管理研究所碩士論文。
Chen, L. C., & Kuo, C. C. (2007), “Automatic TFT-LCD mura defect inspection using discrete cosine transform-based background filtering and ‘just noticeable difference’quantification strategies.” Measurement Science and Technology, vol. 19, No. 1.
Chen, S.L., Chou, S.T. (2008), “TFT-LCD Mura defect detection using wavelet and cosine transforms.” Journal of advanced mechanical design, systems, and manufacturing, Vol. 2, No. 3, pp. 441-453.
Cheng, F. T., Hsieh, Y. S., Zheng, J. W., Chen, S. M., Xiao, R. X., & Lin, C. Y. (2017), “A Scheme of High-Dimensional Key-Variable Search Algorithms for Yield Improvement.” IEEE Robotics and Automation Letters, Vol. 2. No. 1, pp. 179-186.
Chien, C. F., Hsu, S. C., & Chen, Y. J. (2013), “A system for online detection and classification of wafer bin map defect patterns for manufacturing intelligence.” International Journal of Production Research, Vol. 51, No. 8, pp. 2324-2338.
Gaillet, M., Yan, L., & Teboul, E. (2007), “Optical characterizations of complete TFT–LCD display devices by phase modulated spectroscopic ellipsometry.” Thin solid films, Vol. 516, No. 2, pp. 170-174.
Gottman, J. M. (1981), “Time-series analysis.” Cambridge: CUP.
Hansen, M. H., Nair, V. N., & Friedman, D. J. (1997), “Monitoring wafer map data from integrated circuit fabrication processes for spatially clustered defects.” Technometrics, Vol. 39, No. 3, pp. 241-253.
Hsieh, K. L. (2008), “The application of clustering analysis for the critical areas on TFT-LCD panel.” Expert Systems with Applications, Vol. 34, No. 2, pp. 952-957.
48
Hsu, Chia-Yu, et al. (2010), “Data mining for yield enhancement in TFT-LCD manufacturing: an empirical study.” Journal of the Chinese Institute of Industrial Engineers, Vol. 27, No. 2, pp. 140-156.
Hwang, J. Y., & Kuo, W. (2007), “Model-based clustering for integrated circuit yield enhancement.” European Journal of Operational Research, Vol. 178, No. 1, pp. 143-153.
Jeong, Y. S., Kim, S. J., & Jeong, M. K. (2008), “Automatic identification of defect patterns in semiconductor wafer maps using spatial correlogram and dynamic time warping.” IEEE Transactions on Semiconductor manufacturing, Vol. 21, No. 4, pp. 625-637.
Kaempf, U. (1995), “The binomial test: A simple tool to identify process problems.” IEEE Transactions on semiconductor manufacturing, Vol. 8, No. 2, pp. 160-166.
Kuo, C. F., Hsu, C. T. M., Fang, C. H., Chao, S. M., & Lin, Y. D. (2013), “Automatic defect inspection system of colour filters using Taguchi-based neural network.” International Journal of Production Research, Vol. 51, No. 5, pp. 1464-1476.
Li, W. C., & Tsai, D. M. (2011), “Defect inspection in low-contrast LCD images using Hough transform-based nonstationary line detection.” IEEE Transactions on industrial informatics, Vol. 7, No. 1, pp. 136-147.
Lee, J. Y., & Yoo, S. I. (2004), “Automatic detection of region-mura defect in TFT-LCD.” IEICE TRANSACTIONS on Information and Systems, Vol. 87, No. 10, pp. 2371-2378.
Lim, D., & Jeong, D. H. (2007), “Zone-based inspection and defect classification for LCD manufacturing: Trivial defect free procedure for TFT glass inspection.” International Journal of Optomechatronics, Vol. 1, No. 3, pp. 312-330.
Liu, Y. H., Huang, Y. K., & Lee, M. J. (2008), “Automatic inline defect detection for a thin film transistor–liquid crystal display array process using locally linear embedding and support vector data description.” Measurement Science and Technology, Vol. 19, No. 9.
Liu, Y. H., Lin, S. H., Hsueh, Y. L., & Lee, M. J. (2009), “Automatic target defect identification for TFT-LCD array process inspection using kernel FCM-based
49
fuzzy SVDD ensemble.” Expert Systems with Applications, Vol. 36, No. 2, pp. 1978-1998.
Liu, Y. H., Wang, C. K., Ting, Y., Lin, W. Z., Kang, Z. H., Chen, C. S., & Hwang, J. S. (2009), “In-TFT-array-process micro defect inspection using nonlinear principal component analysis.” International journal of molecular sciences, Vol. 10, No. 10, pp. 4498-4514.
Liu, Y. H., Liu, Y. C., & Chen, Y. Z. (2011), “High-speed inline defect detection for TFT-LCD array process using a novel support vector data description.” Expert systems with applications, Vol. 38, No. 5, pp. 6222-6231.
Park, N. K., & Yoo, S. I. (2009), “Evaluation of TFT-LCD defects based on human visual perception.” Displays, Vol. 30, No. 1, pp. 1-16.
Song, Y. C., Choi, D. H., & Park, K. H. (2004), “Multiscale detection of defect in thin film transistor liquid crystal display panel.” Japanese journal of applied physics, Vol. 43, No. 8R.
Taam, W., & Hamada, M. (1993), “Detecting spatial effects from factorial experiments: An application from integrated-circuit manufacturing.” Technometrics, Vol, 35, No. 2, pp. 149-160.
Tsai, D. M., Lin, P. C., & Lu, C. J. (2006), “An independent component analysis-based filter design for defect detection in low-contrast surface images.” Pattern Recognition, Vol. 39, No. 9, pp. 1679-1694.
Wald, A. B. R. A. H. A. M., & Wolfowitz, J. A. C. O. B. (1943), “An exact test for randomness in the non-parametric case based on serial correlation.” The Annals of Mathematical Statistics, Vol. 14, No. 4, pp. 378-388.
Wang, C. H. (2008), “Recognition of semiconductor defect patterns using spatial filtering and spectral clustering.” Expert Systems with Applications, Vol. 34, No. 3, pp. 1914-1923.
Wang, Y. C., & Lin, B. S. (2013), “Critical point defect detection for small‐pixel thin‐film transistor array applied to medical display.” Journal of the Society for Information Display, Vol. 21, No. 9, pp. 376-380.
50
Zhang, Y., & Zhang, J. (2005), “A fuzzy neural network approach for quantitative evaluation of mura in TFT-LCD.” Neural Networks and Brain, ICNN&B'05. International Conference on IEEE, Vol. 1, pp. 424-427