在晶圓生產過程中，瓶頸機台的存在經常是不可避免的，這些機台的產能更直接地影響著產品的交付時間，如何有效的保持機台的產量於一定水準，一直是製造領域中重要的議題，透過深入了解影響產出的諸多因素，從而進行充分的分析和改進，使這些因素的影響能夠被更加精確地控制和調整。半導體製造領域中，經常是以即時性的每小時產出量(Wafer Per Hour，WPH)來監測機台產出是否如預期，因此若能透過辨別管制圖產量異常趨勢，快速且正確地發現異常現象並修正問題，勢必能降低晶圓生產過程所產生的資源浪費，達到生產最大化的效益。故本研究將透過WPH產量管制圖圖形來進行辨識分類，區分出產量異常與正常的類別，並針對各類別資料以特徵萃取取得各項特徵，最後透過解讀決策樹分類方法與RIPPER演算法分類規則，找到影響產量的共同關鍵因素。
本研究通過分析決策樹歸納法和RIPPER演算法在混淆矩陣中的各項評估指標，確定了RIPPER演算法在識別產量異常方面優於決策樹歸納法。研究結果顯示，決策樹歸納法和RIPPER都選擇了偏度(SKEW)、向下累積和(cumulative down)、條件分段線之平均斜率(AS)及最小平方線斜率之絕對值(AB)等特徵作為關鍵因子，透過深入了解這些關鍵因子的變化和影響，能夠為晶圓生產提供更多的實證支持。這些研究成果不僅有助於工程師排除生產條件變異、機台故障和環境變化的影響，也為未來的研究提供了重要的啟示和參考，從而進一步提升了機台管理的科學性和效率。

In semiconductor wafer production, the presence of bottleneck machines is often unavoidable, and these machines directly impact product delivery times. Maintaining consistent machine output is a crucial issue in the wafer production. By thoroughly understanding the influence of each manufacturing factor, comprehensive analysis and improvements can be made to more precisely control and adjust the factors. In the semiconductor manufacturing field, real-time Wafer Per Hour (WPH) monitoring is commonly used to ensure that machine output meets expectations. Identifying and correcting anomalies through control chart patterns can significantly reduce resource waste during wafer production, thereby maximizing production efficiency. This study employs WPH control chart patterns to identify whether production output is normal or abnormal. Feature extraction is first applied on each control chart to obtain attributes for classification. Subsequently, the common key factors affecting output are identified from the learning results of decision tree induction and RIPPER algorithm. Through the evaluation of various metrics in the confusion matrix, the RIPPER algorithm outperforms the decision tree classification method in identifying abnormal patterns. The experimental results indicate that the common attributes identified by decision tree induction and RIPPER algorithm are skewness, cumulative down, average slope of conditional segmented lines, and the absolute value of the least squares line slope. Understanding the variations and impacts of these key factors provides substantial empirical support for wafer production. These findings not only assist engineers in mitigating the effects of production condition variations, machine failures, and environmental changes but also offer valuable insights and references for future research.

Addeh, A., Khormali, A., & Golilarz, N. A. (2018). Control chart pattern recognition using RBF neural network with new training algorithm and practical features. ISA Transactions, 79, 202-216.

Addeh, J., Ebrahimzadeh, A., Azarbad, M., & Ranaee, V. (2014). Statistical process control using optimized neural networks: A case study. ISA Transactions, 53(5), 1489-1499.

Addeh, J., Ebrahimzadeh, A., & Nazaryan, H. (2013). A research about pattern recognition of control chart using optimized ANFIS and selected features. Journal of Engineering & Technology, 3(1), 6-16.

Addeh, J., Ebrahimzadeh, A., & Ranaee, V. (2011). Control chart pattern recognition using adaptive back-propagation artificial Neural networks and efficient features.

Breiman, L., Friedman, J., Olshen, R., & Stone, C. (1984) Classification and Regression Trees. Chapman and Hall.

Cheng, C.-S., Huang, K.-K., & Chen, P.-W. (2015). Recognition of control chart patterns using a neural network-based pattern recognizer with features extracted from correlation analysis. Pattern Analysis and Applications, 18(1), 75-86.

Cheng, C.-S., Ho, Y., & Chen, P.-W. (2022). Control Chart Concurrent Pattern Classification Using Multi-Label Convolutional Neural Networks.

Gauri, S. K. & Chakraborty, S. (2007). A study on the various features for effective control chart pattern recognition. The International Journal of Advanced Manufacturing Technology, 34(3), 385-398.

Gauri, S. K. & Chakraborty, S. (2009). Recognition of control chart patterns using improved selection of features. Computers & Industrial Engineering, 56(4), 1577-1588.

Guh, R.-S. & Tannock, J. D. T. (1999). Recognition of control chart concurrent patterns using a neural network approach. International Journal of Production Research, 37(8), 1743-1765.

Guh, R.-S. (2005). A hybrid learning-based model for on-line detection and analysis of control chart patterns. Computers & Industrial Engineering, 49(1), 35-62.
Guh, R.-S.,Zorriassatine, F., Tannock, J. D. T., & O'Brien, C. (1999). On-line control chart pattern detection and discrimination—a neural network approach. Artificial Intelligence in Engineering, 13(4), 413-425.

Guh, R.-S.& Li, M.-H. (2018). An artificial neural network-based classifier ensemble approach for on-line recognition of concurrent control chart patterns. Journal of Quality, 25(1), 1-28.

Hassan, A., Baksh, M. S. N., Shaharoun, A. M., & Jamaluddin, H. (2003). Improved SPC chart pattern recognition using statistical features. International Journal of Production Research, 41(7), 1587-1603.

Kass, G.V. (1980) An Exploratory Technique for Investigating Large Quantities of Categorical Data.

Lesany, S. A., Koochakzadeh, A., & Fatemi Ghomi, S. M. T. (2014). Recognition and classification of single and concurrent unnatural patterns in control charts via neural networks and fitted line of samples. International Journal of Production Research, 52(6), 1771-1786.

Lu, C.-J., Shao, Y. E., & Li, C.-C. (2014). Recognition of concurrent control chart patterns by integrating ICA and SVM. Applied Mathematics & Information Sciences, 8(2), 681-689.

Michalski et al. (1986). The Multi-Purpose Incremental Learning System AQ15 and Its Testing Application to Three Medical Domains.

Montgomery, D. C. (2009). Introduction to Statistical Quality Control, Sixth Edition. John Wiley & Sons.

Pang & Kao. (2006). Scaling Analysis and Systematic Extraction of Macroscopic Structures in Fluctuating Systems of Arbitrary Dimensions.

Pham, D. T.& Wani, M. A. (1997). Feature-based control chart pattern recognition. International Journal of Production Research, 35(7), 1875-1890.

Quinlan, J.R. (1986) Induction of Decision Trees. Machine Learning, 1, 81-106.

Ranaee, V.& Ebrahimzadeh, A. (2013). Control chart pattern recognition using neural networks and efficient features: a comparative study. Pattern Analysis and Applications, 16(3), 321-332.

Ranaee, V., Ebrahimzadeh, A., & Ghaderi, R. (2010). Application of the PSO–SVM model for recognition of control chart patterns. ISA Transactions, 49(4), 577-586.

Wu, C., Liu, F., & Zhu, B. (2015). Control chart pattern recognition using an integrated model based on binary-tree support vector machine. International Journal of Production Research, 53(7), 2026-2040.

Xanthopoulos, P. & Razzaghi, T. (2014). A weighted support vector machine method for control chart pattern recognition. Computers & Industrial Engineering, 70, 134-149.

Xie, L., Gu, N., Li, D., Cao, Z., Tan, M., & Nahavandi, S. (2013). Concurrent control chart patterns recognition with singular spectrum analysis and support vector machine.Computers & Industrial Engineering, 64(1), 280-289.

Yang, W.-A., Zhou, W., Liao, W., & Guo, Y. (2015). Identification and quantification of concurrent control chart patterns using extreme-point symmetric mode decomposition and extreme learning machines. Neurocomputing, 147, 260-270.