批次控制 (run-to-run control) 在半導體生產製程中扮演極重要的製程監控角色。當生產製程的投入變數對產出變數具有遷移效應 (carry-over effect) 的 transfer function (TF) model 時，文獻上有quasi MMSE控制器被提出來監控此類動態生產製程，唯此控制器僅在製程干擾項服從 IMA(1,1) 模型下被建構出來。因此當製程干擾項服從一般化的 ARIMA(p,1,q) 模型時，其製程的穩定條件及total mean square error (TMSE) 之表現並不甚理想。針對此缺點，本論文提出修正的quasi MMSE動態回饋控制器，並分別推導出此控制器下的製程產出公式、製程穩定條件及製程產出之漸近分配。最後，當實際製程為二階動態模型 TF(2,2) 且其干擾項服從 ARIMA(2,1,2) 時，本文亦探討此控制器與現存控制器之績效表現；結果顯示，本文提出的控制器之優點是不僅能將製程產出值更快速地調整到目標值，同時亦可使製程的TMSE大幅降低。

Run-to-run (R2R) control plays a vital role in monitoring or adjusting the integrated-circuit (IC) manufacturing process. Recently, a dynamic quasi MMSE (qMMSE) controller has been proposed in literature to address R2R problem when the process dynamics and disturbance dynamics exit simultaneously. The controller, however, was constructed only based on a transfer function (TF) model with the process disturbance follows an IMA(1,1) model. Therefore, its control performances on stability conditions and total mean square error (TMSE) may not perform satisfactorily when the true process disturbance follows a general ARIMA(p,1,q) model. To overcome this difficulty, this article first proposed a modified dynamic quasi MMSE (qMMSE) controller and then, the corresponding long-term stability conditions and its limiting distribution are derived. Furthermore, when the process I-O model follows a second-order transfer function (TF) model with the ARIMA(2,1,2) disturbance, we also compared the control performance of our proposed controller with that of existing controllers by using a simulation study. The results demonstrates that using wrong controllers usually cause serious effects on the performances of stability conditions and total mean square error (TMSE).

[1] Box, G. E. P., Jenkins, G. M., and Reinsel, G. C. (1994). Time Series Analysis: Forecasting and Control, 3rd ed. Prentice Hall, Englewood Clifs, NJ.
[2] Butler, S. W. and Stefani, J. A. (1994). Supervisory run-to-run control of polysilicon gate etch using in situ ellipsometry. Semiconductor Manufacturing, IEEE Transactions on Semiconductor Manufacturing, 7(2), 193-201.
[3] Capilla, C., Ferrer, A., Romero, R., and Hualda, A. (1999). Integration of statistical and engineering process control in a continuous polymerization process, Technometrics, 41, 14-28.
[4] Del Castillo, E. (2001). Some properties of EWMA feedback quality adjustment schemes for drifting disturbances. Journal of Quality Technology, 33, 153-166.
[5] Del Castillo, E. (2002). Statistical Process Adjustment for Quality Control, Wiley Interscience.
[6] Fan, S. K. S., Jiang B. C., Jen, C. H., and Wang, C. C. (2002). SISO run-to-run feedback controller using triple EWMA smoothing for semiconductor manufacturing process. International Journal of Production Research, 40, 3093-3120.
[7] Ingolfsson, A. and Sachs, E. (1993). Stability and sensitivity of an EWMA controller. Journal of Quality Technology, 25, 271-287.
[8] Jen, C. H. and Jiang, B. C. (2008). Combining on-line experiment and process control methods for changes in a dynamic model. International Journal of Production Research, 46, 3665-3682.
[9] Tseng, S. T., and Chen, P. Y. (2017). A Generalized Quasi-MMSE Controller for Run-to-Run Dynamic Models. Technometrics, 1-10.
[10] Tseng, S. T., and Lin, C. H. (2009). Stability analysis of single EWMA controller under dynamic models. IIE Transactions, 41(7), 654-663.
[11] Tseng, S. T., and Mi, H. C. (2014). Quasi-minimum mean square error run-to-run controller for dynamic models. IEE Transactions, 46(2), 185-196.
[12] Tseng, S. T., Mi, H. C., and Li, I. C. (2016). A Multivariate EWMA Controller for Linear Dynamic Processes, Technometrics, 58(1), 104-115.
[13] 李淑玲 (2010). “動態批次回饋系統的Quasi MMSE控制器之績效分析,” 國立清華大學統計學研究所碩士論文.
[14] 陳沛瑜 (2015). “批次動態回饋系統的一般化qMMSE控制器分析,” 國立清華大學統計學研究所碩士論文.