在半導體生產製程中，量測延遲 (metrology delay) 是相當常見的現象。Wu et al. (2008) 針對線性之SISO生產製程，沿用single EWMA控制器的方法對下一期的投入變數作調整，但由於製程存在量測延遲的現象，故僅能採用的資訊為量測延遲發生前之製程產出值。當製程干擾項為非平穩型時間序列模型時，所採用的資訊通常不夠準確，故而導致製程的總均方誤差 (Total Mean Squared Error, TMSE) 變大。為克服此困難，本文考慮引進虛擬量測 (Virtual Metrology, VM) 的方法，這種方法不會受量測延遲現象的影響，因此對製程趨勢的預測較為準確，唯其精確度通常較低。本篇論文的研究重點，主要是希望結合量測儀器延遲的觀測值，以及VM的預測值，對下一期投入變數作調整，藉此能降低製程的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 Cliffs, NJ.
[2] Del Castillo, E. (2002), Statistical Process Adjustment for Quality Control, John-Wiley & Sons Inc., New York.
[3] Good, R. and Qin, S. J. (2002), “Stability Analysis of Double EWMA Run-to-Run Control with Metrology Delay,” Proc. Amer. Contr. Conf., Anchorage, AK, pp. 2156-2161.
[4] Good, R. and Qin, S. J. (2006), “On the Stability of MIMO EWMA Run-to-Run Controllers with Metrology Delay,” IEEE Transactions on Semiconductor Manufacturing, 19, pp. 78-86.
[5] Ingolfsson, A. and Sachs, E. (1993), “Stability and Sensitivity of an EWMA Controller,” Journal of Quality Technology, 25, pp. 271-287.
[6] Sachs, E., Guo, R. S., Ha, S. and Hu, A. (1991), “Process Control System for VLSI Fabrication,” IEEE Transactions on Semiconductor Manufacturing, 4, pp. 134-144.
[7] Tseng, S. T., Yeh, A. B., Tsung, F. and Chan, Y. Y. (2003), “A Study of Variable EWMA Controller,” IEEE Transactions on Semiconductor Manufacturing, 16, pp. 633-643.
[8] Wu, M. F., Lin, C. H., Wong, D. S. H., Jang, S. S. and Tseng, S. T. (2008), “Performance Analysis of EWMA Controllers Subject to Metrology Delay,” IEEE Transactions on Semiconductor Manufacturing, 21, pp. 413-425.
[9] 何人杰 (2000), “EWMA控制器非固定折扣因子之研究”, 國立清華大學統計學研究所碩士論文.
[10] 林資程 (2004), “Single EWMA控制器在動態模型下之績效分析”, 國立清華大學統計學研究所碩士論文.
[11] 詹昱宏 (2008), “EI控制器在動態模型下之績效及穩定性分析”, 國立清華大學統計學研究所碩士論文.