在本篇論文中我們針對結合統計製程管制與批次回饋控制的方法進行研究。我們 以現有的方法為基礎，提出一個製程品質控制績效更為提昇之架構。此架構包含兩個主要模 組，在統計製程管制模組中，蕭華特及指數加權移動平均管制圖分別被使用來偵測製程之大 偏移及小偏移。在批次回饋控制模組中，我們使用移動平均的理論來估計製程偏移量的大小 ，並藉此調整製程控制模型，產生出下一批次的製程輸入值。利用蒙地卡羅的模擬結果，我 們提出了較為強健且經濟的控制參數選擇法則，我們也由模擬中驗證對於被速移與漂移所影 響之製程，我們的方法較現有的方法，特別是結合指數加權移動平均管制圖與指數加權移動 平均控制法要來的有效。

　　　　　In this paper we present an enhanced algorithm to integrate the functions of statistical process control and run-by-run feedback control. In the statistical process control function, the Shewhart control chart is used to detect a large process deviation while the EWMA control chart is used to detect a small process deviation. In the run-by-run feedback control function, a moving average formulation is used to update the control model and generate the recipe on a run by run basis. To validate the effectiveness of the proposed algorithm, Monte Carlo simulations are performed to simulate processes that are subject to shifts and drift. Results of the simulations show that the enhanced algorithm is bettr than the current schemes such as a combined EWMA control chart and EWMA feedback control scheme.