近年來，台灣地區空氣污染品質控制與監測問題，已引起大眾相當程度的重視，目前由環保署所制定評估空氣品質的空氣污染指標(Pollution Standards Index，簡稱PSI)皆為事後公佈，並未全然達到事先預警的效果。在多年的研究與推廣下，統計製程管制(Statistical Process Control，簡稱SPC)對於監測製程的穩定性與提昇產品品質均具良好的成效，但傳統的SPC係假設製程觀察值彼此獨立，然而隨著電腦科技的日新月異及資料收集方式不斷改進下，使得許多製程資料收集的間隔時間縮短，彼此間存在著自我相關性。因此，若將此類型資料以傳統的SPC方法進行監測極易產生誤判，而導致不必要的成本浪費。近年來已有學者提出配適原始資料的時間數列模型，若模式配適正確，並假設模式的殘差彼此獨立，即可利用傳統的SPC管制圖監測殘差值，達到監控品質之目的。除了指數平滑移動平均(EWMA)及累和(CUSUM)之殘差管制圖外，另有學者提出自我迴歸(Autoregressive) T2管制圖與自我迴歸移動平均(ARMA)等兩種管制圖。由於環保資料係經長時間不斷收集而得，本身具有自我相關性，故本研究擬針對上述四種管制圖在監測自我相關製程上的表現進行比較分析以期找出一最適合對空氣品質污染情況作有效監控的環保管制圖。若能在空氣品質出現異常現象的第一刻即預示警訊，將可降低其對人體健康可能造成之危害與損失，本研究之成果可作為對未來空氣品質預警與評估的重要參考。

Recently, the air pollution problems in Taiwan have aroused a great public concern. However, the PSI (Pollutant Standards Indices) of air quality stipulated by the EPA (Environmental Protection Administration) of Taiwan has always been announced and posted afterwards, thus it cannot give a timely precaution to the public. Traditionally, SPC (Statistical Process Control) charts are developed assuming that the process observations are independent and follow normal distribution. Unfortunately, this assumption is unrealistic in practice. The autocorrelated process, if mistreated as an independent process, will result in an improper decision and unnecessary cost. The most widely used SPC method for autocorrelated process is residual-based control chart, which involves fitting an appropriate ARMA model to the data and monitoring the residuals. If the model is correct, then the residuals are independent. Consequently, traditional SPC control charts can be used. So far, four different control charts, including EWMA and CUSUM residual control chart, Autoregressive T2 chart and ARMA (Autoregressive moving average) chart have been proposed by researches to monitor autocorrelated data. Due to the fact that environmental data possess the property of autocorrelation, this study compares the performance of these four control charts for monitoring autocorrelated air pollution data and select the most appropriate one for future use. Hopefully, giving a warning signal in advance, the result of this research could be a useful reference for evaluating environmental performance.

【1】 Adams, B. M. and Tseng, L. T. (1998). Robustness of Forecast-Based Monitoring Schemes. Journal of Quality Technology 30, 328~329.
【2】 Alwan, A. J. and Alwan, L.C. (1994). Monitoring Autocorrelated Processes Using Multivariate Quality Control Charts. Proceedings of the Decision Sciences Institute Annual Meeting 3, 2106~2108.
【3】 Alwan, L.C. and Roberts, H. V. (1988). Time-Series Modeling for Statistical Process Control. Journal of Business and Economic Statistics 6, 87~95.
【4】 Apley, D. W. and Tsung, F. (2002). The Autoregressive T2 Chart for Monitoring Univariate Autocorrelated Processes. Journal of Quality Technology 34, 80~89.
【5】 Atienza, O. O., Tang, L. C. and Ang. B. W. (2002). A CUSUM Scheme for Autocorrelated Observations. Journal of Quality Technology 34, 187~199.
【6】 Bagshaw, H. and Johnson, R. A. (1975). The Effect Serial Correlation on the Performance of CUSUM Test II. Technometrics 17, 73~80.
【7】 Box, G. E. P., Jenkins, G. M. and Reinsel, G. C. (1994). Time Series Analysis, Forecasting and Control, 3rd ED., Prentice-Hall, London.
【8】 Corbett, C. J. and Pan, J. N. (2002). Evaluating Environment Performance Using Statistical Process Control Techniques. European Journal of Operational Research 139, 68~83.
【9】 Ebizuka, K. and Akizuki, K. (1991). Prediction of Air Pollution Based Observed Data. Journal of Environmental Protection 24, 66~78.
【10】 Harris, T. J. and Ross, W. H. (1991). Statistical Process Control Procedure for Correlated Observations. Canadian Journal of Chemical Engineering 69, 48~57.
【11】 Jiang, W., Tsui, K. L. and Woodall, W. H. (2000). A New SPC Monitoring Method: The ARMA Chart. Technometrics 42, 399~410.
【12】 Johnson, R. A. and Bagshaw, M. (1974). The Effect of Serial Correlation on the Performance of CUSUM Tests. Technometrics 16, 103~112.
【13】 Krieger, C. A., Champ, C. W. and Alwan, L. C. (1992). Monitoring an Autoregressive Process. Presented at the Pittsburgh Conference on Modeling and Simulation, Pittsburgh, PA.
【14】 Lu, C. W. and Reynolds, M. R., JR. (1999). EWMA Control Charts for Monitoring the Mean of Autocorrelated Processes. Journal of Quality Technology 31, 166~188.
【15】 Lu, C. W. and Reynolds, M. R., JR. (2001). CUSUM Charts for Monitoring An Autocorrelated Processes. Journal of Quality Technology 33, 316~334.
【16】 Lucas, J. M. and Saccucci, M. S. (1990). Exponentially Weighted Moving Average Control Schemes: Properties and Enhancements. Technometrics 32, 1~12.
【17】 Montgomery, D. C. (2001). Introduction to Statistical Quality Control, 4th Ed., Wiley, New York.
【18】 Montgomery, D. C. and Mastrangelo, C. M. (1991). Some Statistical Process Control Methods for Autocorrelated Data. Journal of Quality Technology 23, 179~193.
【19】 Page, E. S. (1961). Cumulative Sum Charts. Technometrics 3, 1~9.
【20】 Paolo, Z. (1990). Time Series Analysis of Venice Air Quality Data. Journal of Environmental Protection 23, 125~134.
【21】 Reynolds, M. R., JR. Arnold, J. C. and Baik, J. W. (1996). Variable Sampling Internal Charts in the Presences of Correlation. Journal of Quality Technology 28, 12~30.
【22】 Roberts, S. W. (1959). Control Chart Tests Based on Geometric Moving Averages. Technometrics 1, 239~250.
【23】 Runger, G. C., Willemain, T. R. and Prabhu, S. (1995). Average Run Lengths for CUSUM Control Charts Applied to Residuals. Communication in Statistics-Theory and Methods 24, 273~282.
【24】 Superville, C. R. and Adams, B. M. (1994). An Evaluation of Forecast-Based Quality Control Schemes. Communications in Statistic -Simulation and computation 23, 645~661.
【25】 Timmer, D. H., Pigantiello, J., JR. and Longecker, M. (1998). The Development and evaluation of CUSUM-based Control Charts for An AR(1) Process. IIE Transactions 33, 525~534.
【26】 Tseng, S. and Adams, B. M. (1994). Monitoring autocorrelated Processes with an Exponentially Weighted Moving Average Forecast. Journal of Statistical Computation and Simulation 50, 187~195.
【27】 VanBrackle, L. N. and Reynolds, M. R., JR. (1997). EWMA and CUSUM Control Charts in the Presence of Correlation, Communications in Statistic -Simulation and computation 26, 979~1008.
【28】 Vander Weil, S. A. (1996). Modeling Process that Wander Using Moving Average Models. Technometrics 38, 139~151.
【29】 Vander Weil, S. A., Tucker, W. T., Faltin, F. W. and Doganaksoy, N. (1992). Algorithmic Statistical Process Control: Concepts and an Application. Technometrics 34, 286~297
【30】 Wardell, D. G., Moskowitz, H. and Palnte, R. D. (1994). Run-Length Distributions of Special-Cause Control Charts for Correlated Processes. Technometrics 36, 3~17.
【31】 Zhang, N. F. (1998). A Statistical Control Chart for Stationary Process Data. Technometrics 40, 24~38.
【32】 Yashchin, E. (1993). Performance of CUSUM Control Schemes for Serially Correlated Observations. Technometrics 35, 37~52.
【33】 呂世宗(1988)。大都會區空氣品質污染潛勢預測之研究，第一階段：台北地區，國立中央大學大氣物理系研究報告。
【34】 林茂文(1993)。時間數列分析與預測─增訂版。華泰書局。
【35】 吳柏林、廖敏治(1993)。大台北都會區空氣污染指標之時空數列分析，
中國統計學報 31，139－167。
【36】 鄭春生(1996)。品質管理─增訂版, 三民書局。
【37】 藺超華(1993)。板橋地區空氣污染預測模式，國立政治大學統計學研究所碩士論文。
【38】 台北市政府環境保護局(2002)。空氣品質改善計劃規劃整合暨成效評核計劃。
【39】 行政院環境保護署(2000)。台灣地區空氣污染防制總檢討。
【40】 行政院環境保護署(2002)。台灣地區空氣品質長期趨勢分析(III)。
【41】 行政院環境保護署(2000)。台灣大都會地區改善空氣品質之經濟效益評估與酸雨風險認知調查。