【正文】
37]Apley, D. W. and Shi, J. (1994). The GLRT for statistical process control of autocorrelated processed, IIE Transactions, 31, 11231134. [38]Shu, L. J., Apley, D. W., and Tsung, F. (2002). Autocorrelated process monitoring using triggered Cuscore charts. Quality and Reliability Engineering International, 18, 411421. [39]Lorden, G. (1971). Procedures for reacting to a change in distribution. Annals of Mathematics and Statistics, 42, 18971908.[40]Zhao Yi, Tsung Fugee, Wang Zhaojun (2005). Dual CUSUM control scheme for detecting a range of mean shift. IIE Transactions, 37, 10471058.[41]Siegmund, D. , and Venkatraman, E. S. (1995). Using the generalized likelihood ratio statistic for sequential detection of a Changepoint, Annals of Statistics, 23, 255271.The Summary and Review of Statistical Process Control Chartsfor Monitoring the Mean of Variable CharacteristicZhaojun WangDepartment of Statistics, School of Mathematical SciencesNankai University, Tianjin 300071.Email: zjwangAbstractSome resent reviews about the static and adaptive statistical process control (SPC) charts for monitoring the mean of continuous variable characteristic and some related reference papers are given in this paper. The author of this paper wish to enhance the research and application of SPC in China.Keywords: Shewhart control chart, CUSUM control chart, EWMA control chart.作者簡介:王兆軍,男,教授(博導(dǎo)),1965年8月生于河北,南開大學(xué)數(shù)學(xué)科學(xué)學(xué)院統(tǒng)計系主任。對其中有興趣的讀者可參見相應(yīng)的文獻。關(guān)于動態(tài)控制圖,有Soumbos和Reynolds提出的SPRT控制圖[35, 36]?,F(xiàn)在討論最多的模型是ARMA,AR(1)及IMA(1,1)模型。二、關(guān)于獨立數(shù)據(jù)的動態(tài)控制圖的研究動態(tài)控制圖的特點是其樣本容量或抽樣區(qū)間的大小與前一組樣本的位置有關(guān),這方面的成果可分為五類:VSI(Variable Sampling Interval)、VSS(Variable Sample Size)、VSSI(Variable Sample Size and Sampling Interval)、VP(Variable Parameter)和VSSIFT (Variable Sample size and Sampling Interval at Fixed Time)控制圖。統(tǒng)計過程控制圖的研究起源于1931年由Shewhart博士提出的Shewhart Xbar控制圖,在這七十年間,它得到了廣泛的研究與發(fā)展,取得了相當豐富的成果。四、針對相關(guān)數(shù)據(jù)控制圖的研究 關(guān)于相關(guān)數(shù)據(jù)的第二階段控制圖,基本有兩大類方法[2229],一類是基于觀測數(shù)據(jù)本身的,另一類是基于一步預(yù)測殘差的。其中僅LRT圖可同時檢測均值與方差的飄移。雖然Han和Tsung僅比較了ARL的大小,并沒有從運行長度的分布和標準差角度進行比較,但這是利用大樣本性質(zhì)研究控制圖的第一篇文章(不包括變點問題。1995年12月晉升副教授,2000年12月晉升教授,2002年4月被選為博士生指導(dǎo)教師。六、檢測區(qū)間飄移的控制圖的研究現(xiàn)有絕大多數(shù)控制圖的最優(yōu)參數(shù)都是針對固定飄移設(shè)計得到的,但在許多實際問題中,飄移的大小是很難事先確定的,于是就有了區(qū)間飄移的檢測問題。理論與模擬結(jié)果均顯示這種控制圖