在監控多變量資料時，除了對品質特徵的平均向量做監控外，還盼望能對品質特徵的變異數和變數間的相關性大小進行監控。在現行的文獻中，對於共變異數或形狀矩陣的監控大多數背後都有要求資料服從多變量常態分配的假設，本文主要探討的議題是當資料的分配不為多變量常態分配時，如何利用無母數方法對資料的形狀矩陣來進行監控。我們利用無母數統計中的空間秩概念，建構一個無母數的EWMA管制圖，並用統計模擬的方式來討論我們所提出的管制圖在監控效率上的表現。最後我們舉一個白葡萄酒資料的實例來說明在實務上如何運用所提出的管制圖。

When monitoring the multivariate production process, we intend to monitor not only the average of quality characteristics but also its covariance or shape matrix. Nowadays, there are many multivariate control schemes for monitoring the covariance or shape matrix, and most of them are based on the normal assumption. However, in reality, the quality characteristics are usually not normally distributed. Therefore, in this article we want to construct a control chart to monitor the shape matrix based on a nonparametric approach. We use the concept of spatial-rank and exponentially weighted moving average control scheme to develop the on-line monitoring control chart. Simulation performance of this chart is given in the article. Finally, we use a real-data example from a white-wine production process to demonstrate the applicability of the proposed control scheme.

[1] Chakraborti, S., Van der Laan, P. and Bakir, S. T. (2001).“Nonparametric control charts: an overview and some results.”Journal of Quality Technology, 33, pp. 304-315.
[2] Cortez, P., Cerdeira, A, Almeida, F., Matos, T. and Reis, J. (2009).“Modeling wine preferences by data mining from physicochemical properties.”Decision Support Systems, 47, pp. 547-553.
[3] Gandy, A. and Kvaloy, J. T. (2013).“Guaranteed conditional performance of control charts via bootstrap methods.”Scandinavian Journal of Statistics, 40, pp. 647-668.
[4] Hawkins, D. M., and Masoudou-Tchao, E. M. (2008).“Multivariate exponentially weighted moving covariance matrix.”Technometrics, 50, pp. 155-166.
[5] Hotelling, H. (1947).“Multivariate quality control.”Techniques of Statistical Analysis, Eisenhart, Hastay, and Wallis (Ed.), McGraw-Hill, New York.
[6] John, S. (1971)“. Some optimal multivariate tests.”Biometrika, 59, pp. 123-127.
[7] Li, Z., Zou, C., Wang, Z. and Huang, L. (2013).“A multivariate sign chart for monitoring process shape parameters.”Journal of Quality Technology, 45, pp. 149-165.
[8] Lowry, C. A., Woodall, W. H., Champ, C. W. and Rigdon, S. E. (1992).“A multivariate exponentially weighted moving average control chart.”Technometrics, 34, pp. 46-53.
[9] Oja, H. (2010).“Multivariate Nonparametric Methods with R.”Springer, New York.
[10] Page, E. S. (1954).“Continuous inspection schemes.”Biometric, 41, pp. 100-115.
[11] Roberts, S.W. (1959)“. Control chart test based on geometric moving averages.” Technometrics, 1, pp. 234-250.
[12] Shewart, W. A. (1924)“. Some application of statistical methods to the analysis of physical and engineering data.”Bell System Technical Journal, 3, pp. 43-87
[13] Sirkia, S., Taskinen, S., Oja, H. and Tyler, D. E. (2009).“Tests and estimates of shape based on spatial signs and ranks.”Journal of Nonparametric Statistics, 21, pp. 155-176.
[14] Stoumbos, Z. G. and Sullivan, J. H. (2002).“Robustness to non-normality of the multivariate EWMA control chart.”Journal of Quality Technology, 34, pp.260-276.
[15] Woodall, W. H. and Ncube, M. M. (1985).“Multivariate CUSUM qualitycontrol procedures.”Technometrics, 27, pp. 285-292.
[16] Yeh, A. B., Huwang, L. and Wu, C.-W. (2005).“A multivariate EWMA control chart for monitoring process variability with individual observations.”IIE Transactions, 37, pp. 1023-1035.
[17] Yeh, A. B., Huwang, L. andWu, Y.-F. (2004)“. A likelihood-ratio-based EWMA control chart for monitoring variability of multivariate normal processes.”IIE Transactions, 36, pp. 865-879.
[18] Yeh, A. B., Lin, D., Zhou, H. and Venkataramani, C. (2003).“A multivariate exponentially weighted moving average control chart for monitoring process variability.”Journal of Applied Statistics, 30, pp. 507-536.
[19] Zou, C. and Tsung, F. (2011).“A multivariate sign EWMA control chart.” Technometrics, 53, pp. 84-97.
[20] Zou, C., Wang, Z. and Tsung, F. (2012).“A spatial rank-based multivariate EWMA control chart.” Naval Research Logistics, 59, pp. 91-110.