Computation of average run length for residual-based T ² control chart for multivariate autocorrelated processes

Abstract

The expression of residual is obtained according to its dynamic response to mean shift, then the distribution of T ² statistic applied to the residual is derived, thus the probability of the T ² statistic lying outside the control limit is calculated. The above-mentioned results are substituted into the infinite definition expression of the average run length (ARL), and then the final finite ARL expression is obtained. An example is used to demonstrate the procedures of the proposed method. In the comparative study, eight autocorrelated processes and four different mean shifts are performed, and the ARL values of the proposed method are compared with those obtained by simulation method with 50,000 replications. The accuracy of the proposed method can be illustrated through the comparative results.

Keywords

autocorrelated process / average run length (ARL) / residual-based T ² control chart

References

[1] Montgomery D. C. Statistical Quality Control: A Modern Introduction [M]. 2009, New York: Wiley.

[2] Alwan L. C., Radson D. Time-series investigation of subsample mean charts[J]. IIE Transactions, 1992, 24(5): 66-80.

[3] Harris T. J., Ross W. H. Statistical process control procedures for correlated observations[J]. Canadian Journal of Chemical Engineering, 1991, 69(1): 48-57.

[4] Noorossana R., Vaghefi S. J. M. Effect of autocorrelation on performance of the MCUSUM control chart[J]. Quality and Reliability Engineering International, 2006, 22(2): 191-197.

[5] Loredo E. N., Jearkpaporn D., Borror C. M. Model-based control chart for autoregressive and correlated data[J]. Quality and Reliability Engineering International, 2002, 18(6): 489-496.

[6] Montgomery D. C., Mastrangelo C. M., Faltin F. W., et al. Some statistical process control methods for autocorrelated data [J]. Journal of Quality Technology, 1991, 23(3): 179-204.

[7] Yang M., Sun Jing. Residual-based T ² control chart for bivariate autocorrelated processes [J]. Journal of Tsinghua University: Science and Technology, 2006, 46(3): 403-406.

[8] Sun J., Yang Mu’er. Residual-based T ² control chart for monitoring multivariate auto-correlated processes[J]. Journal of Tsinghua University: Science and Technology, 2007, 47(12): 2184-2187.

[9] Pan X., Jarrett Jeffrey. Using vector autoregressive residuals to monitor multivariate processes in the presence of serial correlation[J]. International Journal of Production Economics, 2007, 106(1): 204-216.

[10] Jarrett J., Pan Xia. The quality control chart for monitoring multivariate autocorrelated processes[J]. Computational Statistics and Data Analysis, 2007, 51(8): 3862 3870

[11] Ryan T. P. Discussion of some statistical process control methods for autocorrelated data by D. C. Montgomery and C. M. Mastrangelo[J]. Journal of Quality Technology, 1991, 23(3): 200-202.

[12] Wardell D. G., Moskowitz H., Plante R. D. Run-length distributions of special-cause control charts for correlated processes [J]. Technometrics, 1994, 36(1): 3 17

[13] Lu C. W., Reynolds M. R. Jr. EWMA control charts for monitoring the mean of autocorrelated processes[J]. Journal of Quality Technology, 1999, 31(2): 166-188.

[14] Jamal A., Seyed T. A. N., Babak Abbasi. Artificial neural networks in applying MCUSUM residuals charts for AR(1) processes[J]. Applied Mathematics and Computation, 2007, 189(2): 1889 1901

[15] Runger G. C., Willemain T. R., Prabhu S. Average run lengths for CUSUM control charts applied to residuals[J]. Communications in Statistics-Theory and Methods, 1995, 24(1): 273-282.