{"references": ["", "Montgomery D.C. (2005) Introduction to statistical quality control.\nHoboken, N.J. John Weily & Sons, Inc..", "Shewhart, W.A. (1931) Economic Control of Quality of Manufactured\nProduct. Milwaukee, WI: ASQ Quality Press. 1980.", "Hotelling H. (1947) Multivariate quality control-Illustrated by the air\ntesting of sample bombsights. Techniques of Statistical Analysis,\nEisenhart, C., Hastay, M.W., Wallis, W.A. (Eds). McGraw-Hill: New\nYork.", "Woodall WH., Ncube MM. (1985) Multivariate CUSUM qualitycontrol\nprocedures. Technometrics. 27(3): 285-292.", "Healy JD. (1987) A note on multivariate CUSUM procedures.\nTechnometrics. 29(4): 409-412.", "Crosier R.B. (1988) Multivariate Generalization of Cumulative Sum\nQuality Control Schemes. Technometrics. 30(3): 291-302.", "Pignatiello JJ., Runger GC. (1990) Comparisons of multivariate\nCUSUM charts. Journal of Quality Technology. 22(3): 173-186.", "Ngai HM., Zhang J. (2001) Multivariate cumulative sum control charts\nbased on projection pursuit. Statistica Sinica. 11(3): 747-766.", "Chan LK., Zhang J. (2001) Cumulative sum control charts for the\ncovariance matrix. Statistica Sinica. 11(3): 767-790.\n[10] Qiu PH., Hawkins DM. (2001) A rank-based multivariate CUSUM\nprocedure. Technometrics. 43(2): 120-132\n[11] Qiu, PH., Hawkins, DM. (2003) A nonparametric multivariate\ncumulative sum procedure for detecting shifts in all directions. Journal\nof the Royal Statistical Society Series D-The Statistician. 52(2): 151-\n164.\n[12] Runger GC., Testik MC. (2004) Multivariate extensions to cumulative\nsum control charts. Quality and Reliability Engineering International.\n20(6): 587-606.\n[13] Lowry CA., Woodall WH., Champ CW., Rigdon SE. (1992) A\nmultivariate Exponential Weighted Moving Average control chart.\nTechnometrics. 34(1): 46-53.\n[14] Rigdon SE. (1995) An integral equation for the in-control average run\nlength of a multivariate exponentially weighted moving average control\nchart. Journal of Statistical Computations and Simulation. 52(4): 351-\n365.\n[15] Yumin, L. (1996) An improvement for MEWMA in multivariate process\ncontrol Computers and Industrial Engineering. 31(3-4): 779-781.\n[16] Runger GC., Prabhu SS. (1996) A markov chain model for the\nmultivariate exponentially weighted moving average control chart.\nJournal of the American Statistical Association. 91(436): 1701-1706.\n[17] Kramer HG., Schmid W. (1997) EWMA charts for multivariate time\nseries. Sequential Analysis. 16(2): 131-154.\n[18] Prabhu SS., Runger GC. (1997)Designing a multivariate EWMA control\nchart. Journal of Quality Technology. 29(1): 8-15.\n[19] Fasso A. (1999) One-sided MEWMA control charts. Communications in\nStatistics -Theory and Methods. 28(2): 381-401\n[20] Borror CM., Montgomery DC., Runger GC. (1990) Robustness of the\nEWMA control chart to non-normality. Journal of Quality Technology.\n31(3): 309-316.\n[21] Runger GC., Keats JB., Montgomery DC., Scranton RD., (1999)\nImproving the performance of a multivariate exponentially weighted\nmoving average control chart. Quality and Reliability Engineering\nInternational. 15(3): 161-166.\n[22] Tseng S., Chou R., Lee S. (2002) A study on a multivariate EWMA\ncontroller. IIE Transactions. 34(6): 541-549.\n[23] Yeh AB., Lin DKJ., Zhou H., Venkataramani C. (2003) A multivariate\nexponentially weighted moving average control chart for monitoring\nprocess variability. Journal of Applied Statistics. 30(5): 507-536.\n[24] Testik MC., Runger GC., Borror CM. (2003) Robustness properties of\nmultivariate EWMA control charts. Quality and Reliability Engineering\nInternational. 19(1): 31-38.\n[25] Testik MC., Borror CM. (2004) Design strategies for the multivariate\nexponentially weighted moving average control chart. Quality and\nReliability Engineering International. 20(6): 571- 577.\n[26] Chen GM., Cheng SW., Xie HS. (2005) A new multivariate control\nchart for monitoring both location and dispersion. Communications in\nStatistics - Simulation and Computation. 34(1): 203-217.\n[27] Atashgar, K. (2012). Identification of the change point: An overview.\nInternational Journal of Advanced Manufacturing Technology, DOI:\n10.1007/s00170-012-4131-2.\n[28] Mason Robert L., Tracy Nola D., Young John C. (1997) A Practical\nApproach for Interpreting Multivariate T2 control Chart Signals. Journal\nof Quality Technology. 29(4): 396:406.\n[29] Aparisi F., Avendano G., Sanz J. (2006) Techniques to interpret T2\ncontrol chart signals. IIE Transaction. 38(8): 647-657.\n[30] Niaki STA., Abbasi B. (2005) Fault diagnosis in multivariate control\ncharts using artificial neural network. Quality and reliability Engineering\nInternational. 21(8): 825-840.\n[31] Nedumaran G., Pignatiello JJ.Jr., Calvin J.A. (1998) Estimation of the\ntime of a step-change with \u03c7 2 control chart. Quality Engineering.\n13(2):765-778.\n[32] Noorossana R., Arbabzadeh N., Saghaei A., (2008) Painabar K.\nDevelopment of procedure of detection change point in multi\nenvironment. 6th International Conference on Industrial Engineering,\nIran-Tehran. (Written in Persian language)\n[33] Noorossana R., Atashgar K., Saghaee, A. (2011) An integrated solution\nfor monitoring process mean vector. International Journal of Advanced\nManufacturing Technology. 56(5): 755-765.\n[34] Atashgar K. and Noorossana R. (2010) An integrating approach to rootcause\nanalysis of a bivariate mean vector with a linear trend disturbance.\nInternational Journal of Advance Manufacturing Technology. 52(1):\n407-420.\n[35] Guh, RS. (2007). On-line Identification and Quantification of Mean\nShifts in Bivariate Processes using a Neural Network-based Approach.\nQuality and Reliability Engineering International, 23(3), 367-385\n[36] Hwarng, HB. (2008). Toward identifying the source of mean shifts in\nmultivariate SPC: a neural network approach. International Journal of\nProduction Research, 46(20),5531\u20135559."]}

When an assignable cause(s) manifests itself to a multivariate process and the process shifts to an out-of-control condition, a root-cause analysis should be initiated by quality engineers to identify and eliminate the assignable cause(s) affected the process. A root-cause analysis in a multivariate process is more complex compared to a univariate process. In the case of a process involved several correlated variables an effective root-cause analysis can be only experienced when it is possible to identify the required knowledge including the out-of-control condition, the change point, and the variable(s) responsible to the out-of-control condition, all simultaneously. Although literature addresses different schemes to monitor multivariate processes, one can find few scientific reports focused on all the required knowledge. To the best of the author’s knowledge this is the first time that a multi task model based on artificial neural network (ANN) is reported to monitor all the required knowledge at the same time for a multivariate process with more than two correlated quality characteristics. The performance of the proposed scheme is evaluated numerically when different step shifts affect the mean vector. Average run length is used to investigate the performance of the proposed multi task model. The simulated results indicate the multi task scheme performs all the required knowledge effectively.