Advertisement

Comparison of joint control schemes for multivariate normal i.i.d. output

  • Manuel Cabral MoraisEmail author
  • Wolfgang Schmid
  • Patrícia Ferreira Ramos
  • Taras Lazariv
  • António Pacheco
Original Paper
  • 64 Downloads

Abstract

The performance of a product frequently relies on more than one quality characteristic. In such a setting, joint control schemes are used to determine whether or not we are in the presence of unfavorable disruptions in the location (\({\varvec{\mu }}\)) and spread (\({\varvec{\varSigma }}\)) of a vector of quality characteristics. A common joint scheme for multivariate output comprises two charts: one for \({\varvec{\mu }}\) based on a weighted Mahalanobis distance between the vector of sample means and the target mean vector; another one for \({\varvec{\varSigma }}\) depending on the ratio between the determinants of the sample covariance matrix and the target covariance matrix. Since we are well aware that there are plenty of quality control practitioners who are still reluctant to use sophisticated control statistics, this paper tackles Shewhart-type charts for \({\varvec{\mu }}\) and \({\varvec{\varSigma }}\) based on three pairs of control statistics depending on the nominal mean vector and covariance matrix, \({\varvec{\mu }}_0\) and \({\varvec{\varSigma }}_0\). We either capitalize on existing results or derive the joint probability density functions of these pairs of control statistics in order to assess the ability of the associated joint schemes to detect shifts in \({\varvec{\mu }}\) or \({\varvec{\varSigma }}\) for various out-of-control scenarios. A comparison study relying on extensive numerical and simulation results leads to the conclusion that none of the three joints schemes for \({\varvec{\mu }}\) and \({\varvec{\varSigma }}\) is uniformly better than the others. However, those results also suggest that the joint scheme with the control statistics \(n \, ( \bar{\mathbf {X}}-{\varvec{\mu }}_0 )^\top \, {\varvec{\varSigma }}_0^{-1} \, ( \bar{\mathbf {X}}-{\varvec{\mu }}_0 )\) and \(\hbox {det} \left( (n-1) \mathbf{S} \right) / \hbox {det} \left( {\varvec{\varSigma }}_0 \right) \) has the best overall average run length performance.

Keywords

Average run length Joint schemes Multivariate normal Statistical process control 

Notes

Acknowledgements

This work was partially supported by FCT (Fundação para a Ciência e a Tecnologia) through projects UID/Multi/04621/2013 and PEst-OE/MAT/UI0822/ 2014. We are most grateful to the two reviewers who selflessly devoted their time to scrutinize this work and offered very pertinent comments that led to a shorter and improved version of the original manuscript.

References

  1. Akritas, A.G., Akritas, E.K., Malaschonok, G.I.: Various proofs of Sylvester’s (determinant) identity. Math. Comput. Simul. 42, 585–593 (1996)MathSciNetCrossRefzbMATHGoogle Scholar
  2. Alt, F.B.: Multivariate quality control. In: Kotz, S., Johnson, N.L., Read, C.R. (eds.) Encyclopedia of Statistical Sciences, vol. 6, pp. 5312–5323. Wiley, New York (1984)Google Scholar
  3. Anderson, T.W.: An Introduction to Multivariate Statistical Analysis. Wiley, New York (1958)zbMATHGoogle Scholar
  4. Arnold, S.F.: Wishart distribution. In: Kotz, S., Balakrishnan, N., Read, C.R., Vidakovic, B., Johnson, N.L. (eds.) Encyclopedia of Statistical Sciences (Second Edition), vol. 15, pp. 9184–9188. Wiley, New York (2006)Google Scholar
  5. Fujikoshi, Y., Ulyanov, V.V., Shimizu, R.: Multivariate Statistics: High-Dimensional and Large-Sample Approximations. Wiley, New York (2010)CrossRefzbMATHGoogle Scholar
  6. Gupta, A.K., Nagar, D.K.: Matrix Variate Distributions. Chapman & Hall/CRC, Boca Raton (2000)zbMATHGoogle Scholar
  7. Hotelling, H.: Multivariate quality control illustrated by the air testing of sample bombsights. In: Eisenhart, C., Hastay, M.W., Wallis, W.A. (eds.) Techniques of Statistical Analysis, pp. 111–184. McGraw Hill, New York (1947)Google Scholar
  8. Karr, A.F.: Probability. Springer, New York (1993)CrossRefzbMATHGoogle Scholar
  9. Li, K., Geng, Z.: The noncentral Wishart distribution and related distributions. Commun. Stat. Theory Methods 32, 33–45 (2003)MathSciNetCrossRefGoogle Scholar
  10. Montgomery, D.C., Wadsworth, H.M.: Some techniques for multivariate quality control applications. In: Transactions of the ASQC. American Society for Quality Control, Washington DC (1972)Google Scholar
  11. Morais, M.C., Pacheco, A.: Some stochastic properties of upper one-sided x and EWMA charts for \(\mu \) in the presence of shifts in \(\sigma \). Seq. Anal. 20, 1–12 (2001)MathSciNetCrossRefzbMATHGoogle Scholar
  12. Morais, M.C., Okhrin, Y., Pacheco, A., Schmid, W.: On the stochastic behaviour of the run length of EWMA control schemes for the mean of correlated output in the presence of shifts in \(\sigma \). Stat. Decis. 24, 397–413 (2006)MathSciNetzbMATHGoogle Scholar
  13. Morais, M.C., Okhrin, Y., Pacheco, A., Schmid, W.: EWMA charts for multivariate output: some stochastic ordering results. Commun. Stat. Theory Methods 37, 2653–2663 (2008)MathSciNetCrossRefzbMATHGoogle Scholar
  14. Muirhead, R.J.: Aspects of Multivariate Statistical Theory. Wiley, Hoboken (1982)CrossRefzbMATHGoogle Scholar
  15. R Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna (2013). http://www.R-project.org/. Accessed 28 Aug 2017
  16. Ramos, P.A.A.C.F.P.: Performance analysis of simultaneous control schemes for the process mean (vector) and (co)variance (matrix). Ph.D. thesis, Instituto Superior Técnico, Universidade Técnica de Lisboa (2013)Google Scholar
  17. Ramos, P.F., Morais, M.C., Pacheco, A., Schmid, W.: On the misleading signals in simultaneous schemes for the mean vector and covariance matrix of multivariate i.i.d. output. Stat. Pap. 57, 471–498 (2016)MathSciNetCrossRefzbMATHGoogle Scholar
  18. Reynolds Jr., M.R., Cho, G.Y.: Multivariate control charts for monitoring the mean vector and covariance matrix. J. Qual. Technol. 38, 230–253 (2006)CrossRefGoogle Scholar
  19. Reynolds Jr., M.R., Stoumbos, Z.G.: Combinations of multivariate Shewhart and MEMWA control charts for monitoring the mean vector and covariance matrix. J. Qual. Technol. 40, 381–393 (2008)CrossRefGoogle Scholar
  20. Rohatgi, V.K.: An Introduction to Probability Theory and Mathematical Statistics. Wiley, New York (1976)zbMATHGoogle Scholar
  21. Seber, G.A.F.: Multivariate Observations. Wiley, New York (1984)CrossRefzbMATHGoogle Scholar
  22. Sylvester, J.J.: On the relation between the minor determinants of linearly equivalent quadratic functions. Philos. Mag. 1, 295–305 (1851)CrossRefGoogle Scholar
  23. Tiku, M.: Noncentral beta distribution. In: Kotz, S., Balakrishnan, N., Read, C.R., Vidakovic, B., Johnson, N.L. (eds.) Encyclopedia of Statistical Sciences (Second Edition), vol. 7, pp. 5541–5542. Wiley, New York (2006a)Google Scholar
  24. Tiku, M.: Noncentral F-distribution. In: Kotz, S., Balakrishnan, N., Read, C.R., Vidakovic, B., Johnson, N.L. (eds.) Encyclopedia of Statistical Sciences (Second Edition), vol. 7, pp. 5546–5550. Wiley, New York (2006b)Google Scholar
  25. Tong, Y.L.: The Multivariate Normal Distribution. Springer, New York (1990)CrossRefzbMATHGoogle Scholar
  26. Walsh, J.E.: Operating characteristics for tests of the stability of a normal population. J. Am. Stat. Assoc. 47, 191–202 (1952)CrossRefzbMATHGoogle Scholar
  27. Wang, K., Yeh, A., Li, B.: Simultaneous monitoring of process mean vector and covariance matrix via penalized likelihood estimation. Comput. Stat. Data Anal. 78, 206–217 (2014)MathSciNetCrossRefzbMATHGoogle Scholar
  28. Wells, W.T., Anderson, R.L., Cell, J.W.: The distribution of the product of two central or non-central chi-square variates. Ann. Math. Stat. 33, 1016–1020 (1962)CrossRefzbMATHGoogle Scholar
  29. Wetherill, G.B., Brown, D.W.: Statistical Process Control: Theory and Practice. Chapman and Hall, London (1991)CrossRefzbMATHGoogle Scholar
  30. Wolfram Research, Inc.: Mathematica, Version 10.3. Champaign (2015). http://reference.wolfram.com/language/. Accessed 28 Aug 2017

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of Mathematics and CEMAT (Center for Computational and Stochastic Mathematics), Instituto Superior TécnicoULisboaLisbonPortugal
  2. 2.Department of StatisticsEuropean University ViadrinaFrankfurt (Oder)Germany
  3. 3.CEMATUniversidade Autónoma de LisboaLisbonPortugal
  4. 4.Center of Information Services and High Performance ComputingTU DresdenDresdenGermany

Personalised recommendations