A generalized multiple dependent state sampling chart based on ridge penalized likelihood ratio for high-dimensional covariance matrix monitoring

Online monitoring of high-dimensional processes variability in which the number variables is larger than sample size a challenging issue for quality practitioners because covariance matrix not invariable. To deal with this challenge, generalized multiple dependent state sampling (GMDS) chart based on ridge penalized likelihood ratio (RPLR) statistic developed Phase II multivariate process under setting. The control benefits from three advantages: (1) departing conventional charts, it can be efficiently employed both spars and non-spars matrices; (2) able to detect shift patterns only few elements are deviated their nominal values; (3) outperforms detectability RPLR terms average run length (ARL) standard deviation (SDRL). performance RPLR, MDS-RPLR, GMDS-RPLR charts compared using extensive simulation studies by considering different diagonal and/or off-diagonal disturbance. Moreover, sensitivity analysis provided analyze how GMDS parameter affect properties chart.