Multivariate Cuscore control charts for monitoring the mean vector in autocorrelated processes

摘要: In many systems, quantitative observations of process variables can be used to characterize a for quality control purposes. As the intervals between become shorter, autocorrelation may occur and lead high false alarm rate in traditional Statistical Process Control (SPC) charts. this article, Multivariate Cuscore (MCuscore) SPC procedure based on sequential likelihood ratio test fault signature analysis is developed monitoring mean vector an autocorrelated multivariate process. The MCuscore charts transient, steady ramp shift signal are designed; they do not rely assumption known starting time. An example presented demonstrate application chart three online search engine marketing tracking Furthermore, simulation shows that outperforms cumulative sum detecting ...