Semiparametric MEWMA for Phase II profile monitoring

A control chart is one of the statistical process techniques that used to monitor different processes. Some processes are characterized by functions or profiles, and a profile functional relationship between dependent independent variable(s) quality process. Several research studies were conducted on linear profiling where only fixed effects considered. However, in this research, we focus random as they represent differences profiles thus more proper for interpretation. Two approaches proposed study Phase II monitoring; first approach nonparametric via residuals second semiparametric approach, technique combines parametric estimates with portion residuals. Usually, estimations lead biased when model misspecified, whereas may give high variances, preferred. New multivariate exponential weighted moving average (MEWMA) charts introduced their performances compared samples shift sizes, correlation within was The run length (ARL) time signal (ATS) criteria choosing best approach. Simulation real datasets utilized comparing performance MEWMA charts.