Research On MEWMA Control Chart With Dimensional Analysis

Statistical Process Control(SPC)is a collection of statistical methods aimed at monitoring and controlling a process.In many industrial applications,the quality of a product typically depends on several correlated quality characteristics and there may be a physical relationship.It's a challenge to control the quality characteristics of these processes for enterprises and researchers.Monitoring these two quality characteristics independently by two univariate control charts can be misleading.The process could be out-of-control when both variables are monitored simultaneously using a multivariate control chart even though neither univariate control chart shows any out-of-control signal.O ne of the major drawbacks of using many univariate SPC charts together is that the overall type I error(or false alarm)becomes large as the number of the quality characteristics increases.In addition,this.practice ignores the relationship(or correlation)among the quality characteristics,which may prove to be vital for diagnostic purposes.Similar to the development of univariate control charts,many researchers turn their attention to detecting small shifts in the mean vector of a multivariate process.In this research a new multivariate control chart procedure based on multivariate exponentially weighted moving average(MEWMA)statistics with dimensional analysis(DA)is proposed to monitor the engineering and physical process.A case study was conducted to evaluate the performance of the proposed charts and compare them with traditional MEWMA control charts.Results are illustrated with some straightforward applications of DA.The comparative performance in terms of average run length showed that the proposed DA-MEWMA control charts performed better than the MEWMA control charts in the engineering and physical process situations.