Adaptive MEWMA charts for univariate and multivariate simple linear profiles

Abstract

The adaptive multivariate EWMA (MEWMA) charts have been widely recognized as an efficient process monitoring tool because of their ability to quickly respond against the shift of various sizes. In this paper, an adaptive MEWMA (AMEWMA) chart is proposed for monitoring the simple linear profile. In addition, new MEWMA and AMEWMA charts are also proposed to monitor the multivariate simple linear profile. The run length characteristics of these control charts are computed using the Monte Carlo simulation method. A detailed comparative study is conducted to compare the detection abilities of the proposed and existing charts when detecting different shifts in the parameters of the univariate and multivariate simple linear profiles. It is found that the AMEWMA chart performs substantially better than the existing charts. An example is given to explain the implementation of the AMEWMA chart.

Keywords:

• AEWMA
• AMEWMA
• average run length
• control chart
• EWMA
• MEWMA
• Monte Carlo simulation
• profile monitoring
• statistical process control

Acknowledgments

The author is thankful to the anonymous reviewer for providing useful comments that led to an improved version of the article.