Premium
An EWMA control chart for the multivariate coefficient of variation
Author(s) -
GinerBosch Vicent,
Tran Kim Phuc,
Castagliola Philippe,
Khoo Michael Boon Chong
Publication year - 2019
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.2459
Subject(s) - ewma chart , control chart , multivariate statistics , chart , statistic , statistics , variation (astronomy) , coefficient of variation , computer science , statistical process control , process (computing) , mathematics , physics , astrophysics , operating system
Monitoring the multivariate coefficient of variation over time is a natural choice when the focus is on stabilising the relative variability of a multivariate process, as is the case in a significant number of real situations in engineering, health sciences, and finance, to name but a few areas. However, not many tools are available to practitioners with this aim. This paper introduces a new control chart to monitor the multivariate coefficient of variation through an exponentially weighted moving average (EWMA) scheme. Concrete methodologies to calculate the limits and evaluate the performance of the chart proposed and determine the optimal values of the chart's parameters are derived based on a theoretical study of the statistic being monitored. Computational experiments reveal that our proposal clearly outperforms existing alternatives, in terms of the average run length to detect an out‐of‐control state. A numerical example is included to show the efficiency of our chart when operating in practice.