An alternative to the bivariate control chart for process dispersion

In this article, we propose a Shewhart-type control chart for monitoring changes in the process variability of a bivariate process. The sample Gini mean differences based matrix /G/ is used as an estimate of the population variance-covariance matrix /Σ/. The newly proposed control chart, denoted by /G/-chart, is based on the generalized Gini mean differences [image omitted]. For the case of two correlated quality characteristics Y and X, the design structure of the proposed /G/-chart is developed assuming bivariate normality of (Y, X). The performance of the proposed [image omitted]-chart is compared with that of the [image omitted]-chart (a sample generalized variance based control chart).