In this paper the economic-statistical design of a S~2 EWMA control chart for the on-line control of process dispersion is investigated. The EWMA chart operates through a control statistic based on a logarithmic transformation of the sample variance to make possible working on an approximately standard normally distributed random variable. The optimal design of the chart is achieved by respecting a statistical constraint related to the in-control Average Run Length (ARL_0). An extensive data set of process scenarios has been investigated and a comparison with a S~2 Shewhart chart is performed: for proper operating conditions and shifts in process dispersion, the obtained numerical results help the decision maker in the selection of the best performing chart.
展开▼