Quality quandaries: A stepwise approach for setting up a Robust Shewhart location control chart

Shewhart control charts are extensively used in practice to monitor changes in process characteristics from the in-control state. In Phase I, the initial estimate of μ is based on the 10% trimmed mean of the sample trimeans. This estimator is robust against both localized and diffuse disturbances so that the limits of the Phase I control chart are not affected by potential disturbances. The Phase I data are then screened for localized and diffuse disturbances by means of Phase I sample location and individuals control charts. At the end of Phase I, μ is estimated from the screened data using the grand sample mean, thus ensuring efficiency. The stepwise procedure, together with the procedure for estimating the standard deviation delivers an easy-to-implement procedure for setting up a robust monitoring device.