Optimization Design of EWMA Charts with Variable Sampling Intervals Based on Taguchi´s Loss Function

In order to reduce quality loss of the production process, this paper proposed the optimization design of EWMA charts with variable sampling intervals (VSI). Using Rayleigh distribution to approximate to probability distribution of the random process mean shifts, we determined the parameters of the VSI EWMA charts such that the overall mean of Taguchi´s loss function (ML) is minimized. The results show that ML-VSIEWMA charts in this paper is significantly superior to the conventional VSI EWMA charts in terms of the overall loss of ML. The optimization design method which proposed in this paper can reduce the quality loss greatly and the general idea can be applied to other control charts.