A New EWMA Control Chart for Monitoring Poisson Observations

Quality control charts based on exponentially weighted moving average (EWMA) has been widely used for monitoring continuous process data. However, many quality characteristics of interest are in the form of counts for nonconformities and are often monitored by a Poisson model. In this article, we introduce a new design structure for the Poisson EWMA charts for monitoring Poisson processes. The proposed scheme is based on a well-structured sampling technique, ranked set sampling instead of the traditional simple random sampling. Using Monte Carlo simulations, we compute the run length properties of the new Poisson EWMA chart and compare their relative performance with the existing schemes for monitoring increases and decreases at the Poisson rate. It is found that the new scheme significantly improves the classical procedures for detecting changes in the Poisson processes. Finally, we illustrate the practical application of the proposed scheme through numerical example.