Design and Analysis of Adaptive Homogeneously Weighted Moving Average Control Chart

Adaptive control charts have recently gained more attention in industries due to their superiority in detecting mean shifts of different sizes compared to classical control charts. This study proposes an alternative efficient control chart for monitoring shifts within a range of sizes in the process mean. The proposed chart, known as an adaptive homogeneously weighted moving average (HWMA) hart, is a homogeneously weighted moving average that is adaptively optimized through an unbiased estimator of the mean shift with the help of an exponentially weighted moving average (EWMA) statistic, such that the smoothing parameter of the chart depends on an estimate of the mean shift. We structure the chart and examine its performance in terms of run-length profiles. Through the Monte Carlo simulation study, we compare the run-length profiles of the chart with those of existing charts that can detect mean shifts of different sizes. The simulation results reveal that the proposed chart performs better at detecting shifts of different sizes than its competitors. Lastly, we apply the proposed chart to a real dataset characterized by the substrate’s manufacturing process.