Design of an EWMA control chart by adaptation of smoothing constant based on a function of estimated shift.

This study introduces a novel Adaptive EWMA (AEWMA) control chart designed to monitor the mean of a normally distributed process with enhanced responsiveness. The proposed methodology dynamically adjusts the smoothing constant based on a proposed continuous function of the estimated mean shift derived from the EWMA statistic. The Monte Carlo simulations are conducted to assess the performance of the AEWMA chart across various magnitudes of process mean shifts, using run-length profiles as the primary evaluation metric. The results indicate that the AEWMA chart outperforms traditional methods in terms of detection efficiency. To demonstrate its practical applicability, the AEWMA chart is applied to a real-world manufacturing dataset, specifically analyzing the flow width resistance of substrates. The findings highlight the efficiency of the proposed chart, making it a valuable tool for improving process monitoring and quality control in industrial environments.