Process location monitoring using a new adaptive memory control chart with Generalized Gauss-Weight function

Score functions are widely used for outlier detection and robust estimation. Recently, they have been incorporated into control charts to improve shift detection performance. The Exponential Weighted Moving Average (EWMA) chart is effective for detecting small-to-moderate shifts in the Process Location Parameter (PLP). Integrating score functions with EWMA further enhances its ability to identify shifts of different magnitudes. This study proposes an Adaptive EWMA (AEWMA) chart based on the Generalized-Gauss-Weight (GGW) function, termed the chart. The GGW function dynamically assigns weights to observations, improving the chart's sensitivity to process shifts. The performance of the proposed chart is evaluated through Monte Carlo simulations using average, median and standard deviation of run lengths. Additional measures, including extra quadratic loss, relative average run length and performance comparison index, are also examined. Numerical results and graphical analyses demonstrate the superior shift-detection capability of the chart, which is further validated through a real-world application.