A resilient S2 monitoring chart with novel outlier detectors.

While researchers and practitioners are seamlessly trying to develop methods for minimizing the effect of outliers in control charts, detecting and screening these outliers continue to pose serious challenges. Keeping in view, the researchers rely on robust estimators to modify the detection limits structure so that the chart can be more sensitive against outliers. In this study, we propose a robust S2${S}^2$ control chart based on Pn${P}_n$, MAPD,$\ MAPD,$AADM$AADM$, GMD$\ GMD$, PSD$PSD$, and τau$\tau au$ estimators, whilst the process parameter is estimated from Phase‐I. Through intensive Monte‐Carlo simulations, the study presents how the estimation of parameter(s) and presence of outliers affect the efficacy of the S2${S}^2$ chart, and then how the proposed outlier detectors bring the chart back to normalcy by restoring its efficacy and sensitivity. Average ARL$ARL$ properties are used as the performance measures. The AARL$AARL$ properties establish the superiority of the proposed scheme over MAD$\ MAD$ and Tukey's outlier detectors. The applicability of the study includes the effectiveness of the proposed detectors in industrial data set but is not limited to manufacturing industries. [ABSTRACT FROM AUTHOR]