Simulation approach for assessing the performance of the γEWMA control chart

Purpose The purpose of this paper is to evaluate the performance a modified EWMA control chart ( γ chart), which considers data distribution and incorporate its correlation structure, simulating in-control out-of-control processes select an adequate value for smoothing parameter with these conditions. Design/methodology/approach This based on simulation approach using methodology evaluating statistical methods proposed by Morris et al. (2019). Data were generated from considering two factors that associated data: (1) quality variable skewness as indicator distribution; (2) autocorrelation structure type relationship between observations modeled AR(1). In addition, one factor process was considered, shift in mean. following step, when modeled, fourth intervenes. parameter. Finally, three indicators defined Run Length are used their interactions. Findings Interaction analysis four evidence modeling selection parameters different therefore considerations selected each case must be carefully analyzed. For processes, it better preserve original features (mean variance) calculation limits. It makes sense highly autocorrelated require smaller since enables preservation relevant information past data. Originality/value there has advantages because gathers, single control: modelling characteristics, process. Although other proposals EWMA, none them simultaneously analyze nor allows setting appropriate considered.