Answers are grounded in this paper and cite their sources.
Shewhart-type control charts, despite their simplicity and satisfactory performance in detecting large shifts, are not sufficiently sensitive to small process changes. On the other hand, in many real-world applications, an assignable cause simultaneously affects both the location and dispersion parameters of the process. Furthermore, despite advances in data-collection technologies, measurement errors remain unavoidable, and their effects on control-chart performance should therefore be considered. Motivated by these considerations, this paper has two objectives. First, two enhanced versions of the multivariate generalized likelihood ratio (MGLR) control chart are developed by incorporating Klein’s 2-of-2 and 2-of-3 sensitizing rules to accelerate the detection of small shifts in the process mean, dispersion, or both. Second, the improved charts are modified under an additive covariate model to account for the effect of measurement error. The effectiveness of the proposed monitoring schemes, and the adverse effects of measurement error are evaluated through extensive simulations, considering various out-of-control scenarios and error variance levels using average run length ( ARL ), standard deviation of the run length ( SDRL ), and median run length ( MRL ). Additionally, an extensive sensitivity analysis is conducted to evaluate the effects of subgroup size ( \(n\) ) and process dimensionality ( \(p\) ), revealing that increasing the sample size substantially improves the detectability of all charts, whereas process dimensional growth degrades their detection sensitivity. The results indicate that measurement error diminishes the sensitivity of all control charts to process changes, even at low levels of error variance. Nevertheless, the 2-of-2 and 2-of-3 schemes substantially outperform the classical MGLR chart across all shift scenarios and error variance levels, with the 2-of-3 scheme performing slightly better than the 2-of-2 counterpart.
No takes yet. Share an insight, caveat, or question.
Maleki et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: