Improved MGLR control charts based on Klein’s sensitizing rules for joint mean and dispersion monitoring under measurement error

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.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-16
DOI
https://doi.org/10.1038/s41598-026-81124-6
Primary Topic
Advanced Statistical Process Monitoring
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Improved MGLR control charts based on Klein’s sensitizing rules for joint mean and dispersion monitoring under measurement error

Hossein Eghbali, Mohammad Reza Maleki, Majid Jafari, Mohammad Nikoogoftar
2 links
Scientific Reports
Advanced Statistical Process Monitoring
article

Improved MGLR control charts based on Klein’s sensitizing rules for joint mean and dispersion monitoring under measurement error

Hossein Eghbali, Mohammad Reza Maleki, Majid Jafari, Mohammad Nikoogoftar
article en

Abstract

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.

Scientific Reports
Isfahan University of Technology (IR), Golpayegan University of Engineering (IR), University of Eyvanekey (IR)
Advanced Statistical Process Monitoring
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon