EWMA Control Charts for the Percentiles of an Inverse Pareto Distribution Under Complete and Middle‐Censored Data

This paper introduces exponentially weighted moving average (EWMA) control charts for monitoring process percentiles of the inverse Pareto distribution (IPD) using complete and middle‐censored samples. The proposed methodology extends existing Shewhart‐type percentile charts by incorporating the memory property of the EWMA statistic, thereby improving sensitivity to persistent small and moderate shifts in the target percentile. The proposed methodology comprises two complementary frameworks. The first develops a classical EWMA percentile control chart based on maximum likelihood (ML) estimation under complete and middle‐censored sampling schemes, with control limits derived from the asymptotic distribution of the estimators. The second extends this framework within a Bayesian paradigm by assigning a gamma prior to the shape parameter of the IPD. Under complete data, closed‐form posterior inference gives the Bayes estimator of the shape parameter, while under middle censoring a maximum a posteriori (MAP) approximation is obtained with a MAP‐expectation‐maximization (EM) algorithm. The resulting plug‐in estimator of the percentile forms the basis for the EWMA statistic and its control limits, thereby enabling the incorporation of prior information and reducing the impact of parameter uncertainty. The performance of the proposed classical and Bayesian EWMA control charts is evaluated through extensive Monte Carlo simulations. In‐control (IC) performance is assessed using the average run length (ARL) and the standard deviation of the run length (SDRL), while out‐of‐control (OOC) performance is examined under various shifts in the shape parameter of the IPD. The effectiveness of the proposed methods is further demonstrated using a real dataset. The proposed framework provides an efficient approach for monitoring IPD percentiles under complete and middle‐censored sampling in classical and Bayesian settings.

Authors

Institutions

Publication Details

Journal
Quality and Reliability Engineering International
Published
2026-11-18
DOI
https://doi.org/10.1002/qre.70509
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

EWMA Control Charts for the Percentiles of an Inverse Pareto Distribution Under Complete and Middle‐Censored Data

N. J. Joshi, Kashinath Chatterjee, Advait Prashant Rege
Quality and Reliability Engineering International
Advanced Statistical Process Monitoring
article

EWMA Control Charts for the Percentiles of an Inverse Pareto Distribution Under Complete and Middle‐Censored Data

N. J. Joshi, Kashinath Chatterjee, Advait Prashant Rege
article en

Abstract

This paper introduces exponentially weighted moving average (EWMA) control charts for monitoring process percentiles of the inverse Pareto distribution (IPD) using complete and middle‐censored samples. The proposed methodology extends existing Shewhart‐type percentile charts by incorporating the memory property of the EWMA statistic, thereby improving sensitivity to persistent small and moderate shifts in the target percentile. The proposed methodology comprises two complementary frameworks. The first develops a classical EWMA percentile control chart based on maximum likelihood (ML) estimation under complete and middle‐censored sampling schemes, with control limits derived from the asymptotic distribution of the estimators. The second extends this framework within a Bayesian paradigm by assigning a gamma prior to the shape parameter of the IPD. Under complete data, closed‐form posterior inference gives the Bayes estimator of the shape parameter, while under middle censoring a maximum a posteriori (MAP) approximation is obtained with a MAP‐expectation‐maximization (EM) algorithm. The resulting plug‐in estimator of the percentile forms the basis for the EWMA statistic and its control limits, thereby enabling the incorporation of prior information and reducing the impact of parameter uncertainty. The performance of the proposed classical and Bayesian EWMA control charts is evaluated through extensive Monte Carlo simulations. In‐control (IC) performance is assessed using the average run length (ARL) and the standard deviation of the run length (SDRL), while out‐of‐control (OOC) performance is examined under various shifts in the shape parameter of the IPD. The effectiveness of the proposed methods is further demonstrated using a real dataset. The proposed framework provides an efficient approach for monitoring IPD percentiles under complete and middle‐censored sampling in classical and Bayesian settings.

Quality and Reliability Engineering International
Augusta University (US), Indian Institute of Technology Delhi (IN)
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