Profile Control Chart based on Maximum Entropy: Accepted - December 2024

Seyedeh Azadeh Fallah Mortezanejad; Ruochen  Wang; Gholamreza Mohtashami Borzadaran; Renkai  Ding; Kim Phuc Tran

Forthcoming

Profile Control Chart based on Maximum Entropy

Accepted - December 2024

Authors

Seyedeh Azadeh Fallah Mortezanejad Jiangsu University https://orcid.org/0000-0002-4358-0768
Ruochen Wang Jiangsu University https://orcid.org/0000-0003-2014-4282
Gholamreza Mohtashami Borzadaran Ferdowsi University of Mashhad https://orcid.org/0000-0002-8841-1386
Renkai Ding Jiangsu University https://orcid.org/0000-0002-7782-8776
Kim Phuc Tran Université de Lille https://orcid.org/0000-0002-6005-1497

Keywords:

linear profile, maximum entropy, control chart, nonparametric chart, average run length

Abstract

Monitoring a process over time is essential in manufacturing. In some cases, process quality is defined by different profile types. The goal is to monitor profile coefficients to detect small undesired shifts in the response variable. Traditional Control Charts (CCs) cannot detect small shifts. This paper addresses the issue using a nonparametric CC based on the Maximum Entropy (ME) principle for Linear Profile (LP) data. The method novelty lies in the use of ME to approximate the joint distribution of profile coefficients. A comprehensive simulation and two real data examples are presented to demonstrate the chart’s applicability and efficient performance based on Average Run Length (ARL).

Downloads

FP_REVSTAT2024_dez1

Published

2024-12-05

How to Cite

Fallah Mortezanejad, S. A., Wang, R. ., Mohtashami Borzadaran, G., Ding, R. ., & Phuc Tran, K. (2024). Profile Control Chart based on Maximum Entropy: Accepted - December 2024. REVSTAT-Statistical Journal. Retrieved from https://revstat.ine.pt/index.php/REVSTAT/article/view/849