Monitoring process variation using modified EWMA

Aamir Saghir; Muhammad Aslam; Alireza Faraz; Liaquat Ahmad; Cedric Heuchenne

doi:10.1002/qre.2576

Monitoring process variation using modified EWMA

Aamir Saghir, Muhammad Aslam, Alireza Faraz, Liaquat Ahmad, Cedric Heuchenne

Research Center Steyr

Research output: Contribution to journal › Article › peer-review

20 Citations (Scopus)

Abstract

A new control chart, namely, modified exponentially weighted moving average (EWMA) control chart, for monitoring the process variance is introduced in this work by following the recommendations of Khan et al.¹⁵ The proposed control chart deduces the existing charts to be its special cases. The necessary coefficients, which are required for the construction of modified EWMA chart, are determined for various choices of sample sizes and smoothing constants. The performance of the proposed modified EWMA is evaluated in terms of its run length (RL) characteristics such as average RL and standard deviation of RL. The efficiency of the modified EWMA chart is investigated and compared with some existing control charts. The comparison reveals the superiority of proposal as compared with other control charts in terms of early detection of shift in process variation. The application of the proposal is also demonstrated using a real-life dataset.

Original language	English
Pages (from-to)	328-339
Number of pages	12
Journal	Quality and Reliability Engineering International
Volume	36
Issue number	1
DOIs	https://doi.org/10.1002/qre.2576
Publication status	Published - 1 Feb 2020

Keywords

control charts
modified EWMA statistic
Monte Carlo simulation
process variation
run length

Access to Document

10.1002/qre.2576

Cite this

@article{0c0caa04dd3841338f1bbb1a98eacc5d,

title = "Monitoring process variation using modified EWMA",

abstract = "A new control chart, namely, modified exponentially weighted moving average (EWMA) control chart, for monitoring the process variance is introduced in this work by following the recommendations of Khan et al.15 The proposed control chart deduces the existing charts to be its special cases. The necessary coefficients, which are required for the construction of modified EWMA chart, are determined for various choices of sample sizes and smoothing constants. The performance of the proposed modified EWMA is evaluated in terms of its run length (RL) characteristics such as average RL and standard deviation of RL. The efficiency of the modified EWMA chart is investigated and compared with some existing control charts. The comparison reveals the superiority of proposal as compared with other control charts in terms of early detection of shift in process variation. The application of the proposal is also demonstrated using a real-life dataset.",

keywords = "control charts, modified EWMA statistic, Monte Carlo simulation, process variation, run length",

author = "Aamir Saghir and Muhammad Aslam and Alireza Faraz and Liaquat Ahmad and Cedric Heuchenne",

note = "Publisher Copyright: {\textcopyright} 2019 John Wiley & Sons, Ltd.",

year = "2020",

month = feb,

day = "1",

doi = "10.1002/qre.2576",

language = "English",

volume = "36",

pages = "328--339",

journal = "Quality and Reliability Engineering International",

issn = "0748-8017",

publisher = "Wiley",

number = "1",

}

TY - JOUR

T1 - Monitoring process variation using modified EWMA

AU - Saghir, Aamir

AU - Aslam, Muhammad

AU - Faraz, Alireza

AU - Ahmad, Liaquat

AU - Heuchenne, Cedric

PY - 2020/2/1

Y1 - 2020/2/1

N2 - A new control chart, namely, modified exponentially weighted moving average (EWMA) control chart, for monitoring the process variance is introduced in this work by following the recommendations of Khan et al.15 The proposed control chart deduces the existing charts to be its special cases. The necessary coefficients, which are required for the construction of modified EWMA chart, are determined for various choices of sample sizes and smoothing constants. The performance of the proposed modified EWMA is evaluated in terms of its run length (RL) characteristics such as average RL and standard deviation of RL. The efficiency of the modified EWMA chart is investigated and compared with some existing control charts. The comparison reveals the superiority of proposal as compared with other control charts in terms of early detection of shift in process variation. The application of the proposal is also demonstrated using a real-life dataset.

AB - A new control chart, namely, modified exponentially weighted moving average (EWMA) control chart, for monitoring the process variance is introduced in this work by following the recommendations of Khan et al.15 The proposed control chart deduces the existing charts to be its special cases. The necessary coefficients, which are required for the construction of modified EWMA chart, are determined for various choices of sample sizes and smoothing constants. The performance of the proposed modified EWMA is evaluated in terms of its run length (RL) characteristics such as average RL and standard deviation of RL. The efficiency of the modified EWMA chart is investigated and compared with some existing control charts. The comparison reveals the superiority of proposal as compared with other control charts in terms of early detection of shift in process variation. The application of the proposal is also demonstrated using a real-life dataset.

KW - control charts

KW - modified EWMA statistic

KW - Monte Carlo simulation

KW - process variation

KW - run length

UR - http://www.scopus.com/inward/record.url?scp=85074988721&partnerID=8YFLogxK

U2 - 10.1002/qre.2576

DO - 10.1002/qre.2576

M3 - Article

AN - SCOPUS:85074988721

SN - 0748-8017

VL - 36

SP - 328

EP - 339

JO - Quality and Reliability Engineering International

JF - Quality and Reliability Engineering International

IS - 1

ER -

Monitoring process variation using modified EWMA

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this