Premium
On the performance of coefficient of variation charts in the presence of measurement errors
Author(s) -
Tran Kim Phuc,
Heuchenne Cédric,
Balakrishnan Narayanaswamy
Publication year - 2019
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.2402
Subject(s) - ewma chart , control chart , statistics , shewhart individuals control chart , chart , coefficient of variation , x bar chart , covariate , variation (astronomy) , mathematics , computer science , process (computing) , physics , astrophysics , operating system
In the literature, coefficient of variation control charts have been introduced under the assumption of no measurement errors. However, measurement errors always exist in practice, and they do affect the performance of control charts in the detection of an out‐of‐control situation. In this paper, we therefore study the performance of a coefficient of variation Shewhart‐type control chart (Shewhart‐CV chart) and also one‐sided coefficient of variation exponentially weighted moving average–type control charts (EWMA‐γ 2 charts) using a model with linear covariates. Moreover, we propose and study the performance of a two‐sided EWMA‐γ 2 chart using a model with linear covariates. Several figures and tables are provided and analyzed to evaluate the statistical performance of these control charts for different sources of measurement errors. The obtained results show that the precision and accuracy errors significantly affect the performance of both the Shewhart‐CV and EWMA‐γ 2 control charts. An example illustrating the use of this study is finally presented.