Designing Exponentially Weighted Moving Average Control Charts under Failure Censoring Reliability Tests

Mohammadipour, P.; Farughi, Hiwa; Rasay, Hasan; Arkat, Jamal

doi:10.5829/ije.2021.34.11b.03

Cited by 1 publication

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, Abbasi et al [8] presented a new monitoring scheme for non-parametric profiles using an adaptive Exponentially Weighted Moving Average (EWMA) control chart. This control chart, EWMA is developed under a type II censoring life test by mohammadipour et al [9]. For the sake of brevity, other related research about profile monitoring is referred to Maleki et al [10], in which an overview is performed on research published during the period 2008-2018.…”

Section: Introductionmentioning

confidence: 99%

A Robust Control Chart for Monitoring Autocorrelated Multiple Linear Profiles in Phase I

Sogandi

Amiri

2023

IJE

View full text Add to dashboard Cite

Many problems do not have one or more variables that determine quality characteristics. In these situations, as a solution method, a profile is descibed by linking independent variables to the response variable. One of the common assumptions in most monitoring schemes is the assumption of independent residuals. Contravention of this assumption can lead to misleading results of the control chart. On the other hand, when the data are contaminated, the classical methods of estimating the parameters do not perform well. Such situations require robust estimation methods. Hence, this paper proposes a robust method to estimate the process parameters for Phase I monitoring autocorrelated multiple linear profiles. The developed control chart is appraised in the absence and presence of contaminated data through comprehensive simulation studies. The results showed that the robust estimator decreases the impact of contaminated data on the performance of the proposed control chart for all outlier percentages and shift magnitudes. Generally, in all three scenarios, including outliers in the model parameters and error variance, the robust approach performs better than the comparative method.

show abstract

Section: Introductionmentioning

confidence: 99%