Bayesian Life Test Acceptance Criteria for Progressively Censored Competing Risks Data Using Copulas

In many practical situations, more than one failure mechanism may contribute to product failure. Many studies assume independence between the different competing risks of failure. Nevertheless, the assumption of independence is not always justified in various practical applications. When the competi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of reliability, quality, and safety engineering quality, and safety engineering, 2022-12, Vol.29 (6)
1. Verfasser: Salem, Maram Magdy
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In many practical situations, more than one failure mechanism may contribute to product failure. Many studies assume independence between the different competing risks of failure. Nevertheless, the assumption of independence is not always justified in various practical applications. When the competing risks are assumed dependent, it is important to identify models that describe their dependence structure. Copulas are considered a powerful tool to model such dependence structures. This paper addresses the problem of developing Bayesian life test acceptance criteria through two-sample prediction of future observations based on another independent Weibull progressively Type-II censored sample with binomial random removals. It is assumed that unit failure occurs due to only one of two competing risks. Dependence among the competing risks of failure is modeled using Archimedean copulas with nonconjugate prior distributions. A Metropolis–Hastings Markov chain Monte Carlo algorithm is implemented to derive the prediction intervals that define the proposed acceptance criteria. The derived acceptance criteria enable manufacturers to conform to the required quality specifications and help their clients to properly set their quality expectations. A real data example is provided to illustrate the proposed life test acceptance criteria.
ISSN:0218-5393
1793-6446
DOI:10.1142/S0218539322500152