Bayesian Estimation for Reliability Engineering: Addressing the Influence of Prior Choice

Abstract

Over the last few decades, reliability analysis has attracted significant interest due to its importance in risk and asset integrity management. Meanwhile, Bayesian inference has proven its advantages over other statistical tools, such as maximum likelihood estimation (MLE) and least square estimation (LSE), in estimating the parameters characterizing failure modelling. Indeed, Bayesian inference can incorporate prior beliefs and information into the analysis, which could partially overcome the lack of data. Accordingly, this paper aims to provide a closed-mathematical representation of Bayesian analysis for reliability assessment of industrial components while investigating the effect of the prior choice on future failures predictions. To this end, hierarchical Bayesian modelling (HBM) was tested on three samples with distinct sizes, while five different prior distributions were considered. Moreover, a beta-binomial distribution was adopted to represent the failure behavior of the considered device. The results show that choosing strong informative priors leads to distinct predictions, even if a larger sample size is considered. The outcome of this research could help maintenance engineers and asset managers in integrating their prior beliefs into the reliability estimation process.

Towards system-theoretic risk assessment for future ships: A framework for selecting Risk Control Options

“…In case the data contains a failure rate, the exponential probability distribution is recommended by (Pui et al, 2017) to account for the randomness of the failure data. The exponential cumulative failure probability is given by (Leoni et al, 2021):…”

A hierarchical Bayesian regression framework for enabling online reliability estimation and condition-based maintenance through accelerated testing

“…In the first model, there are three hyper-parameters (the intercept a, and the slopes b and c), while in the second model, there are two hyper-parameters (the intercept a and the slope b). Adopting non-informative priors is strongly recommended when enough data is available to avoid a powerful influence (of the prior choice) on the posterior distribution (Leoni et al, 2021). For each hyper-parameter, a diffusive normal prior is selected:…”

Predicting Extreme Dynamic Response of Offshore Jacket Platform Subject to Harsh Environment

“…According to the structural reliability analysis basis, the level of the proposed criteria through different standards would experience a change. Reliability analysis has been carried out widely in different engineering fields to improve the systems' serviceability [9][10][11][12][13].…”