Bayesian updating of reliability of civil infrastructure facilities based on condition-state data and fault-tree model
Journal
Reliability Engineering and System Safety
Journal Volume
94
Journal Issue
12
Pages
1962-1974
Date Issued
2009
Author(s)
Abstract
This paper considers a difficult but practical circumstance of civil infrastructure management-deterioration/failure data of the infrastructure system are absent while only condition-state data of its components are available. The goal is to develop a framework for estimating time-varying reliabilities of civil infrastructure facilities under such a circumstance. A novel method of analyzing time-varying condition-state data that only reports operational/non-operational status of the components is proposed to update the reliabilities of civil infrastructure facilities. The proposed method assumes that the degradation arrivals can be modeled as a Poisson process with unknown time-varying arrival rate and damage impact and that the target system can be represented as a fault-tree model. To accommodate large uncertainties, a Bayesian algorithm is proposed, and the reliability of the infrastructure system can be quickly updated based on the condition-state data. Use of the new method is demonstrated with a real-world example of hydraulic spillway gate system. © 2009 Elsevier Ltd. All rights reserved.
Subjects
Bayesian update; Civil infrastructure; Inspection; Life-cycle analysis; Reliability
Other Subjects
Arrival rates; Bayesian algorithms; Bayesian update; Bayesian updating; Civil infrastructure; Civil infrastructures; Fault-trees; Infrastructure systems; Life-cycle analysis; Novel methods; Poisson process; Real-world; Target systems; Time varying; Bayesian networks; Life cycle; Poisson distribution; Reliability analysis; Time varying systems; Quality assurance
Type
journal article