A nonparametric Bayesian network approach to assessing system reliability at early design stages. (March 2018)

• Record Type: Journal Article
• Authors: Lee, Dongjin; Pan, Rong
• Is Part Of: Reliability engineering & system safety. Volume 171(2018)
• Page Start: 57
• Page End: 66
• Publication Date: 2018-03
• Subjects: Product design -- Complex system -- Graphical models -- Copula -- Reliability prediction; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2017.11.009
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: NPBN for evaluating system reliability at early design stages is proposed. This copula-based framework is more flexible and computationally manageable. Expert opinion can be augmented by the historical data from parent products. The prognosis and diagnosis functions of the NPBN are demonstrated. Abstract: It is important to predict a system's reliability at its early design stages because modifying design to improve reliability and maintainability at a later time in the system's lifecycle will be costly and, oftentimes, impossible. However, this early prediction is challenging because of the lack of reliability data and the incomplete knowledge of a complex system's reliability structure. To tackle this problem, this paper presents a nonparametric Bayesian network approach. Employing nonparametric Bayesian network, the limitation of discrete Bayesian network can be overcome, and it can be used as a useful tool for decision support. The proposed methodology is applied to a case study to demonstrate its prognostic and diagnostic capabilities.