Bayesian Methods for Estimating System Reliability Using Heterogeneous Multilevel Information

Abstract

Abstract We propose a Bayesian approach for assessing the reliability of multicomponent systems. Our models allow us to evaluate system, subsystem, and component reliability using multilevel information. Data are collected over time, and include binary, lifetime, and degradation data. We illustrate the methodology through two examples and discuss extensions. Supplementary materials are available online. KEY WORDS: Degradation dataHierarchical modelLifetime dataMulticomponent system ACKNOWLEDGMENTS Alyson Wilson's work on this article took place while she was at IDA Science and Technology Policy Institute Washington, DC, and Iowa State University, Ames, IA. Notes 1Assume that a system has only two states: functioning or failed. A system is coherent when if all the components have failed, the system has failed; if all the components are functioning, the system is functioning; if the system has failed, an additional component failure does not cause the system to function; if the system is functioning, replacing a failed component with a functioning component does not cause the system to fail (Rausand and Høyland Citation2004).