A Hierarchical Nonparametric Bayesian Model That Integrates Multiple Sources of Lifetime Information to Model Large-Scale System Reliability

• Activity Number: 179
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Richard Warr* and Brandon Greenwell
• Companies: and AFIT
• Keywords: Beta-Stacy Process ; Big Data ; Complex Systems ; Dirichlet Process ; Right Censoring

Abstract

The need for new large-scale reliability models is becoming apparent as the amount of available data is expanding at a dramatic rate. Often complex systems have thousands of components. Each of these components and their respective subsystems could have many, few, or no test data. The large number of components creates a massive estimation project that challenges the computational feasibility of traditional reliability models. The solution presented in this work suggests a hierarchical nonparametric Bayesian framework, using beta-Stacy processes. In this Bayesian framework, time-to-event distributions are estimated from sample data (which may be randomly right censored), and possible expert opinion. These estimates can be used to compute and predict system reliability.