A Bayesian Reliability Growth Grounded Assurance Test

• Activity Number: 370
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Defense and National Security
• Author(s): Rebecca Dickinson* and Kassie Fronczyk and Alyson Wilson and Caleb Browning and Laura Freeman
• Companies: Institute for Defense Analyses and Institute for Defense Analyses and North Carolina State University and North Carolina State University and IDA
• Keywords: Assurance Testing ; Combining Information ; Defense Acquisition ; Reliability ; Reliability Growth

Abstract

Reliability growth, which tracks the change, and ideally improvement, in the reliability of a system as it moves through test events and undergoes periodic corrective actions, is a primary objective in the Department of Defense test community. Future test events should be planned based on the current reliability and growth of the system, as estimated from observed test data. The proposed framework is motivated by data from the Joint Light Tactical Vehicle (JLTV), a family of vehicles with common parts intended to replace one-third of the legacy high-mobility multipurpose wheeled vehicle (Humvee) fleet. The JLTV has been through a series of three test events as it has been developed. We present a Bayesian hierarchical model to leverage the information learned from the three phases to determine the length of testing needed to demonstrate a given reliability threshold in subsequent testing. The methods employed are similar to the assurance testing as discussed by Hamada et al. (2008), but exploit the reliability growth context and incorporate potential covariates.