Track 3 Session 10
9:10 to 10:10 a.m. Thursday June 11, 2009

A Methodology for Predicting Operational Reliability Growth for a Complex Multi-stage System

Traditionally, reliability growth models are used only during system design to project growth as the system proceeds through the Test-Analyze-And-Fix (TAAF) stages that characterize development, and growth models are not utilized once the system is fielded. In most cases, demonstrated reliability replaces predicted reliability at this point. Single-shot systems such as weapons or launch vehicles present a unique challenge in this area. Often these systems remain in storage or standby for long periods of time, and operational use for these systems is infrequent. Determination of failure cause and subsequent correction during operational use is difficult or impossible for most of these types of systems; therefore the reliability prediction challenges that existed during system development remain throughout the life cycle of a single-shot system. Continued testing of representative articles remains the only method of measuring reliability throughout the life of the system. This presentation uses a Bayesian model to look at continuing the application of reliability growth models throughout the life a single-shot system. The model will be applied to a real system data set to examine the programmatic value that would have been gained by predictions.

Key Words: Reliability Growth, Single-shot systems, Operational Reliability

Jennifer L. Nicholls
United States Navy
Solomons, Maryland