System reliability prediction prioritization strategy

Abstract

A procedure is presented to prioritize system reliability prediction activities once a preliminary reliability prediction has been determined. Time and budgetary constraints impose limitations on the extent of analyses and testing needed to determine component reliability estimates. Therefore, it becomes necessary to allocate limited resources in accordance with their system-level impact. The variance of the system reliability estimate is decomposed so that the effect of individual components within the system can be compared and ranked. The reliability prediction prioritization index (RPPI) is defined to allow a comparison of components and to provide a relative ranking of components that can be used to separate the components into two priority groups. The separation is based on whether a decrease of the component reliability estimate variance meaningfully decreases the system reliability estimate variance. The procedure is demonstrated on an automatic train control example.