Estimating System Reliability from Heterogeneous Data

August, 2016
IDA document: D-5515
FFRDC: Systems and Analyses Center
Division: Operational Evaluation Division
Caleb Browning, Laura Freeman, Alyson Wilson, Kassandra Fronczyk, Rebecca Dickinson See more authors
The reliability of complex systems is often best measured by multiple task-level reliabilities. In this paper we describe the challenge of estimating the reliability of a complex system using the information from task-level reliability measures. We propose an experimental design approach for assessing the impact of different usage profiles on overall system reliability. We consider appropriate models for accounting for correlation between heterogeneous task-level reliabilities and for integrating data from multiple tests.