A challenging aspect ofa system reliability assessment is integratingmultiple sources of information, such as component, subsystem, and full-system data,along with previous test data or subject matter expert (SME) opinion. A powerfulfeature of Bayesian analyses is the ability to combine these multiple sources of dataand variability in an informed way to perform statistical inference. This feature isparticularly valuable in assessing system reliability where testing is limited and only asmall number of failures (or none at all) are observed.The F-35 is DoD’s largest program; approximately one-third of the operations andsustainment cost is attributed to the cost of spare parts and the removal, replacement,and repair of components. The failure rate of those components is the drivingparameter for a significant portion of the sustainment cost, and yet for many of thesecomponents, available estimates of the failure rate are poor. For many programs, thecontractor produces estimates of component failure rates based on engineering analysisand legacy systems with similar parts. While these estimates are useful, the actualremoval rates provide a more accurate estimate of the removal and replacement ratesthe program will experience in future years.In this document, we show how we applied a Bayesian analysis to combine theengineering reliability estimates with the actual failure data to estimate componentreliability. Our analysis technique also allows for us to overcome the problems of caseswhere few or no failures have been observed. We are able to show that combining theengineering knowledge of reliability with the observed operational reliability results inboth a more informed estimate of each individual component’s reliaiblity and a moreinformed estimate of overall F-35 maintenance costs.The technique presented is broadly applicable to any progam where multiple sourcesof reliability information need to be combined for the best estimation of componentfailure rates, and ultimately of sustainment costs.
Suggested Citation
Medlin, Rebecca M, and V. Bram Lillard. Bayesian Component Reliability Estimation: An F-35 Case Study. IDA Document NS D-10561. Alexandria, VA: Institute for Defense Analyses, 2019.