Sensitivity of Availability Estimates to Input Data Characterization
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING
Pagination or Media Count:
Reliability analysts are often faced with the challenge of characterizing the behavior of system components based on limited data. Any insight into which model input data is most significant and how much data is necessary to achieve desired accuracy requirements will improve the efficiency and cost effectiveness of the data collection and data characterization processes. This thesis assesses potential significant factors in the probabilistic characterization of component failure and repair behavior with respect to the effect on system availability estimates. Potential factors were screened for significance utilizing fractional. factorial and Plackett-Burman experimental designs for several system models developed using an AFOTEC simulation program entitled RAPTOR. Two input data characterization factors were found to have a significant affect on availability estimation accuracy the size of the structure and the number of data points used for component failure and repair distributional fitting. Estimation error was minimized when the structures analyzed were small and many data points in this case, 25 were used for the distributional fittings. Assuming constant component failure rates and using empirical repair distributions were found to be equally effective component characterization methods pertaining to model availability estimation error compared to using automated software fitting tools or wizards. The results of this study also indicate that there is no apparent benefit in concentrating on important components for the highest fidelity distributional fillings.