Statistical aspects of the F/A-18 AGE Exploration Program

摘要

Implementation of the AGE Exploration Program (AEP) for F/A-18 aircraft by the Naval Air Systems Command involves sampling fleet leader aircraft emphasizing inspection of selected structural components. Sample size, and the interpretation of sample results, are the subjects of this report. When the objective of sampling of is reliability estimation, one can, in addition to single point estimates, construct confidence bounds for fleet reliability. These reflect the quality of the estimate in terms of how big a sample was taken. In AEP inspection to date, the usual sampling result is that no discrepancies are found, hence point estimates of reliability are 1.0. The functional relations and graphs developed in this report permit one to, for the case of a discrepancy-free sample, place a lower bound on fleet reliability as a function of how many aircraft were inspected. During inspection, some discrepancies may go undiscovered. When this happens, sampling results overstate reliability. In this paper a method is developed to adjust sample size or reliability estimates to account for the chance of inspection error, and curves are provided to simplify this adjustment