Nonparametric Statistical Analysis of the Reliability of a Naval Aviation Propulsion System

Abstract

This paper reports the results of applying lifetime (or reliability database) statistical analysis methods to engine removal data recorded over the initial 8 years’ service of the General Electric F414-GE-400 turbofan engines propelling the F/A-18 E/F Super Hornet in service with the US Navy, with the intent of better understanding the reasons for engine removal and their impact on engine time on wing (TOW), and to gain an insight into the reliability of the modules and components from which the engine is assembled. It was found that ‘‘coloring’’ the engine removals into three classes of reasons for removal enabled lifetime data analysis revealing interesting and useful features of in-service engine reliability. Nonparametric statistical analysis provided actionable information on engine removal probability as a function of TOW and removal cause that should be applicable to planning flight operations, line maintenance, and support logistics. The analysis of engine removals due to hot section distress appears to disconfirm the presumption of independence between the three classes of removal. Opportunistic maintenance of modules made accessible due to engine removals to service other modules may significantly affect the observed engine removal distribution of the module of interest, i.e., competing risk masks the underlying module hazard functions.