ABSTRACT
In situations where sophisticated prognostic models are not warranted due to the lower level of criticality or low rates of failure occurrence and/or there is an insufficient sensor network to assess health condition, a statistical reliability or usage-based prognostic approach may be the only alternative. This form of prognostic algorithm is the least complex and requires component/LRU failure historical data of the fleet and operational usage profile data. Although simplistic, a statistical reliability-based prognostic distribution can be used to drive interval-based maintenance practices that can then be updated on regular intervals [4]. Typically, fault and inspections data are compiled, and employed to fit a probability distribution of time-to-failure. Weibull distributions are typically employed for this purpose [5].
