An Estimation Method Using Periodic Inspection of Indicators

TL;DR

This paper introduces a novel estimation method for failure time distributions using periodic indicator data from redundant systems, employing forward recurrence times and interpolation for accurate estimation.

Contribution

It presents a new approach to estimate failure time distributions from indicator data obtained through periodic inspections, enhancing reliability analysis methods.

Findings

Method performs well on Weibull distributions

Estimates the failure time distribution accurately

Uses interpolation to refine CDF estimation

Abstract

This paper proposes a new approach for estimating the failure time distribution using the indicator data. The indicators, which are checked by periodic inspection of a standby redundant system, only convey whether at least one failure occurs per interval. The estimation procedure first obtains the estimation of the forward recurrence time using the indicator data. Then the mean is estimated based on its relationship with the forward recurrence time. And the estimation of the sampled Cdf is thus derived based on its relationship with the forward recurrence time and the mean. Finally, the Cdf function is estimated using interpolation method. The simulation results showed that the estimation method performed well for the four Weibull distributions.