Napoved zbirnega števila okvar popravljivega izdelka na podlagi poteka delovanja
Prediction of the cumulative number of failures for a repairable system based on past performance

english
slovenian

2007

volume 53, issue 10

Univerza v Ljubljani, Fakulteta za strojništvo

The prediction of the cumulative number of failures for a repairable system isan important topic in reliability theory. A repairable system may end up in one of the three possible states after a repair: 'as good as new', 'as bad as old' and 'better than old but worse than new'. Current probabilistic models used in repairable system analysis account for the first two states, but they do not properly apply to the last one, which is more common in practice. In this paper, a robust solution to a probabilistic model that is applicable to all of the three after repair states, called generalized renewal process (GRP), is presented. This research demonstrates that the GRP based on an m-fold Weibull mixture offers a general approach to modeling complex repairable systems and discusses application of the EM algorithm to estimation of the GRP parameters. This paper also presents a review of the standard GRP based on two-parameter Weibull distribution. The GRP with m mixture components distributions is compared to the standard GRP by calculating the expected cumulative number of failures and the error function.It is shown that the proposed GRP solution with a Weibull mixture accurately describes the failure data and it is suitable for predicting failures based on the past performance of the system, even when a small amountof failure data is available.

16.09.2009