Premium
Application of weibull proportional hazards modelling to bad‐as‐old failure data
Author(s) -
Love C. E.,
Guo R.
Publication year - 1991
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.4680070306
Subject(s) - weibull distribution , covariate , failure rate , statistics , poisson distribution , hazard , reliability engineering , reset (finance) , parametric statistics , hazard analysis , mathematics , econometrics , engineering , economics , chemistry , organic chemistry , financial economics
In this paper, an analysis is conducted of the equipment failure regime known as bad‐as‐old. In such a regime, the repair of the machine after an unexpected emergency breakdown does not reset the hazard rate. Only a planned overhaul refreshes the hazard rate. This bad‐as‐old regime is modelled as a non‐homogeneous Poisson process. The structure is comprised of a base‐line hazard rate, here assumed to be Weibull, plus a covariate structure, since such (covariate) factors are found to influence equipment failure times. The results are compared to similar results 1 when the underlying failure regime is assumed to be good‐as‐new. The analysis performed here is also compared to a previous analysis 2 in which the base‐line hazard rate was of a non‐parametric form. Several tests are presented as well as an examination of the residuals in order to verify the appropriateness of the model chosen.