Statistics Seminars: Stochastic Models for Repairable Systems
6 June 2003 00:00 in CM221
"Interpolating between renewals and minimal repairs of complex systems, we propose a parametric framework of stochastic point processes as mathematical models for reliability and imperfect maintenance. The underlying model for successive inter- failure times is assumed to be a nonhomogeneous Poisson process. We consider extensions to allow for preventive maintenance actions and to determine strategies for optimal scheduling of these actions. We refer to these as the additive (AIM) and geometric (GIM) intensities models. Subjective prior distributions are developed for dealing with unknown model parameters. Our theory is illustrated by application to valve maintenance data from a continuous process industry.
Contact email@example.com for more information