Abstrakt
When operating a device which is a subject to degradation, we want to estimate the distribution of the time to failure for maintenance optimization. Our aim is to describe the dependency of the failure time distribution on applicable regression variables.
Models commonly used in survival analysis, such as the Cox model or the Accelerated failure time model, need to be adjusted to accommodate repairs and maintenance. For instance, we may use the number of repairs or maintenance actions or their cost as time-varying covariates.
In this work we describe such models and demonstrate their application on real data.