Statistical modeling of maintenance on offshore oil and gas installations

Abstract

In this thesis it has been analyzed how the maintenance will evolve over time for different systems on oil and gas installations. Several statistical models have been proposed to analyze this, minimal repair models (NHPP), perfect repair models (HPP/RP) and imperfect repair models (ARA$\infty$/ARA1). The focus has been on the relationships between these models, the state which the system is left in after maintenance and how good each model fit the given data. The maximum likelihood method has been used for all models when finding estimates for the parameters.

The result after fitting the models to the data are also used to simulate how maintenance will evolve during a 30 year period for a specific plant.