Bayesian framework for reliability prediction of subsea processing systems accounting for influencing factors uncertainty

Abstract

A Bayesian network probabilistic framework is suggested for reliability prediction of two conceptual subsea processing systems. A suitable reliability prediction method for mechanical equipment is selected and then illustrated stepwise to estimate the total failure rate of the main subsea equipment from reliability data available for similar offshore topside equipment. First, a Failure Mode and Effects Analysis is conducted, then the main Reliability Influencing Factors on equipment failure are identified and the corresponding influence diagrams are developed. Bayesian network models are developed to describe the uncertainty on the Reliability Influencing Factors and to assess their effect on the equipment failure and subsequently on the reliability of the subsea processing systems. A sensitivity analysis is then conducted to evaluate the relative importance of Reliability Influencing Factors on the equipment and systems reliability. The paper contributes with a highly flexible approach that is suggested to model the uncertainty on the expert judgments when predicting the failure rates of new equipment operating in new scenarios, which can be updated as more information becomes available or more knowledge is provided by experts.