Prognostic model for reliability assessment of emergency shutdown valves used in oil & gas facility

The surveillance of Emergency Shutdown Valves (ESVs) used in oil and gas facilities is of paramount importance for both process safety and production assurance. ESVs are vulnerable to degradation and their testing often requires production shutdown impeding production availability. Currently used reliability quantification methods are oversimplified and unable to capture the degradation of ESVSs. In this work, the Markov framework is used to derive the degradation model for ESVs based on the empirical condition monitoring data. The approach shows how to use condition monitoring data as an input for degradation model and derive the output in form of a standard reliability measure Probability to Fail on Demand (PFD). In addition, the proposed phase-type extension of the model improves the predictive capacity of the model compared to common approaches. It has been demonstrated on real condition monitoring data that the proposed approach is competitive with existing methods for quantification of ESVs reliability.