Dynamic risk analysis of marine and offshore systems suffering microbial induced stochastic degradation

This research paper presents a dynamic methodology that integrates the dynamic Bayesian network (DBN) with a loss aggregation technique for microbial corrosion risk prediction. The DBN captures the dynamic interrelationships among microbial corrosion influencing variables to predict the rate of system degradation and failure probability. The model captures the dynamic and time-evolution effect of the degradation propagation on the consequences of failure. The loss aggregation technique is used to forecast the expected economic loss associated with the different loss scenarios. The proposed methodology is tested on a subsea pipeline to assess risks of failure upon microbial corrosion. The outcomes reveal that the interplay among the vital variables results in severe deterioration of the offshore/marine system; thus, it increases the risk in terms of economic losses. Three critical loss scenarios are examined as the consequences of microbial corrosion-induced failure to capture the effect of the soft and hard failures of the safety barriers/actions on the expected total economic loss. At the 95% confidence interval, the upper and lower bound economic losses (value at risk) increase by 40.3% and 57.5%, respectively. The proposed methodology provides a risk-based prognostic tool for offshore and marine systems suffering from microbial corrosion.

Automated summary

This research introduces a dynamic methodology that combines DBN with loss aggregation technique for microbial corrosion risk prediction. The study shows that the interaction among key variables can result in an increase in economic loss.