Risk-based operation safety analysis during maintenance activities of subsea pipelines

Abstract Maintenance activities are essential to restoring the structural integrity of damaged subsea pipelines. However, many risk factors exist in maintenance operations and may result in some unexpected incidents. This paper proposes a novel methodology to risk analysis during maintenance operations of subsea pipelines, integrating Job Safety analysis (JSA) with Bayesian network (BN). In this approach, JSA method is used to find risk factors and consequences existing in maintenance operations. These identified factors and consequences are classified into several types. Subsequently, some individual BNs are developed for each type of influencing factor and consequence. Meanwhile, the flowchart of maintenance operation is transformed into a main BN directly. Eventually, a complete BN model is established by adding individual BNs to the main BN. The fuzzy-based approach is utilized to determine the probabilities of basic factors, which are used as the prior briefs in the model to conduct a robust probability reasoning. This methodology could be utilized to perform a safety analysis of maintenance operations. It can also serve as a helpful tool to reduce the potential of accident occurrence. Its applicability and effectiveness are illustrated through two typical cases of maintenance operations.