An evaluation method for pipeline corrosion risk index weighting in beach and sea oil fields based on combined weighting with improved hierarchical analysis and Bayesian networks

Abstract

Coastal oil and gas pipelines are prone to corrosion due to complex environmental factors, while traditional risk assessment methods often suffer from subjective bias and oversimplified, one-sided weighting strategies. To address these limitations, this study proposes a risk evaluation model based on game theory, combining an improved analytic hierarchy process (AHP) and a Bayesian network for composite weighting. First, a coastal pipeline corrosion leakage risk indicator system is constructed. Next, a gray relational analysis-improved AHP model is introduced for hierarchical weighting. A Bayesian network model is constructed using the T-S fuzzy fault tree, and probability weighting is performed through hierarchical importance analysis. Finally, based on game theory, the above two weightings are combined to calculate the composite weight. Additionally, to verify the model's applicability, a corrosion risk assessment for a coastal pipeline was conducted. The results indicated that the model could effectively mitigates the over-reliance on expert subjective judgment and fragmented analysis of risk factors in corrosion risk assessment for coastal oil and gas pipelines, Critical factors such as high sand content and external coating failure were identified as dominant risks, which were overlooked in conventional models. The evaluation results were generally consistent with the actual situation.