Risk Assessment Models for Pipeline Infrastructure Failure

Pipeline failures caused by aging, corrosion, and damage may result in irreparable societal, economic, and environmental consequences. Therefore, pipeline safety and integrity are crucial to the robustness of modern societies. However, current pipeline risk analysis and management have some limitations in predicting pipeline holistic failure for integrity management. In this paper, we develop a machine learning-based risk model to predict the failure type of pipelines. We train and test the risk models based on the historical data from a report covering 50 years of spillage data in Europe. Various factors, such as pipeline diameter, age, pipeline location, and land use, are considered in the models. We propose nine risk models and apply the support vector machine (SVM) approach to obtain the best model type with the highest prediction accuracy. The best risk model will provide pipeline operators with an efficient and effective way to evaluate pipeline conditions and guide pipeline rehabilitation and maintenance procedures.