Risk-based asset integrity management in the oil and gas industry from traditional to machine learning approaches: A systematic review

Abstract

Oil and gas operations are categorized as high-risk because they involve numerous equipment and complex processes. Asset integrity management (AIM) aims to mitigate the risk of failure resulting from degradation with corrosion as the primary cause and to maintain equipment safety and functionality. The risk-based inspection (RBI) methodology is one of the AIM processes that considers risks in decision-making to prioritize inspection and maintenance. This paper provides a comprehensive review of risk-based studies in the context of AIM activities. Risk-based AIM is categorized and reviewed based on risk analysis methods, including quantitative, qualitative, semi-quantitative, probabilistic, deterministic, hybrid probabilistic-deterministic, and dynamic or traditional risk. Most research areas used in case studies focus on pipeline applications. Analysis tools for risk assessment and control applied in risk-based AIM, including the evolution of tools from traditional to machine learning approaches, are examined. The current trends and future research opportunities for applying risk-based AIM are also discussed. This study offers risk assessment models for researchers and oil and gas industry practitioners that fit their specific requirements.