Expert Knowledge and Risk-Based Safety-Critical Equipment Assessment: Use of AI for Life Extension Analysis in Ageing Offshore Facilities

As industrial infrastructure and critical systems age, the reliability of safety systems becomes increasingly crucial in preventing catastrophic failures, minimizing health, safety, and environmental damage, and ensuring production availability. Risk-based safety assessment enables optimal resource allocation for maintaining aging safety-critical equipment (SCE) in the oil and gas industry. The risk-based safety assessment of ageing SCE is often hindered by vagueness, uncertainties, and variability, which can affect the accuracy of risk estimation and subsequent decisions. This study proposes an artificial intelligence (AI)-based approach that utilizes expert knowledge for risk-based safety evaluation employing fuzzy set theory. It offers a strategy for evaluating the risk associated with potential safety and production loss failures concerning identified failure modes in ageing safety systems with minimal variability. Firstly, this study examines material, electronic, and mechanical degradation challenges over time alongside evolving safety standards and regulations. Secondly, it presents the development of a risk matrix to evaluate the criticality level of different items concerning failure modes, considering the probability of potential failure (PoF) and the consequences of such failures (CoF). Thirdly, it presents the development of the Fuzzy Inference System for risk ranking, aiming to minimize variability, with membership functions developed using experts' knowledge. Finally, the proposed framework's applicability, significance, and validity are discussed. The suggested approach enables practicing engineers to evaluate risks with minimal variability in assessing the critical levels of SCE, thereby ensuring consistency in the safety strategy approach.