AI DRIVEN PREDICTIVE MAINTENANCE IN PETROLEUM AND POWER SYSTEMS USING RANDOM FOREST REGRESSION MODEL FOR RELIABILITY ENGINEERING FRAMEWORK

This study systematically reviews the application of artificial intelligence (AI)-driven predictive maintenance in petroleum and power systems, with a focus on Random Forest regression as a reliability engineering tool. Predictive maintenance, defined as the integration of real-time monitoring with analytical forecasting, has become essential for minimizing downtime, reducing costs, and improving safety in energy infrastructures. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology, 92 peer-reviewed studies published between 2000 and 2024 were identified and analyzed across major databases. The review synthesized literature on conceptual frameworks, including distinctions between corrective, preventive, condition-based, and predictive maintenance, as well as core reliability metrics such as mean time to failure (MTTF), mean time between failures (MTBF), and remaining useful life (RUL). The findings demonstrated that Random Forest regression consistently balanced predictive accuracy, robustness, and interpretability compared with other machine learning methods, including neural networks, support vector machines, and gradient boosting. Applications in petroleum systems emphasized drilling reliability, well integrity, pipeline monitoring, and refinery optimization, while power system studies focused on turbine reliability, transformer fault prediction, renewable energy components, and smart grid stability. The integration of predictive maintenance with Internet of Things (IoT) sensors, digital twins, and cloud-based platforms was identified as a key enabler of real-time reliability analytics. However, persistent challenges remain in terms of scalability, interpretability, and sector-specific customization. This review contributes by consolidating current evidence, identifying research gaps, and offering practical recommendations for enhancing reliability and sustainability in petroleum and power industries.