MINIMIZATION OF CORROSION IN CARGO OIL TANKS USING ARTIFICIAL NEURAL NETWORK MODELING

ABSTRACT Corrosion in cargo oil tanks remains a persistent challenge in the maritime and petroleum industries, posing serious risks to structural integrity, environmental safety, and operational efficiency. Traditional corrosion control methods often rely on reactive maintenance strategies, which can be costly and insufficient in preventing long-term degradation. This study investigates the application of Artificial Neural Network (ANN) modeling as a predictive and preventive tool for minimizing corrosion in cargo oil tanks. By analyzing critical variables such as temperature, humidity, chemical composition, exposure time, and coating conditions, the research develops a data-driven model capable of predicting corrosion rates and identifying high-risk conditions. The methodology involves data collection from simulated and real-world conditions, preprocessing, ANN training, and validation. Results demonstrate that ANN models can accurately forecast corrosion trends, allowing for proactive maintenance decisions. The findings suggest that integrating ANN into corrosion management systems can significantly reduce maintenance costs, enhance safety, and extend the lifespan of cargo oil tanks. This research contributes to the growing body of knowledge on intelligent maintenance systems and highlights the potential of machine learning in industrial corrosion management.