A MODEL TO DETECT OIL THEFT IN NIGERIAN PIPELINES USING ARTIFICIAL INTELLIGENCE

This study delves into the critical issue of pipeline oil theft in Nigeria, a nation heavily reliant on its oil industry. By harnessing the power of artificial intelligence (AI), mainly through the utilization of artificial neural networks, the research aims to counteract the economic and environmental impact of oil theft. The study conducts an extensive analysis of various machine learning models, focusing on accuracy, actual positive rate, false negative rate, and Receiver Operating Characteristics (ROC). Leveraging feature scaling, permutation, and advanced techniques like PCA, the study develops an Artificial Neural Network (ANN)-based model to detect pipeline oil theft. The new system showcases success in detecting pipeline oil theft incidents with an accuracy rate of 90.75%. The integration of crucial attributes like time, pressure, flow rate, and temperature, enriches the model's precision, even if it leads to a slightly moderated ROC value. The study contributes to the broader knowledge by emphasizing the potential of A.I to secure natural resources, reduce losses, and foster the sustainability of Nigeria’s oil industry.