Machine Learning-Based Porosity Cube Prediction from Integrated Well Logs and 3D Seismic Attributes: Egypt’s Offshore Mediterranean

This study focuses on enhancing porosity prediction in a clastic reservoir of Egypt’s Offshore Nile Delta by integrating well log data and 3D seismic attributes through advanced machine learning techniques. Traditional methods often fail to capture heterogeneity and extend interpretations across larger volumes, but by leveraging data-driven approaches such as Ridge Regression, Random Forest, and XGBoost, a high-resolution 3D porosity cube was generated. The results show strong predictive accuracy, aligning with geological interpretations, facies trends, and stratigraphic architectures, while reducing subsurface uncertainty. This workflow improves reservoir characterization, supports well placement, and aids in early-stage field development planning, highlighting the growing role of artificial intelligence in modern upstream workflows