While data-driven reservoir characterization has gained traction among operators, most analyses remain constrained to operator-specific acreage or single-basin datasets. This limited scope may introduce systematic bias in feature importance estimation, particularly for geological parameters. This study leverages a comprehensive dataset of 100,000 horizontal wells across seven major U.S. basins to evaluate the robustness of machine learning approaches in identifying true production drivers and their interactions.

This episode is based on the technical paper ” Cross-Basin Analysis of Production Drivers: Insights from Machine Learning Applied to 100,000 Unconventional Wells” presented at URTeC 2025. Download the full paper here.