A Comprehensive Integrated AI-Workflow to Optimize Future Well Placement in a Complex Clastics Reservoir, Kuwait

The complexity of heterogeneous reservoir rocks necessitates robust and efficient workflows for accurate reservoir parameter estimation. This study presents an integrated evaluation from core scale analysis to a comprehensive 3D-sector model, aimed at discerning reservoir quality within the Clastics of the Wara Formation in Magwa Field, South East Kuwait. Magwa is one of the three sectors of the Greater Burgan Oil Field, subdivided into the Burgan, Magwa, and Ahmadi sectors based on three structural traps. The mixed fluvial and deltaic succession in Wara Upper along with offshore mudstones and shoreline deposits with limited sand development in Lower Wara, contribute to the overall geological complexity of the field. Among the structural challenges given by the fault system, the existing stratigraphic compartmentalization at small scale, caused by the juxtaposition of the channel fills increase the uncertainty of selecting new well locations. Using only 29 wells, this study leverages extensive datasets from a single well to construct a model of the Cretaceous Wara Formation, employing advanced AI and machine learning techniques with the primary objective to develop a workflow that harnesses machine learning algorithms to optimize future well placements.