Abstract Heterogeneity is a vital spatial feature for subsurface resource recovery predictions, such as mining grade tonnage functions, hydrocarbon factor, and water aquifer draw-down predictions. Feature engineering presents the opportunity to integrate heterogeneity information, but traditional engineered features like Dykstra-Parsons Lorenz coefficients ignore context; therefore, are not suf...

Facies cube is a required part of static model, especially concerning fields characterized by complicated geological structure. The important quantitative limitations for modeling are facies proportions in the formation volume. Nowadays these calculated using standard geostatistical methods without considering particular properties data. These specific geometrical characteristics sedimentologic...