Prediction of Ultimate Bearing Capacity of Shallow Foundations on Cohesionless Soils: A Gaussian Process Regression Approach

This study examines the potential of soft computing technique—namely, Gaussian process regression (GPR), to predict ultimate bearing capacity (UBC) cohesionless soils beneath shallow foundations. The inputs model are width footing (B), depth (D), geometry (L/B), unit weight sand (γ), and internal friction angle (ϕ). results present were compared with those obtained by two theoretical approaches reported in literature. statistical evaluation shows that presently applied paradigm is better than competing well for prediction UBC (qu). developed GPR a robust qu foundations on soil. Sensitivity analysis was also carried out determine effect each input parameter.