Machine learning techniques to predict the dimensionless bearing capacity of circular footing on layered sand under inclined loads

The purpose of this research is to use machine learning approaches to make predictions about the dimensionless bearing capacity (DBCp) of a circular footing on layered sand subjected to an inclined load. To achieve this objective, the finite element method was applied to the literature in order to collect 2400 data sets for the circular footing on layered sand under inclination loads. As independent variables, the embedment ratio (u/D), thickness ratio (H/D), load inclination angle (alpha(1)/90 degrees), unit weight ratio of the loose sand layer to the dense sand layer (gamma(2)/gamma(1)) and friction angle ratio of the loose sand layer to the dense sand layer (phi(2)/phi(1)) were used to predict the dimensional bearing capacity (DBCP). The impact of each independent variable on the overall structural integrity was also analyzed through sensitivity analysis. The results demonstrate that load inclination is the primary factor impacting in the dimensionless bearing capacity. Finally, the performance of the machine learning model was assessed by means of six statistical parameters. The developed model was proven to be effective for predicting the dimensionless bearing capacity of the circular footing on layered sand under inclined loading.