Reliability Analysis of Circular Footing by CubiST, MARS and ANN PSO

Reliability analysis of shallow foundation is very important in civil engineering practices. To estimate the ultimate bearing capacity of circular foundation diameter of 2.0 m and depth of foundation 1.3 m three soft computing techniques like Cubing with Statistics Tress (CubiST), Multi Adaptive Regression Spline (MARS) and artificial neural network particle swarm optimization (ANN PSO) were used. To design circular footing, extensive database of soil like cohesion (c), unit weight and angle of internal friction were used to train and validate the models. The performance of these models was tested using different fitness parameters value account for (VAF), root mean square error (RMSE), mean absolute percentage error (MAPE), R
²
, and relative percent difference (RPD). Based on analysis, it was found that ANN PSO model significantly predicts the ultimate bearing capacity of circular footing and used for design. Furthermore, the performances of Cubist model also infer that this model can also be used as new soft computing technique for prediction of circular footing.