Forecasting Bearing Capacity, Error Analyses and Parametric Analysis of Circular Footing Seating on the Limited Thick Sand-Layer with Eccentric-Inclined Load

Bearing Capacity (BC) of the soil is one of the crucial parameters to construct any structure. A consistent soft computing models can reduce the cost and time by swiftly generate the required experimental data. This research presents, M5P model tree and feedforward backpropagation ANN model have been used to predict the BC of the circular footing resting on the limited thick sand-layer with eccentric-inclined load. To generate the proposed model, a set of 120 data are gathered from the literature. The results of M5P model tree achieved a coefficient of determination (R2) of 0.96 for both training and testing phases. The Mean Absolute Percentage Error (MAPE) was 19.83% for training and 21.46% for testing. Whereas, for ANN model, R2 is 0.98 and 0.97; MAPE is 18.20 and 16.29 for training and testing, respectively. The R2 and MAPE results reveals that, the ANN model is better substitute method for predict the BC of the Circular Footing (CF) resting on the limited thick sand-layer with eccentric-inclined load than the M5P model. Further, model equations are developed to calculate the BC of the circular footing for the both the methods. Finally, sensitivity analysis concludes that the input parameter ratio of depth of the rigid rough base to width of footing (H/B) is the most influencing parameter to predict the desired output.