A Comparative Study of Reinforced Soil Shear Strength Prediction by the Analytical Approach and Artificial Neural Networks

For the prediction of shear strength reinforced soil many approaches are utilized which complex and they depend on laboratory tests several parameters. In this study, we aim to investigate compare ability Gray Ohashi (GO) model Artificial Neural Networks (ANNs) predict soil. To achieve objective, work was divided into two parts. first part in order evaluate impact different fiber reinforcing parameters behavior soil, direct experiments were carried out. The results revealed a significant improvement values with reinforcement. increase is function length, proportion, direction. second part, used our experimental study develop ANN model. obtained agree reasonably well experiment ones, very acceptable error (RMSE =1.714, MAE=5.981, R2= 0.960, E = -1.601%). comparative showed that more accurate statistically stable than GO model, took all conditions one equation. On other hand, does not take reinforcement failure uses equations.