Predicting axial capacity of driven piles in cohesive soils using intelligent computing

Abstract

An accurate prediction of pile capacity under axial loads is necessary for the design. This paper presents the development of a new model to predict axial capacity of pile foundations driven into cohesive soils. Gene expression programming technique (GEP) has been utilized for this purpose. The data used for development of the GEP model is collected from the literature and comprise a series of in-situ driven piles load tests as well as cone penetration test (CPT) results. The data are divided into two subsets: training set for model calibration and independent validation set for model verification. Predictions from the GEP model are compared with experimental data and with predictions of number of currently adopted CPT-based methods. The results have demonstrated that the GEP model performs well with coefficient of correlation, mean and probability density at 50% equivalent to 0.94, 0.96 and 1.01, respectively, indicating that the proposed model predicts pile capacity accurately.