Development of Combined Pile-CPT Methods for Estimating the Ultimate Axial Capacity of PPC Piles Driven in Different Soil Categories in Louisiana

Abstract

The cone and piezocone penetration tests (CPT, PCPT) have been widely acknowledged as useful in-situ testing tools for subsurface investigation, characterization of soil type, and evaluation of different soil properties. Because of similarity between the cone and pile, the evaluation of axial pile capacity was one of initial applications of the CPT/PCPT. A previous study conducted by the authors on 80 pile load tests of precast prestressed concrete (PPC) piles demonstrated that some pile-CPT methods are able to predict the ultimate axial pile capacity with better accuracy than other methods. These methods include: Schmertmann, De Ruiter and Beringen, Laboratoire Central des Ponts et Chaussées (LCPC), European Regional Technical Committee 3 (ERTC3), University of Western Australia (UWA), probabilistic, and University of Florida (UF) methods. The results of these seven pile-CPT methods were compared and their performance was examined for different soil categories where different percentages of pile capacity contribution is because of sandy layers. The log-normal distribution of the estimated to measured pile capacity for these pile-CPT methods was adopted to develop combined pile-CPT methods that optimize the estimation accuracy of axial pile capacity in different soil categories. Reliability analysis using Monte Carlo Simulation (MCS) was used to evaluate the resistance factors (ϕ) and efficiency (ϕ/λ_R) of the individual and combined pile-CPT methods. Results of analysis of 80 pile load tests demonstrated the advantage of using the combined pile-CPT methods over the individual methods in relation to improving the accuracy of estimating the ultimate axial pile capacity and having better resistance factors.