Performance Evaluation of a Large Scale Pile Load Testing Program in Light of Newly Developed LRFD Parameters

Abstract

A large load test program was performed by the Connecticut Department of Transportation (ConnDOT) as part of the Interstate I-95 New Haven Harbor Crossing Corridor Improvement Project. The extensive program included 23 statically load tested piles to failure in unique subsurface conditions containing glacio-deltaic deposits of silts and fine sands. The majority of the piles were dynamically monitored during driving and subsequent restrike tests. This paper focuses on the evaluation of design methodologies, making use of the information related to the piles that were installed in two test areas (A and B). An extensive evaluation is carried out via the static and dynamic testing results. The outcomes are compared with existing pile databases, recommendations developed for the American Association of Highway and Transportation Officials (AASHTO) Load and Resistance Factor Design (LRFD) Specifications for Deep Foundations (NCHRP 507, see Paikowsky, 2004), and current AASHTO Allowable Stress Design (ASD) Specifications (AASHTO, 2002). These comparisons enable assessment of the design and the construction prediction methods as well as the associated resistance factors. An attempt to compare ASD and LRFD parameters is made considering the load treatment of the different methodologies. The accuracy of the static methods used during the design phase (in this specific site) was found to be marginally in agreement with the existing pile databases used for the LRFD parameter development (1 std dev removed). However, for the construction phase, the dynamic method capacity predictions compared well with the existing pile databases. Overall, the LRFD recommendations resulted with a more consistent design methodology when directly compared to the AASHTO ASD Specifications.