Development of New Shaft Resistance Models for Piles Driven in the Puget Sound Lowlands

Early stage design of driven piles primarily relies on static analysis based on one or a combination of total and effective stress approaches, the latter known as the method to estimate axial capacity. A database of 85 piles driven in the Puget Sound Lowland region and monitored at End of Driving (EOD) and Beginning of Restrike (BOR) was compiled along with capacities estimated using the CAPWAP procedure. Established soil profiles based on borings with standard penetration tests (SPT), cone penetration tests, and geologic maps were matched with their corresponding unit shaft resistance profiles from the CAPWAP procedure and resulted in more than 1,200 observations. In this study, eight new models were developed to predict the coefficients at EOD and BOR, based on the corresponding USCS categorization. Each coefficient model is formulated using a three-parameter exponential function, calibrated to capture the profile of coefficients with depth. These models could benefit the designer by providing them with a simple, empirical model that accounts for soil type and setup in a smooth, continuous function with depth. A parallel analysis was performed to assess the accuracy and variability of the formulated prediction models to facilitate implementation in load and resistance factor design (LRFD). INTRODUCTION The design of the axial capacity of driven piles during the planning stages of a project is typically based on a static analysis approach. The goals of this approach are to predict the components of capacity, consisting of the shaft and toe bearing resistances, and to assess project concept feasibility and preliminary lengths. In most cases, this analysis is based on empirically-developed methods, such as the total stress method, also known as the method (Tomlinson 1987) or the effective stress method known as the method (Esrig and Kirby 1979; Fellenius 1991). Many other methods, such as those based on in-situ test results exist and rely on correlations of blow counts from the standard penetration test (SPT), and cone tip and sleeve friction resistance from the cone penetration test (CPT), to shaft and toe pile resistances (Schmertmann 1978; Meyerhof 1976; Nottingham and Schmertmann 1975; Nordlund 1963; Eslami and Fellenius, 1997; De Ruiter and Beringen 1979). Although these methods were generally developed using data from specific geological regions, they are often recommended in national design manuals (e.g., FHWA, Hannigan et al. 2006).