Probabilistic analysis of drilled shaft service limit state using the "t-z" method

The utility of the load and resistance factor design (LRFD) approach is being increasingly recognized for the design of drilled shafts. The current LRFD methodologies of drilled shaft design would be more efficient if reliability based design approaches were used for service limit state design. In this paper, the "t-z" methodology is utilized to develop probabilistic approaches for axial service limit state analysis of drilled shafts. Two different models of the soil-shaft interaction are implemented for load displacement calculations: (1) an ideal elastoplastic model, and (2) a hyperbolic model. For both of these soil-shaft interactions, Monte Carlo simulation is used to obtain a large set of load-displacement curves assuming lognormal distributions for the shaft-soil interface properties. The load-displacement curves are analyzed to develop two alternate methods for determining the probability of drilled shaft failure at the service limit state. As a result, cumulative distribution histograms are developed for drilled shaft load capacities at allowable head displacements. These results may be utilized to obtain resistance factors that can be applied to LRFD based service limit state design.