STOCHASTIC SIMULATION OF THE SEISMIC RESPONSE OF PILE SUPPORTED WHARF STRUCTURES CONSIDERING INHERENT SOIL VARIABILITY

Abstract

The aim of this research is to study the effect of inherent soil variability on the structural performance of laterally loaded pile supported wharf structures with an emphasis on assessing the current deterministic design practices by using an uncertainty quantification approach. In order to represent the soil pile interaction, the p-y method was incorporated in two distinct modeling approaches: Deterministic approach where the upper and lower bounds p-multipliers were used to represent the current design practices and a probabilistic approach that treats some of the p-y parameters as random variables to represent the inherent uncertainty in soil which will in turn be transformed into stochastic structural response. The outcomes of the first approach were assessed based on the probabilistic distributions drawn from the second approach. Additionally, a performance based design was done for both approaches where the deterministic demand to capacity ratios and their corresponding probabilities of failure were calculated. Moreover, a sensitivity analysis for the correlation length was performed and its effect on the probability of failure was studied. The results showed a compliance between both approaches in most cases except at the Operating Level Earthquake where the lower bound limit didn’t cover the range of the probabilistic distribution. Also, it was shown that when the correlation length increases, the probability of failure decreases.