A probabilistic boundary element method applied to the pile dislocation problem

In this paper a probabilistic approach is presented where the boundary element method is efficiently used to study the effect of a random shift of a given pile within a particular pile cap from its original position–the so-called pile dislocation problem–on selected output design parameters such as pile loads and bending moments in the pile cap. A new circular internal element is developed to simulate the true geometric modeling of piles. The boundary element method for the shear-deformable (thick) plate theory is employed to analyze the pile cap. The plate–pile interaction forces are considered to have constant variation over the circular pile domain. The probabilistic approach presented herein incorporates a Monte Carlo simulation technique for generating random shifts in the original position of a given pre-selected pile. The procedure has been applied to some exemplar pile caps with given pile layouts typically adopted in bridge construction.The results demonstrate that the random dislocation of piles within practical ranges/values as customarily encountered for example in pile caps pertinent to bridge applications will cause limited variations in the output design parameters investigated herein and mentioned above. In other words, it has been illustrated that the resulting dispersion in the output values due to random dislocation of piles is less than the possible intrinsic dispersion that may be practically triggered in the pile locations due to common construction inaccuracies and/or unanticipated problems during pile driving process. The study further emphasizes the efficiency and reliability of the Boundary Elements Method adopted herein for such application.

► Boundary Elements Method is used to study the random shift of a pile in a pile cap.
► Effect of random pile dislocation on pile loads and bending moments is explored.
► Exemplar pile caps typically adopted in bridge construction are presented.
► The study emphasizes the efficiency and reliability of BEM for such application.
► Random dislocations in practice cause limited variations in design parameters.