Probabilistic Analysis of Laterally Loaded Piles using Commercial Finite Difference Software-PPT Presentation

PPT Presentation

Pdf Summary

The presentation describes <strong>AllPile Randomizer</strong>, a tool that adds Monte Carlo probabilistic analysis to commercial finite-difference software for laterally loaded piles. It uses AllPile’s solver to run many analyses with randomly sampled input properties, then estimates the probability of unsatisfactory performance and reliability index.

The project examined a 50-foot concrete pile embedded in sand and developed a practitioner-oriented workflow for varying soil properties, pile elastic modulus, and water-table depth. Because AllPile permits only ten soil layers, the tool initially uses random-variable inputs rather than spatially varying random fields; its current implementation is limited to one sand/gravel layer.

A parametric study found that uncertainty in corrected SPT values had a substantial effect on the probability of exceeding a one-inch pile-deflection limit. In contrast, varying the water-table depth had relatively little influence. The presentation also demonstrates calibrating deflection limits against reliability: project-specific response thresholds can be assessed using risk-based methods to support safer, more efficient design.

The authors conclude that probabilistic analysis can be incorporated into familiar commercial software and that uncertainty in soil properties should be considered in pile design. They recommend expanding the tool to handle more soil layers and encourage commercial software to better support probabilistic methods and spatial variability.

Keywords

AllPile Randomizer

Monte Carlo simulation

laterally loaded piles

probabilistic analysis

reliability index

soil uncertainty

pile deflection

SPT values

water table depth

finite-difference software