Bayesian updating of LRFD resistance factor for CPT-based pile design methods using combined static and dynamic pile load tests

Document Type

Article

Publication Date

4-1-2026

Abstract

This study presents a hybrid Bayesian updating framework for calibrating the resistance factors (ϕ) in Load and Resistance Factor Design (LRFD) of driven piles using CPT-based pile design methods. The method integrates prior regional (or state) distributions of the resistance bias factor with site-specific data from static pile load tests (SPLTs) and dynamic pile load tests (DPLTs). A series of numerical simulations were conducted across varying site spatial variabilities, variabilities of static/dynamic pile load tests, and bias factor combinations to evaluate their effect on the updated ϕ factors. The results demonstrate that early evidence from either SPLTs and/or DPLTs substantially influence the ϕ updates, with the values of static and/or dynamic bias factors exerting a strong pull-on initial calibration outcome. The results also show that the ϕ factors can be reliably estimated with limited testing in homogeneous sites but require more extensive calibration under high spatial variability. This research provides a probabilistic basis platform for rational update of resistance factors (ϕ) in geotechnical engineering practice that supports the need of using Bayesian models in site-specific LRFD calibration.

Publication Source (Journal or Book title)

Transportation Geotechnics

This document is currently not available here.

Share

COinS