Probabilistic analysis of site-specific load-displacement behaviour of cement-fly ash-gravel piles

Abstract

Cement-fly ash-gravel (CFG) piles comprise one of the novel ground-improvement methods. They are attracting considerable attention, as their beneficial role in accessible soft foundations is being increasingly exploited by the surging urban industrial development in China. Despite their recognised significance, however, the load-settlement behaviour of CFG piles is poorly understood, particularly in the context of probabilistic assessments. This study focuses on six full-scale CFG pile load test datasets, consisting of 245 samples collected at six sites in the Beijing region, under static axial compression loading. As a part of this investigation, a regression curve is applied to the load-settlement data for each load test in the database using a two-parametric hyperbolic or power law curve-fitting equation. Moreover, an assessment of the multiple load-displacement curves, based on the full set of pile load measurements conducted at a particular test site, reveals that the scatter observed in the regression parameter values is mainly caused by the inherent soil variability.Thus, a bivariate copula-based mixed distribution is chosen to represent the dependence between these regression parameters. A simple copula-based simulation model is used to estimate the reliability index at any specific allowable settlement for the serviceability limit state (SLS) design. The correlation coefficients in the copula-based distributions of the regression parameters are proven to have an impact on the reliability index of this pile foundation. A scatter analysis of the load-displacement behaviour provides insight into the probabilistic design of site-specific CFG pile foundations.