Abstract

In the quality assurance procedure of the deep cement mixing method, the statistical parameters of the unconfined compressive strength of core samples, core strength, are adopted to assess the quality of the cement-treated soil ground. Since the statistical parameters of the core strength are the sample statistical parameters, the statistical uncertainty emerges when estimating the population parameters. Moreover, the spatial correlation of the core strength should be considered on the evaluation of the overall strength of cement-treated soil ground. The paper presents a reliability-based assessment for the deep cement mixing soil column based on the core strength. The analysis method in which the statistical uncertainty included in the core strength and the spatial variability of the strength are considered simultaneously is adopted to calculate the overall failure probability of the cement-treated soil column. The statistical uncertainty is estimated using a Bayesian inference method and the random fields of the strength are generated with the statistical parameters involving the statistical uncertainty. The random finite element method with the generated random fields is performed to simulate the compression failure behaviour of a cement-treated soil column. The analysis result provides the cumulative distribution function of the overall strength of the cement-treated soil column. The reliability-based assessment is performed on the basis of the cumulative distribution function of the overall strength.