Geotechnical probabilistic analysis and reliability-based design require statistics（such as mean

standard deviation

quantiles

etc.）and probability distributions of geotechnical parameters as input information.However

the number of soil data obtained during geotechnical site investigation is generally too sparse to generate meaningful statistics and probability distribution.This poses a challenge in the implementation of geotechnical probability analysis method and reliability-based design codes in practice.This paper aims to develop a Markov chain Monte Carlo simulation-based Bayesian approach to determine the statistics and probability distributions of undrained Young′s modulus（Eu）using limited information obtained from geotechnical site characterization.The proposed approach integrates

systematically

the prior knowledge and limited site observation data to characterize Euprobabilistically.Finally

the proposed approach is illustrated by using a real case history in this paper.