Probabilistic methods are traditionally used to account for the uncertainty in engineering design. However, conventional probabilistic methods have limitations when representing uncertainty. There is an alternative approach based on bayesian statistical methods that has advantages in treating uncertainty in the geotechnical model for slope design. Probabilistic data analysis using the bayesian approach involves numerical procedures for estimating parameters from posterior probability distributions. These distributions are the result of combining prior information with available data through bayes equation. The posterior distributions are often complex, multidimensional functions whose analysis requires the use of markov chain monte carlo (mcmc) methods. These methods are used to draw representative samples of the parameters investigated, providing information on their best estimate values, variability and correlations. The paper describes a methodology in which typical data from laboratory tests and site investigations are used to define representative distributions of the geotechnical parameters, and the use of these results for the evaluation of the reliability of a slope. The first order reliability method (form) is a common technique used for reliability analyses of geotechnical structures such as slopes and tunnels. The form typically considers predefined probability distributions to represent the variability of uncertain parameters and a limit state surface (lss) defining the condition of failure of the structure. The lss is derived from a performance function that may be available in explicit form, or alternatively, could be approximated with a response surface (rs) for complex models. The paper presents an example of a slope evaluated with a rs based on limit equilibrium analyses with the slope model. The example is used to highlight the advantages of using the posterior distributions from the bayesian analysis for the assessment of the slope reliability using the form approach.
