Assessment of Geotechnical Parameter Uncertainty Using Probabilistic Methods in Slope Stability Analysis

Navarro W.S.A.; Garces T.; Ruiz G. · International Conference on Construction in the 21st Century (2024)

Resumen

This paper shows the results of an evaluation of geotechnical parameter uncertainty using probabilistic methods in slope stability analysis. These evaluations were conducted on a series of uniform soil slope profiles, considering both static and pseudo-static conditions. The obtained results were then compared to those derived from deterministic analyses applied to the same profiles. The analyses were conducted using limit equilibrium methods and the Mohr-Coulomb criterion, implemented through RocScience Slide v6.0 software. The simulations for each scenario focused on identifying circular slip surfaces, utilizing two widely employed methods in professional practice: the Spencer method and the Morgen-stern-Price method. In the probabilistic analyses, input parameters were modeled using a normal distribution. The reliability and probability of failure were deter-mined using the Latin Hypercube Sampling (LHS) and the First Order - Second Moment (FOSM) methods. Additionally, the methodology described here was applied to a real case. The results showed that not every slope considered stable, analyzed only from a deterministic approach, is safe, even when it meets the minimum safety factor (SF) requirement.

Instituciones: Universidad de Piura