Performance-based Foundation Design Using a Probabilistic Framework

1. Home
2. Publications
3. Online Library
4. Performance-based Foundation Design Using a Probabilistic Framework

This article presents a framework for quantifying the probability of exceeding a performance criterion in a geotechnical foundation design. The novelty of the proposed framework is that it takes into account the non-linear behaviour of the soil-foundation-system, which allows the prediction of the probability of a predefined state of a foundation. This is achieved by integrating a normalized stiffness-deformation curve into a mathematical response surface that describes the dependence of the deformation on a set of independent random variables. Each variable is defined by a probability distribution function including mean value and standard deviation. The non-linear curve and the response surface are calibrated in a deterministic parametric study. By means of the first-order reliability method, the combined effect of all independent random variables is evaluated, and partial safety factors are determined for a predefined target probability of exceedance of a given performance criterion. The proposed framework can be applied for all design limit states including, but not limited to, serviceability, foundation stiffness, and foundation capacity. In this article, the framework is demonstrated for a serviceability limit state of a monopile foundation for an offshore wind turbine.

5th International Symposium on Frontiers in Offshore Geotechnics (ISFOG2025)

8 - Monopile design to lateral monotonic loads