Uncertainty quantification in the bearing capacity estimation for shallow foundations in sandy soils

In geotechnical engineering, is well known that the theoretical Bearing capacity (BC) differs from the real behaviour of the foundation, especially when theoretical BC was obtained from in-situ test equations or through shear strength properties obtained by laboratory or correlations. The scope of this work is to quantify the uncertainty of the different BC methods of a shallow foundation supported by an anthropic sandy soil tested in the Texas A&M University. Uncertainty was classified according to a degree associated with human errors, laboratory tests, traditional BC models and BC obtained by in-situ test equations. The results show that the choice of the most appropriate friction angle (ϕ') correlation is ambiguous and can generate important uncertainties with possible BC overestimations. As the exploration and/or calculation method becomes more sophisticated (e.g. Finite elements (FE) and Random Finite elements (RFEM)), the bias will be lower. However, as evidenced, the FE and RFEM bias is linked to the quality of the laboratory results and the correlation length estimation in RFEM. Finally, it is concluded that the BC uncertainty is directly related to the model selection and the derivation of the input parameters and not the method degree of uncertainty.