Abstract
This paper evaluates the uncertainties and quality of bearing capacity factor prediction models of shallow foundations. The development of bearing capacity factor prediction models is a field of extensive research and many different models have been proposed. Sixty models with different modeling approaches such as the analytical model, semi-empirical model, empirical model, finite difference model, upper bound limit model and lower bound with finite element model etc. are connected through a statistical framework that aids in uncertainty quantification and model quality evaluation. First, uncertainty in the estimation of input parameters studies is performed using multivariate information through multiple correlations, in order to determine the parameters that contribute to the uncertainties of the model prediction. Second, the uncertainties of the bearing capacity factor prediction for all models are compared and significant differences are revealed. Due to the consideration of parameter and model uncertainties, a measure for the total variation of the model response is achieved. Results show that the more inaccurate the input parameters are, the more uncertain the quality of the estimated model prediction becomes. With increasing model uncertainty, the quality of the model also decreases. It has been found that the quality of the model decreases as the friction angle increases. A comparison of the models using total model uncertainty appears to be a reliable and economical method for selecting a stochastic model