Analysis of Soil-Steel Bar Mat Pullout Models Using a Statistical Approach

Steel bar mat is a common reinforcing material in mechanically stabilized earth (MSE) walls. The current method to calculate the pullout capacity of these materials is an empirical-based formulation found in AASHTO and Federal Highway Administration (FHWA) design guidance documents. The accuracy of the current formulation is evaluated using a large database of laboratory pullout tests including data that were not available at the time the current model was developed. A new model is proposed, and it is shown to have advantages over the current formulation particularly when pullout capacity design and analysis will be carried out in a load and resistance factor design (LRFD) framework. Calibration of the new model and quantitative comparisons of the accuracy of the two models are evaluated using statistical characteristics of bias values where bias is the ratio of measured pullout capacity to predicted pullout capacity. The accuracy of both models is evaluated for two cases: (1) when project-specific pullout data are not available (the typical case) and presumptive default coefficients must be used; and (2) when project-specific laboratory pullout box data are available. The influence on calibration outcomes of data sets corresponding to markedly different soils and test methodology is demonstrated in the paper. The data reported in the paper are a necessary precursor to future LRFD calibration of the pullout limit state in the internal stability design of bar mat reinforced soil walls.