A Parametric Study of Live Load Distribution Factors for Steel Girder Integral Abutment Bridge: Camera-Ready Guidelines

Abstract

This research examines the relativity of both AASHTO Standard Specification and AASHTO LRFD girder distribution factors (GDF) for use with designing integral abutment bridges. To evaluate the GDFs, the Scotch Road integral abutment bridge was modeled in the finite element software Abaqus/Cae. The model was verified using temperature-displacement data recorded from April, 2003 to May, 2006. Following the validation of the finite element model (FEM), three loading cases including one, two, and three lanes, were run in Abaqus. The stress data obtained from each case was used to calculate the GDF for each girder. Lane one loading provided the most reasonable results for AASHTO LRFD, while the AASHTO standard equation was overly conservative in all cases. The GDF calculated using the finite element model was only about 30% lower than the AASHTO LRFD value for one lane loaded, while being 50% lower than the AASHTO standard specifications. Both AASHTO GDF equations were overly conservative for both two and three lanes loaded. The finite element calculated GDF was approximately 50% and 60% lower for two and three lanes loaded compared to AASHTO LRFD. A limited parametric study was conducted to investigate the effect of number of piles, girder spacing, and boundary conditions on the GDF ratios. The GDF ratios decreased in accuracy as the girder spacing decreased for one lane loaded. A simply supported bridge was compared with an equivalent IAB. It was found that IABs are less conservative than simply supported bridges for interior GDFs in all loading cases