The torsional rigidity of tub-girder makes them ideal for use in curved bridges. The use on skewed support applications by bridge designers is limited as the behavior is complex and requires the use of advanced analysis tools. In consequence, a simplified analysis method to account for the effects of skew on tub-girder twist rotations and internal torques and how these affect the internal component forces was proposed. The combined effects of skew and curvature are studied by examining the results for analysis with different levels of sophistication for 18 representative bridges. The data generated constitutes the first systematic study on a large set of curved and skewed tub-girder bridges using consistent, refined 3D FEA models to model construction forces and deformations. Comparisons of the simplified analysis method to the refined 3D FEA analysis display the limitations of the simplified analysis and present potential sources of error. Furthermore, the results from the 3D FEA helped identify interactions between components and, therefore, an improved simplified procedure was proposed to account for the effects of the resulting increased stresses. In addition, the bridge erection procedures are discussed and specific examples illustrating the calculation of the fit-up forces is presented. These findings provided additional tools for the analysis process and erection engineering to account for the effects of skew. Lastly, further research needs considering the analysis of additional loading conditions and construction procedures are described.

The exposure from a localized vehicle fire has been observed to produce excessive damage onto highway bridge structural elements including complete collapse of the infrastructure. The occurrence of a fire beneath a bridge can lead to significant economic expense and loss of service even if the bridge does not collapse. The focus of the current research is to assess and evaluate the effect of realistic localized fire exposures from vehicles on the bridge structural integrity and to guide future development of highway bridge design with improved fire resistance. In this research, the bridge structural element response was predicted through a series of three loosely coupled analyses: fire analysis, thermal analysis, and structural analysis. Two different types of fire modeling methodologies were developed in this research and used to predict the thermo-structural response of bridge structural elements: one to model the non-uniform exposure due to a vehicle fire and another to predict response due to a standard uniform furnace exposure. The vehicle fire scenarios required coupling the computational fluid dynamics (CFD) code Fire Dynamics Simulator (FDS) with Abaqus while the furnace exposure scenarios were all done within Abaqus. Both methodologies were benchmarked against experimental data. Using the developed methodologies, simulations were initially performed to predict the thermo-structural response of a single steel girder-concrete deck composite assembly to different local, non-uniform fires and uniform standard furnace fire exposures. The steel girder-concrete deck composite assembly was selected since it is a common bridge design. Following this, a series of simulations were performed on unprotected highway bridges with multiple steel plate girders and steel tub girders subjected to localized fires. The analyses were used to evaluate the influence of a fire scenario on the bridge element response, identify the factors governing the failure of bridge structural elements subjected to a localized fire exposure, and provide guidance in the design of highway bridge structural elements against fire hazards. This study demonstrates that girder geometry affected both the dynamics of the fire as well as the heat transfer to the bridge structural elements which resulted in a different structural response for the bridge. A heavy goods vehicle (heat release rate of 200 MW) and tanker fires (heat release rate of 300 MW) were predicted to cause the bridge to fail due to collapse, while smaller fires did not. The geometric features of the plate girders caused the girder elements to be exposed to higher heat fluxes from both sides of the girder resulting in collapse when exposed to a HGV fire. Conversely, the closed feature of the box girder does not allow the interior surfaces to be in direct contact with the flames and are only exposed to the internal reradiation from surfaces inside the girder. As a result, the single and double lane tub girder highway bridge structure does not fail due to a heavy goods vehicle fire exposure. / Ph. D.