Date of Graduation


Document Type


Degree Type



Statler College of Engineering and Mineral Resources


Civil and Environmental Engineering

Committee Chair

Karl E. Barth

Committee Co-Chair

Fei Dai

Committee Member

Udaya B. Halabe


The use of skew in bridges is becoming increasingly more popular with the number of urban or geographical restraints that require unique abutment and pier orientations. The increasing transportation needs in highly-populated areas require more complicated interchanges, along with the use of skewed or even curved bridges. However, the use of skew complicates the design and performance of the bridge. In straight bridges, girder stress and rotations are fairly easy to predict. However, the use of skew in steel I-girder bridges can cause uneven loading and detailing issues with girders and cross-frames. In particular, skew can result in increased warping, which produces a stress phenomenon known as lateral flange bending.;Lateral flange bending (LFB) is the torsional effect in flanges of an I-section that results from warping. Since the st. Venant torsional stiffness for an open cross-section is low, torsional loads are resisted by the girder in the form of lateral bending stresses. The current AASHTO LRFD Specifications use a fixed-end moment approximation to account for LFB in the design phase. The method assumes that cross-frames act as fixed supports and employs fixed-end moment equations to compute LFB moments in respective unbraced segments. During this study, it was found that this approximation is quite accurate for estimating LFB stresses at cross-frame locations; however, the method tends to overestimate LFB in between cross-frame locations.;Therefore, the goal of this project was to assess the AASHTO LRFD approximation for LFB. To accomplish this, a commercial finite element software package (Abaqus) was employed. The finite element modeling technique was used in several parametric matrices of simple-span bridges to determine the key parameters that affect LFB. Once key parameters were identified and assessed, a modification factor was developed which includes the effect of these parameters and directly adjusts the AASHTO LFB approximation. Observing the data developed in this study, it can be seen that the empirical modification significantly improves the accuracy of the approximation in those regions between the cross-frames, which can improve the efficiency of the design of simple span I-girder bridges.