Sensitivity of the seismic performance of two typical bridges to the modeling variation of their abutment parameters is investigated and compared through a comparison parameter proposed in this study. One of the bridges is a two-span two-column-bent bridge with seat-type skew abutment and the other is a three-span three-column-bent bridge with seat-type skew abutment. A set of 40 pulse-like ground motions is applied to the bridges for nonlinear time-history analysis.

In the transverse direction, two force-deformation models, which are based on strut-and-tie and sliding shear friction mechanisms, are used. The analytical models are based on the extensive experimental research previously conducted.

A hyperbolic force-deformation model (General Hyperbolic Force Deformation) is used to represent the passive lateral resistance of the abutment backfill. For considering the possible variation in the backfill geotechnical property, three typical abutment backfill is chosen from the exciting data that was collected from multiple highway bridges in California. Two alternative methods are used to account for the effect of the abutment skew angle on the backfill reaction. The methods include an empirical relationship derived from experimental data, and an analytical method developed based on assumed log-spiral soil failure mechanism.

A comparison parameter is proposed in this study, which is a representative of ductility demand of a skewed bridge to the same non-skewed bridge. For each case, the parameter is computed using the data derived from the nonlinear time-history analysis to investigate the seismic performance of bridges and compare the ductility demand of the specimen bridges.

The outcome of this research reveals the significance of shear keys and abutment backfill on the global response of bridges. The sensitivity of the comparison parameter to the specimen bridges’ geometry is also discussed in detail in this study.