PSI - Issue 78

Stefano Bozza et al. / Procedia Structural Integrity 78 (2026) 1213–1220

1219

Fig. 5: Maximum displacements of load bearings and joints.

The mean longitudinal displacement demand in the load bearings exceeds the nominal capacity in all devices, with mean values about 0.10 m in the first half of the bridge and 0.08 m in the second half, with demand to capacity ratios equal to 2.0-1.60. As highlighted for shear forces, one time history shows very large demand with respect to the other six, and median values are smaller than mean values (respectively, about 0.08 m for the first half and about 0.06m for the second half). Similar results were obtained for the transversal displacements, with higher mean demand in the curved deck than in the straight one. Demand to capacity ratios for transverse displacements show higher values, around 3.1-3.7, due to a very low design capacity. Both longitudinal and transversal displacements in load bearings are not large enough to physically induce loss of support of the deck, since the devices have dimensions of approximatively 0.7m, but once the displacement capacity of the devices is exceeded, their load bearing capacity is no longer guaranteed. More refined analyses should be carried out in order to investigate any residual capacity of the devices. Due to higher displacements in the curved deck than the other one, joint closure is higher in the initial and mid span joints (mean values about 0.09-0.11 m) than the final joint. Although in the design drawing very small gaps were reported between elements (only 3 cm between the deck and the abutments and 6 cm in the mid-span joint), a visual inspection showed gaps of 15 cm (capacity reported in Fig. 5), which is exceeded only in a single time history. Thus, pounding between elements did not represent an issue for the preliminary model of the presented case study. Instead, the nominal displacement capacity of the shear keys (0.06 m) is not sufficient to avoid local pounding nor loss of support in the devices, although the actual capacity of the shear keys should be further investigated. 4. Conclusions The preliminary seismic assessment of an in-service curved PC box girder bridge with a mid-span hinge was evaluated, focusing on load bearings, shear keys and deck joints in order to evaluate devices resistance, pounding between elements or loss of support of the deck. A FE model of the bridge was developed, with frame elements for piers and the box girder, nonlinear fiber-based hinges at the base of each pier, elastic link elements for load bearings and shear keys, and gap link element at deck joints. Both nonlinear time history analyses and simplified static analyses for the spatial variability of the seismic actions were performed for the LS limit state seismic intensity, and maximum forces and displacements were calculated for load bearings, shear keys and deck joints. Then, seismic demands were compared to the nominal capacity of the elements obtained by design drawings of the bridge. Both load bearings and shear keys showed an inadequate nominal resistance against horizontal forces, with demand/capacity ratios higher in unidirectional bearings and shear keys than in fix bearings. Displacements in all bearings exceeded their nominal displacement capacity, not guaranteeing the vertical load-bearing capacity (even though estimated displacements were too small to induce loss of support of the deck). Axial displacements in shear keys exceeded the nominal capacity, and local pounding or loss of support for the devices could be possible. Instead,