Steel and reinforced concrete buildings are popular structural systems. The design of these buildings is regulated by deterministic building codes. In this context, it is established that if building codes are followed, the structure will resist demands without collapsing. However, no regulation is required to control the damage of structures in terms of performance criteria. In this paper, the seismic performance and structural reliability of both steel and reinforced concrete buildings, respectively, are analyzed as a benchmark case of study. Both buildings are designed in an earthquake-prone area for two soil types, respectively. Subsequently, nonlinear dynamic analyzes are conducted and the seismic responses of the models are determined in terms of inter-story drift. To obtain seismic responses, eleven characteristic ground motions of the region are selected corresponding to three performance levels: (1) immediate occupancy, (2) life safety, and (3) collapse prevention, respectively. It was documented that the resulting maximum inter-story drift was much lower than the one obtained from modal analysis. In addition, the risk was computed in terms of reliability index integrating a novel probabilistic approach with performance-based design criteria. According to the results, a small variation in the structural risk among the buildings under consideration is observed. However, buildings designed for rigid soil proved to be more reliable. Additionally, it is observed that the buildings designed with current regulations are too conservative based on the performance criteria limits. Hence, structures located on earthquake-prone areas may be overdesigned when implementing deterministic building codes.

Compared with circular, arched, and pipe-arched soil-steel structures, box-type soil-steel structures (BTSSSs) have the advantages of high cross- utilization and low cover depth. However, the degree of influence of the crown and haunch radii on the mechanical performance of BTSSSs is still unclear. Therefore, two full-scale BTSSS models with a span of 6.6 m and a rise of 3.7 m but with different crown and haunch radii were established, and the mechanical properties during backfilling and under live load were tested. Afterward, 2D finite element models (FEMs) were established using the ABAQUS 2020 software and verified using the test data. The influence of cross- geometric parameters on mechanical performance was analyzed by using the FEM, and a more accurate formula for calculating the bending moment during backfilling was proposed. The results show that the BTSSS with a smaller crown radius has a stronger soil-steel interaction, which promotes more uniform stress on the structure and makes the structure have smaller relative deformations, bending moments, and earth pressure. The span and arch height greatly influence the bending moment and deformation of the structure. Based on the CHBDC, the crown and haunch radii were included in the revised calculation formula.