Seismic reliability of steel SMFs with deep columns based on PBSD philosophy

Steel special moment frames (SMFs) represent a common structural solution for buildings in earthquake-prone areas. However, when SMFs are selected by structural engineers, one of their main concerns is to satisfy the permissible drifts recommended by codes. As a common option, the use of deep columns may be a feasible alternative to reduce lateral deformation of SMFs. In this paper, the authors explore the performance of steel SMFs with deep columns by evaluating their seismic reliability. Since earthquake-resistant design is moving from the prescriptive-code to performance-based seismic design (PBSD), an alternative safety approach is integrated with the PBSD philosophy. In this way, SMFs are represented by finite elements and excited by seismic loading incorporating all major sources of nonlinearity as material behavior, geometric deformations, and connections of structural members. The novel approach is developed using the first-order reliability method, response surface method, and an advanced probabilistic scheme. The computational benefits and accuracy of the proposed method are validated using traditional Monte Carlo simulation. The implementation potential is showcased with the numerical evaluation of three 9-story steel SMFs: the first one using columns of medium size and the other two considering deep columns. The seismic reliability is extracted for every model considering serviceability performance functions correlated with performance levels of collapse prevention, life safety, and immediate occupancy. In addition, the contribution of the post-Northridge connection in the steel SMFs is incorporated. Finally, without being too much critical, the use of deep columns in the models of this paper seems to be a step in the right direction to reduce weight, decrease cost, and increase structural reliability. However, it must be stated that deep columns considered in the models have no instability concern because of their low slenderness ratios.