PSI - Issue 70

Edavalath Nadeem et al. / Procedia Structural Integrity 70 (2025) 19–26

20

1. Introduction Bridges are critical components of a society’s transportation infrastructure, ensuring connectivity and facilitating economic activity. However, they are particularly susceptible to damage during natural hazards such as earthquakes. The failure or loss of functionality of bridges during seismic events can result in substantial socio-economic consequences. Consequently, a comprehensive understanding of the seismic behavior of bridges and the implementation of preventive measures prior to such events is of paramount importance (HAZUZ, 2003; Caltrans, 2019; Chang et al ., 2000). A widely adopted method for evaluating the seismic performance of structures is fragility assessment. Seismic fragility analysis provides a probabilistic framework to quantify the likelihood that a structure or its components will reach or exceed specified damage states under varying levels of seismic intensity. Fragility curves derived from such assessments are invaluable for estimating economic losses, guiding emergency response strategies, and informing retrofitting decisions (Padgett, 2007; Ramanathan, 2012; Dukes, 2013; Mangalathu, 2017; Soleimani, 2017). Traditionally, fragility assessments for bridge columns have focused on engineering demand parameters (EDPs) based on peak structural responses, such as maximum drift and maximum curvature ductility. While effective in characterizing the severity of ground motion demands, these indices often fail to represent the post-earthquake condition of a structure, particularly its repairability or residual functionality (Wei et al ., 2023). For example, reinforced concrete (RC) bridge columns may undergo large displacements during seismic excitation but exhibit minimal residual deformation, thus retaining good self-centering capabilities and post-event usability (Han et al ., 2019). To address this shortcoming, recent research has explored the inclusion of residual displacement as a complementary or alternative EDP in fragility evaluations, offering a more comprehensive understanding of damage states and recovery potential. Nonetheless, using either maximum or residual parameters in isolation may still result in an incomplete representation of seismic damage and structural functionality (Wei et al ., 2023). In this context, the present study incorporates both maximum and residual column curvature ductility as key EDPs to evaluate the combined seismic fragility of the structural system. Recognizing that curvature ductility effectively captures localized deformation demands in bridge columns, this dual-parameter approach offers enhanced insight into seismic performance. A conventional RC bridge model from prior research is adopted, and a nonlinear finite element model is developed using the OpenSees framework. Nonlinear time history analyses are performed using a selected suite of ground motion records recommended in earlier studies (Ramanathan, 2012; Mangalathu, 2017; Srivastava et al ., 2022; Srivastava et al ., 2024). From these simulations, the maximum ( µ ϕ -m ) and residual ( µ ϕ -r ) column curvature ductility values are extracted, and probabilistic seismic demand models (PSDMs) are developed in logarithmic space. Fragility curves corresponding to each demand parameter are then constructed for various damage states. Subsequently, combined fragility curves are established using first-order reliability theory. This integrated approach enables a more refined evaluation of seismic response, leading to improved estimates of the probability of exceeding performance thresholds and offering better support for resilience-based decision-making. 2. Bridge Modelling Details and Seismic Ground Motions As previously mentioned, a RC bridge designed in accordance with the Caltrans Seismic Design Criteria (SDC) is adopted from the prior studies (Caltrans, 2019; Ramanathan, 2012; Mangalathu, 2017; Srivastava et al. , 2022; Srivastava et al. , 2024). The bridge comprises two spans, each measuring 36.58 meters in length, as depicted in Figure 1. The superstructure is supported by a single circular RC column with a height of 6.79 meters and includes a three cell box girder with a depth of 1.46 meters and an effective deck width of 10.75 meters. A detailed nonlinear finite element model of the bridge is developed using the OpenSees framework (McKenna, 2011), incorporating rigorously validated inelastic components. The bridge deck and diaphragm are modelled as linear elastic elements using elastic beam-column elements, assuming that these components remain within the elastic range under seismic loading (Pandikkadavath et al ., 2022; Dukes et al ., 2018; Soleimani, 2020). These elastic components are rigidly connected to the inelastic column cross-sections using massless rigid links, allowing accurate force transmission and realistic modelling of column confinement. The validated inelastic cyclic behaviour of concrete and reinforcing steel in the RC columns is captured using the Concrete07 material model (Chang and Mander, 1994) and the Steel02 model, a modified version of the Menegotto-Pinto model (Menegotto and Pinto, 1973), respectively. The specified design strengths of concrete and steel reinforcement are 33.50 MPa and 464.70 MPa, with corresponding elastic moduli of