This study aims to analyze and compare the importance of feature affecting earthquake fatalities in China mainland and establish a deep learning model to assess the potential fatalities based on the selected factors. The random forest (RF) model, classification and regression tree (CART) model, and AdaBoost model were used to assess the importance of nine features and the analysis showed that the RF model was better than the other models. Furthermore, we compared the contributions of 43 different structure types to casualties based on the RF model. Finally, we proposed a model for estimating earthquake fatalities based on the seismic data from 1992 to 2017 in China mainland. These results indicate that the deep learning model produced in this study has good performance for predicting seismic fatalities. The method could be helpful to reduce casualties during emergencies and future building construction.

Road networks are one of the vital components of a transportation system that influence the traffic capacity and disaster losses after the earthquakes. The road network reliability is crucial for the postearthquake emergency decision‐making. In this study, a method is proposed to assess the seismic damage of road networks considering individual road damage state and reliability of road networks in emergency conditions. First, we assess the importance of the factors that affect seismic road damage using the AdaBoost algorithm. In addition, artificial neural networks are used to evaluate the damage state of an individual road based on the factors that are selected with higher importance. Then, the improved estimation for the reliability of road networks is adopted to assess the damage of road networks. Last, the method is demonstrated using the road networks in Karamay, China.

This article proposes a seismic fragility analysis framework of bridges, which considers the freeze–thaw cycles and scour comprehensively. The time-varying seismic fragility analysis method is used to analyse the bridge system based on the degradation law of mechanical properties of concrete under the freeze–thaw cycles, combined with the uncertainty of earthquake, scour depth and freeze–thaw cycles. Under the condition of the scour, with the growth of service time, the seismic fragility of piers and bearings increased and the exceeding probability of bearing is higher than the pier, which makes the bearing more easily damaged than the pier in the case of an earthquake, thus reduces the seismic damage of the pier. The plastic hinge of the bridge pier transfers from the immersion zone to the freeze–thaw zone in freeze–thaw cycles. In this process, the curvature in the freeze–thaw zone grows up, dissipates seismic energy and leads the pier’s fragility to decrease before the plastic hinge transfer. When combined with the scours and freeze–thaw cycles, the failure probability of the pier is slightly lower than that of only the scour action due to the phenomenon of plastic hinge transfer.

A hysteretic model is essential for the study of the seismic performance of structures. Due to the ability to capture the component strength and stiffness deterioration, the modified Ibarra–Medina–Krawinkler (ModIMK) hysteretic model has been widely used recently. Previous studies on the ModIMK model have mainly focused on reinforced-concrete (RC) square columns, and very few have focused on circular columns. In contrast, studies on the ModIMK model specifically for corroded circular RC columns have not yet been conducted. For this reason, a database of 35 corroded and uncorroded comparison columns was gathered, and the hysteretic model parameters of each column were calibrated. The ratios of the model parameters of the corroded and uncorroded comparison components are defined as the corrosion-induced deterioration coefficients (CIDCs). After the CIDCs were calculated for all the corroded columns, the empirical equations for the CIDCs were established using regression methods. Through determining the ModIMK model parameters of the corroded columns when they were uncorroded, combined with these equations, the ModIMK model parameters of the corroded circular columns were calculated and their hysteretic models defined. Finally, the accuracy of the ModIMK model’s prediction method was demonstrated using the hysteretic response analyses of four components in the database and the application of two components outside the database. The prediction method lays the foundation for applying the ModIMK model to structures containing corroded circular RC columns.

The heights and height differences of pier structures in different terrains vary, and their damage characteristics under seismic conditions also differ. In order to further explore the influence of pier height on the seismic resistance of rigid frame bridges, this study takes a group of equally high rigid frame bridges with different pier heights and a group of unequally high rigid frame bridges with different pier height differences as the research objects. The piers are taken as the main vulnerable components, and their seismic fragility is calculated and compared through Incremental Dynamic Analysis (IDA) method. The results show that the higher the piers of equally high rigid frame bridges, the smaller the probability of damage under seismic action; when the pier height is increased from 70 meters to 80 meters, the seismic performance of the piers is significantly improved compared to other pier heights.