The seismic connectivity of bridge networks, crucial for post-earthquake relief and reconstruction, is strongly affected by the fragility correlation among bridges, which is largely attributed to their similar dynamic characteristics. Previous research merely evaluated the effect of this correlation without addressing its control. This study proposes a connectivity-based seismic design strategy that adjusts design parameters to enhance and reduce fragility correlation on the same and different paths, respectively, while ensuring the bridge safety against the design seismic action. A design method is established, which involves identifying network information, deriving fragility and correlation models regarding bridge design parameters and site-specific earthquake conditions, and designing bridges within the network using these models. This method is demonstrated on a hypothetical network including bridges with isolation bearings. The results show that the fragility correlation mainly arises from the correlated demand related to dynamic characteristics, and impacts connectivity more than the bridge performance. The fragility and correlation of bridges are affected by seismic input angles and bearing parameters, and can be controlled by altering the latter based on the corresponding models. Compared to the network design focusing on fragility reduction, the proposed strategy effectively achieves higher connectivity reliability with lower costs for bridge improvements. © 2025 The Author(s).

Since the 2011 Great East Japan earthquake, tsunami evacuation towers have been constructed in coastal regions based on worst-case predictions to provide additional safety margins against unprecedented tsunamis. However, the extent to which such assumptions effectively reduce casualty risk remains unclear. This study proposes a comprehensive risk-based framework for estimating tsunami casualty risk, incorporating uncertainties associated with seismic and tsunami hazard intensities, structural vulnerabilities, and evacuation behaviors. The risk assessment focuses on the probability of evacuation failure, accounting for the loss of road network functionality due to seismic impacts and variations in evacuation behavior shaped by the conflicting influence of congestion and disaster education. Ground motion-induced damage to road networks prior to tsunami arrival may delay evacuations, potentially leading to casualty rates exceeding initial estimates. Conversely, disaster education can facilitate earlier evacuations in response to seismic shaking, thereby increasing survival rates. As an illustrative example, the proposed framework is applied to a coastal region projected to be significantly affected by the anticipated Nankai Trough earthquake. Furthermore, this study evaluates the appropriateness of incorporating worst-case scenarios in decisions concerning the placement of tsunami evacuation towers.

Tsunamigenic earthquakes have triggered damage to structures and infrastructure systems due to both ground motion and subsequent tsunamis, resulting in a huge amount of disaster waste. Since the disaster waste must be processed and transported through both the waste disposal system (WDS) and the road network system (RNS), the functionality loss of the two systems subjected to spatially correlated ground motion and tsunamis hinders the disaster waste removal. Therefore, disaster waste management must be developed by considering the effects of the cascading hazards on not only the capacity of processing facilities, but also the time required to transport disaster waste to the processing facilities (i.e., the interdependency between WDS and RNS). This paper presents a framework for estimating the disaster waste disposal time considering the interdependency between WDS and RNS under seismic and tsunami hazards. As a measure of a coastal community resilience after the occurrence of a strong earthquake, the probability of waste disposal completion at a time elapsed from the earthquake occurrence is estimated considering the uncertainties associated with predicting the time-dependent functionalities of the two systems. As an illustrative example, the proposed framework is applied to a coastal community in Mie Prefecture, Japan, to investigate the effects of the interdependency between the two systems on the disaster waste disposal time. © 2025 The Author(s)

Structural performance assessments of corroded prestressed concrete (PC) beams using numerical models that account for spatial corrosion distribution and are validated against experimental results remain limited compared to those of reinforced concrete (RC) beams. This study proposes a probabilistic analysis method to evaluate the structural performance of corroded PC beams, incorporating the spatial corrosion distribution of strands and wires. The method is further applied to compare the structural performance of corroded PC and RC beams. Three finite element (FE) models are developed and compared for their accuracy in predicting the structural behavior of PC beams: (a) using the mean steel weight loss of the strand, (b) incorporating the spatial corrosion distribution of the strand, and (c) considering the spatial corrosion distribution of the six outer wires. The model incorporating the spatial corrosion distribution of the six outer wires achieves the highest accuracy, as it effectively simulates the first wire breakage that governs the flexural load-bearing and deflection ductility capacities of PC beams. The probabilistic distribution parameters representing the spatial variability of corrosion are derived from experimental data. Using this distribution, Monte Carlo simulation-based spatial corrosion samples are integrated into the most accurate FE model to obtain the probability density functions (PDFs) of corroded PC beams. The results indicate that PC beams with the same total steel weight loss can exhibit significantly different flexural load-bearing and deflection ductility capacities due to spatial variability, underscoring the importance of a probabilistic assessment. Furthermore, the PDFs of PC members are shifted to the left compared to those of RC members with the same degree of corrosion. Notably, early wire breakage results in lower mean values and standard deviations for the deflection ductility of corroded PC beams compared to RC beams. © 2025 The Author(s)

Combined effects of rainfall and seismic hazards pose significant threats to structures and infrastructure systems. Additionally, climate change is projected to impact the intensity and frequency of future rainfall, increasing the likelihood of landslides. However, evaluating long-term landslide probability under the combined effects of rainfall and seismic hazards, while considering nonstationary climate change, presents significant challenges due to the distinct characteristics of their occurrence processes. This study introduces a novel framework for probabilistic life-cycle landslide assessment that systematically integrates climate change effects on rainfall hazard along with seismic hazard. Probabilistic nonstationary rainfall and seismic hazard models are developed by leveraging stochastic renewal process theory based on occurrence probability and the associated hazard intensity distribution. Slope fragility assessments are conducted for four event scenarios: individual rainfall, individual earthquake, rainfall followed by an earthquake, and an earthquake followed by rainfall, using seepage and equivalent linear analysis through Monte Carlo simulation. Finally, using the total probability theorem, life-cycle landslide probability is numerically evaluated by convolving nonstationary rainfall and seismic hazards with slope fragilities. An illustrative example is provided by applying the proposed framework to a slope in Hiroshima city, Japan, to explore how the combined effects between nonstationary rainfall and seismic hazards impact life-cycle landslide probability. © 2025 The Authors

To enhance the seismic performance of arch-shaped segmental piers, a novel rocking mechanism has been developed to maintain a low-damage state. The proposed system combines a double sliding mechanism with rocking behavior, supported by the arch-shaped pier to achieve self-centering capability. Horizontal cyclic loading tests demonstrate its feasibility and effectiveness. Finite element simulations confirm the reliable activation of rocking behavior under severe seismic excitations, significantly reducing horizontal displacements. Fragility analyses using a simplified model further indicate that the rocking mechanism improves the pier's resilience against short-period and long-period ground motions.

Predicting the performance and remaining service life of deteriorated reinforced concrete (RC) structures based on observational information obtained through methods such as visual inspection and monitoring remains a significant challenge within the structural engineering community. Experimental studies on the process of steel corrosion and its subsequent impact on the structural performance and integrity of deteriorated RC structures have demonstrated the need for improved numerical simulation techniques to accurately replicate observed phenomena. Addressing this complex task requires the integration of various advanced technologies and a proper account of the uncertainties involved in estimating and evaluating each stage of structural deterioration and its progression in both space and time. This paper provides a state-of-the-art review of the methods and technologies applied in the probabilistic remaining service life assessment of corroded concrete structures. These include: (a) corrosion detection methods, (b) experimental investigation and finite element (FE) analysis of the structural performance of corroded concrete components, (c) probabilistic time-dependent reliability assessment using stochastic simulation methods, and (d) Bayesian updating and machine learning techniques. This review aims to highlight the current status and obstacles of these methods and technologies in assessing the remaining service life of corrosion-affected concrete structures.

The present study enhances a previously established framework for estimating the load-bearing capacity of corroded reinforced concrete (RC) beams from corrosion crack width, now focusing on practical application to real RC bridges. The method's improvements include: (1) assessing the feasibility of using corrosion crack images captured by an unmanned aerial vehicle (UAV) for capacity estimation; (2) expanding the artificial database for machine learning models by incorporating new relationships between corrosion crack width and corrosion product thickness in RC beams with various structural details; and (3) testing the model's applicability to large RC beams. Six-meter-long RC beams were corroded via the galvanostatic method, with additional beams used to study the effects of rebar spacing and diameter on corrosion crack width given a steel weight loss. The improved method successfully predicted the probabilistic density function (PDF) of the beams' load-bearing capacity, independent of corrosion levels. The PDF estimated from the corrosion crack images taken by the UAV was compared with that obtained from close-up images. Finally, the requirements for the image quality of corrosion crack widths in relation to load-bearing capacity estimation are discussed.

To enhance the seismic performance of bridges, a novel arch-shaped concrete masonry segmental pier, constructed solely from nylon and concrete, is proposed by integrating an annular double sliding system utilizing a 3D printer. The proposed precast pier leverages the arch-shaped design and double sliding system to remain elastic under substantial horizontal forces. Horizontal cyclic loading tests validate the theoretical hysteretic behavior of the proposed pier. The impact of various design parameters is investigated through experimental methods. Numerical simulation results indicate that the proposed pier significantly reduces the acceleration of the superstructure while maintaining adequate stability, even under severe seismic excitations.

The effects of corrosion methods, galvanostatic (GS) vs. artificial climate environment (ACE), on the relationships between corrosion products and corrosion cracks of corroded reinforced concrete beams are investigated. The experimental results show that the GS method induces smaller corrosion crack width. A novel quantitative detection method of corrosion product thickness using X-ray and digital image processing techniques is proposed for investigating the continuous development of corrosion products. As the corrosion level increases, a smaller steel-to-rust volume expansion ratio caused by continuous leakage and a lower oxidation degree of corrosion products for the GS specimens are the major reasons for the smaller corrosion cracks. Regarding the effect of rebar arrangement on the development of corrosion-induced cracks, the expansion stress from corner-located rebars restrained the growth of corrosion products and further limited the propagation of corrosion-induced cracks of the center rebar. Estimated corrosion crack width via FE analysis using measured corrosion products as input show good agreement with experimental results. © 2024 The Authors