Critical infrastructure systems (CISs) are increasingly exposed to natural and anthropogenic hazards, and their interdependencies heighten cascading failure risk across multiple sectors. Traditional hazard-specific models inadequately capture systemic complexities and deep uncertainties of CIS disruptions. Moreover, their validation is frequently hampered by limited historical data. As a complementary approach, worst-case disruption analysis offers a disruption-source-agnostic perspective by focusing on component availability post-disruption rather than specific failure mechanisms. These analyses are typically formulated as attacker-operator (AO) models, which identify worst-case disruption scenarios that maximize disruption consequences, or as defender-attacker-operator (DAO) models, which incorporate defense decisions to minimize worst-case disruption consequences. However, the expanding scale of CISs and their increasingly complex interdependencies, driven by rapid urbanization, pose significant challenges to worst-case disruption analysis. In this review, we synthesize recent advancements in worst-case disruption identification and mitigation for CISs within multi-level optimization, categorizing AO and DAO models based on defender strategies, attacker behaviors, and operator responses. We further assess solution algorithms by distinguishing between exact and approximate methods, including heuristic and machine learning-based approaches. By identifying knowledge gaps, we aim to chart future directions for developing scalable, robust, and integrated modeling frameworks to address multifaceted CIS risks from a worst-case disruption perspective.

Railway systems, key drivers of national connectivity and economic growth, are increasingly exposed to meteorological multi-hazards under a changing climate. Despite growing recognition of multi-hazard impact analyses, the mechanisms by which meteorological multi-hazards translate into operational impacts on railway systems remain poorly understood. This study develops an integrated methodology for assessing railway resilience to meteorological multi-hazards, combining a data-driven method to systematically extract railway-relevant multi-hazard events with a resilience model that jointly quantifies physical damage and operational disruption. This methodology is demonstrated using China’s high-speed railway (CHSR) system by analyzing the spatiotemporal impacts of a cold-wave–blizzard event and a compound wind–rain event. Applying the methodology to the CHSR from 2000 to 2024 shows that 87.8% of meteorological events affecting the system are embedded within multi-hazard episodes. Multi-hazard events generate larger annual train delays than any single-hazard event set, with losses reaching up to 8.72 times those caused by Rain-induced single-hazard events in specific years. Spatially, coastal rail corridors are dominated by Wind–Rain events, whereas northern and northwestern rail lines are more vulnerable to cold-related events. These findings highlight the need to incorporate multi-hazard metrics and location-specific climate adaptation strategies into railway resilience planning and operational management.

Amidst the accelerating pace of global urbanization, urban centers play a pivotal role in shaping the functionality and spatial organization of cities. Here, we introduce a systematic framework that combines a suite of morphological and spatial indicators with a harmonized multitemporal dataset derived from GHS-SMOD R2023A settlement layers (spanning 1975 to 2020 at five-year intervals). Applying this framework to 388 Chinese cities, we rigorously validate a series of classical urban expansion theories. In particular, our results confirm that city size distributions, scaling laws relating urban area to population, the exponential decay of radial population density, and the fractal dimension of urban centers largely keep consistent with existing findings. Our analysis further reveals that for 68% of the studied cities, the centroid of the largest patch of urban center fluctuated within 5 km over the past 45 years, indicating a notable spatial stability amid dynamic urban growth. Moreover, we introduce an entropy-based Urban Isotropy Index (UII) to quantify the evenness of urban expansion across all radial directions; the progressive increase in average UII from 2.97 in 1975 to 3.72 in 2020 underscores a trend toward more uniform spatial development. Interestingly, only 42% of the cities exhibited a directional bias with greater area expansion toward adjacent core cities. Finally, our influence analysis suggests that urban economic and geographic factors correlate complexly with expansion patterns. Collectively, our framework not only advances the empirical validation of urban growth theories but also provides a robust reference for future studies on sustainable urban development worldwide.

Natural hazards such as earthquakes, floods, and tropical cyclones pose significant threats to the operation of critical infrastructure systems (CISs) in urban environments. Rapid recovery of post-disaster CISs is essential not only for mitigating immediate socio-economic impacts but also for strengthening urban resilience against future shocks. A key challenge in this recovery process is the efficient scheduling of resources to repair damaged infrastructure, a task complicated by the dynamic and uncertain post-disaster environment, the interdependencies within infrastructure networks, and the diverse priorities and demands of various stakeholders. Given the multifaceted nature of these challenges, numerous repair resource scheduling models have been developed, each incorporating distinct algorithmic strategies tailored to different disaster types and infrastructure systems. Despite a growing body of literature on optimization problems in disaster recovery, a comprehensive understanding of the variations in these models and methods remains lacking. This review aims to systematically explore and synthesize the landscape of repair resource scheduling models, highlighting model variants and their solution algorithms. In particular, it addresses the emerging challenges in post-disaster recovery, exacerbated by the coupled effects of climate change and rapid urbanization. By categorizing the variants and extensions of existing models, this study seeks to refine current frameworks and inspire the development of more comprehensive models, ultimately contributing to more informed restoration decisions and enhanced resilience of urban infrastructure systems.

Accurately calculating travel time-to map transport accessibility and assess transport resilience to inform sustainability-oriented decisions-remains a critical modeling and computational challenge if considering large-scale all-modal transport. Vector-based methods cannot capture movement across off-network areas and are still computationally expensive in some applications for large-scale networks, whereas conventional raster-based methods may bring considerable inaccuracies if computationally acceptable. Here, we propose an optimal local connectivity-based method to rasterize transportation networks and enable a smooth integration of all travel modes to support fastest travel time-based accessibility and resilience analysis. Experimental studies on road networks in cities worldwide, together with theoretical analyses of lattices and simulations of random planar graphs, show its capability for remarkably accurate and rapid estimation of travel time in various network conditions. Successful applications in accurately mapping national-scale accessibility to healthcare facilities and rapidly estimating the worst-case resilience against local disruption demonstrate its utility to support many research and policy needs.

National multimodal transport systems (NMTSs) are vital to intercity mobility and economic development, yet increasingly vulnerable to extreme events. Existing resilience assessments largely focus on single-mode or regional-scale systems and fail to address the resilience of NMTSs, considering their large-scale, multi-modal, timetable-dependent nature. In response to this challenge, the paper proposes a passenger-oriented framework to assess the resilience of NMTSs under extreme events. A dynamic functionality network is constructed, integrating scheduled services, intermodal transfers, and time-dependent travel paths. Based on this network, a resilience model is formulated using daily passenger demand and minimum travel time matrices. Two computational methods, a critical nodes-based method and a segment computation-based method, are designed to ensure computational efficiency. The framework is applied to the coupled road, high-speed rail, and airline system in mainland China under COVID-19 lockdowns and a heavy rainstorm in Zhengzhou. Results show that the daily proportion of affected passengers during city lockdowns ranges from 0.01% to 10.07%, with impacts strongly correlated with city GDP and network size. In the rainstorm case, 6.44% of passengers were affected, with an average delay of 167.12 min. The proposed framework offers a transferable and computationally efficient approach for national-scale resilience assessment and transport planning.

Critical infrastructure systems (CISs) sustain modern societies, yet their interdependencies allow local disruptions to cascade across systems and amplify socio-economic losses. Hazard-specific models represent physical mechanisms but often struggle to capture the full uncertainty and complexity of disruption impacts, while worst-case disruption analysis complements them by identifying upper-bound consequences under the most adverse conditions. However, existing worst-case analyses usually optimize system performance metrics and overlook a logical interdependency created by people who jointly depend on multiple CISs’ services. We propose a people-centric worst-case disruption modelling framework to identify failure scenario that leads to the largest impacts on people under both localized and non-localized disruptions, while capturing the new logical interdependency. Applied to power, gas, water and road-transport systems in a region, results reveal that worst-case impacts and single- versus multi-system outage patterns vary with disruption intensity and interdependency strength. In contrast, traditional performance-centric worst-case analyse identifies different disruption scenarios and underestimates affected populations by up to 114.65%. Sensitivity analyses on CIS topologies and interdependencies, people-centric objective functions, and correlations in service states across zones further demonstrate how input parameters shape worst-case disruption scenarios. Together, these findings underscore the importance of integrating a people-centric perspective into worst-case disruption analyses to inform disaster risk reduction.

Security vulnerability identification is vital for safeguarding industrial control systems (ICSs). In recent years, machine learning (ML) techniques has been increasingly integrated into ICSs to enhance control performance and reliability. However, this integration increases system complexity and cyber-physical interactions, thereby introducing new security risks. Existing vulnerability identification methods for ICS generally identify either cyber vulnerabilities or physical vulnerabilities in isolation, with limited attention to vulnerabilities stemming from cyber-physical interactions. To solve this problem, this study proposes a systems-theoretic approach for identifying security vulnerabilities in ML-enhanced ICS. Building on systems-theoretic process analysis (STPA), the proposed approach incorporates the ML development process into the modeling of functional control structures and applies causal analysis across five dimensions in the security vulnerability derivation. We applied the proposed approach to a reactor-regenerator system (RRS) and found that, compared with existing approaches, it identified at least 19.7 % more vulnerabilities. Finally, three representative security vulnerabilities are validated using an operational training simulation system (OTSS), and their impact and hazard are examined and discussed.

Early warning systems are central to disaster resilience, yet impact-based alerts often falter under cross-agency fragmented mandates, semantic silos, and rigid approval chains that delay protective action. Despite advances in hazard forecasting, few platforms convert predictions into coordinated government response at scale. We present MAESTRO, a multi-agent system that mirrors institutional roles, mediates cross-agency semantics through natural-language reasoning and grounded tool-use, and integrates forecasts, impact models, and situational awareness under human oversight. Across 100 typhoon scenarios, MAESTRO matched expert alert levels in 98% of cases, reduced decision latency by over 85%, and produced reports rated clearer and more actionable by emergency professionals. In a 72-h replay of Typhoon Lekima (2019), earlier warnings enabled relocation of ~180,000 residents with eight additional hours of lead time. Deployed in a provincial government platform for multiple typhoon responses over 1 year, MAESTRO demonstrates one of the first AI-orchestrated early warning systems operating as live infrastructure—offering a scalable, inclusive pathway toward the global goal of Early Warning for All.

Power outages caused by tropical cyclones (TCs) pose serious risks to electric power systems and the communities they serve. Accurate, high-resolution outage forecasting is therefore critical for both proactive mitigation planning and real-time emergency response. Most existing outage prediction models operate in open-loop or event-level settings and cannot update forecasts as storm conditions and system states evolve. To address this limitation, we propose the SpatioTemporal Outage ForeCAST (STO-CAST) model, a spatiotemporal deep learning framework that performs state-dependent, observation-updated rolling inference throughout TC events. STO-CAST enables outage forecasts to evolve in response to updated meteorological projections and newly observed outage information during runtime. The model integrates static environmental and infrastructure attributes with dynamic meteorological and outage sequences and produces hourly outage forecasts at a 4 km by 4 km resolution. STO-CAST supports dual-horizon forecasting, providing short-term nowcasting with a 6-hour lead time for real-time situational awareness and long-term forecasting with a 60-hour lead time to inform proactive planning and resource staging. A case study of Typhoon Muifa (2022), evaluated under a Leave-One-Storm-Out framework, demonstrates the model's operational value, including its ability to track evolving outage hotspots and to provide diagnostic insight through error decomposition that distinguishes the effects of model limitations, meteorological uncertainty, and observation gaps. Overall, STO-CAST offers a scalable and interpretable framework to support risk-informed emergency response and enhance power system resilience under intensifying TC threats.

The complexity of coupled risks, which refer to the compounded effects of interacting uncertainties across multiple interdependent objectives, is inherent to cities functioning as dynamic, interdependent systems. A disruption in one domain ripples across various urban systems, often with unforeseen consequences. Central to this complexity are people, whose behaviors, needs, and vulnerabilities shape risk evolution and response effectiveness. Realizing cities as complex systems centered on human needs and behaviors is essential to understanding the complexities of coupled urban risks. This paper adopts a complex systems perspective to examine the intricacies of coupled urban risks, emphasizing the critical role of human decisions and behavior in shaping these dynamics. We focus on two key dimensions: cascading hazards in urban environments and cascading failures across interdependent exposed systems in cities. Existing risk assessment models often fail to capture the complexity of these processes, particularly when factoring in human decision-making. To tackle these challenges, we advocate for a standardized taxonomy of cascading hazards, urban components, and their interactions. At its core is a people-centric perspective, emphasizing the bidirectional interactions between people and the systems that serve them. Building on this foundation, we argue the need for an integrated, people-centric risk assessment framework that evaluates event impacts in relation to the hierarchical needs of people and incorporates their preparedness and response capacities. By leveraging real-time data, advanced simulations, and innovative validation methods, this framework aims to enhance the accuracy of coupled urban risk modeling. To effectively manage coupled urban risks, cities can draw from proven strategies in real complex systems. However, given the escalating uncertainties and complexities associated with climate change, prioritizing people-centric strategies is crucial. This approach will empower cities to build resilience not only against known hazards but also against evolving and unforeseen challenges in an increasingly uncertain world. (c) 2025 THE AUTHORS. Published by Elsevier LTD on behalf of Chinese Academy of Engineering and Higher Education Press Limited Company. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

The emerging tropical cyclone (TC)-blackout-heatwave compound risk under climate change is not well understood. In this study, we employ projections of TCs, sea level rise, and heatwaves, in conjunction with power system resilience modeling, to evaluate historical and future TC-blackout-heatwave compound risk in Louisiana, US. We find that the return period for a compound event comparable to Hurricane Ida (2021), with approximately 35 million customer hours of simultaneous power outage and heatwave exposure in Louisiana, is around 278 years in the historical climate of 1980-2005. Under the SSP5-8.5 emissions scenario, this return period is projected to decrease to 16.2 years by 2070-2100, a ~17 times reduction. Under the SSP2-4.5 scenario, it decreases to 23.1 years, representing a ~12 times reduction. Heatwave intensification is the primary driver of this increased risk, reducing the return period by approximately 5 times under SSP5-8.5 and 3 times under SSP2-4.5. Increased TC activity is the second driver, reducing the return period by 40% and 34% under the respective scenarios. These findings enhance our understanding of compound climate hazards and inform climate adaptation strategies.

Tropical cyclones (TCs) increasingly threaten the operational stability of high-speed railway (HSR) system, particularly through compound wind-rain hazards, yet their impacts on HSR passengers remain poorly quantified. This study develops the first HSR resilience assessment framework to quantify the effects of TC-induced compound wind-rain hazards on intercity travels, integrating train schedules, meteorological observations, and mobility data. Analysis of Chinese HSR during eight TC events reveals that compound hazards cause 2.96 times more affected passengers and 3.94 times greater delays than wind-only scenarios. Urban-scale amplification ratios reach up to 59.25 times for affected passengers and 790.44 times for delays. Moreover, compound hazards expand the spatial footprint of disruption, shifting impacts from coastal hotspots into inland regions. Sensitivity analyses confirm the persistence of amplification effects across variations in railway response strategies. These findings inform targeted adaptation strategies and offer a scalable framework for railway systems facing intensifying climate-related risks.

With rapid urbanization and the expansion of national transportation system, versatile service facilities have emerged, ranging from urban services meeting daily needs to national facilities catering to specialized demands. Ensuring efficient and equitable access to multi-level facilities is critical for promoting a nation's sustainable development. Despite extensive research on urban facility accessibility, an integrative assessment of multi-level facility access—particularly inter-regional access to national facilities—remains underexplored. To fill this gap, this study proposes a nationwide hierarchical classification system, categorizing facilities into three levels—urban, provincial, and national—based on service range and spatial distribution. Tailored accessibility calculation methods are developed for each level, incorporating road, rail, and air transport. Results from mainland China reveal stark disparities across facility levels: Urban healthcare services exhibit the highest accessibility but significant spatial inequality, whereas national-level theme parks show the lowest accessibility yet the highest spatial equality. Additionally, accessibility determinants display a hierarchical pattern: facility density dominates urban facility accessibility, while administrative area size and GDP govern access to provincial and national facilities, respectively. Furthermore, HSR and air transport enhance access to national facilities while reducing inequality. These findings provide actionable insights for optimizing facility distribution and accessibility across hierarchical spatial scales.

The coupled national-scale railway and airline systems (CRASs) have drastically improved inter-city connectivity and national economics, but they remain susceptible to various extreme events. These events, including flooding and typhoon, often manifest as localized events because all direct interrupted components lie in a small region relative to the large-scale distribution of CRASs. Instead of the hazard-specific modeling of each localized event, this paper introduces four types of localized disruption models to simulate localized events with various locations, impact coverages, and time spans. A multi-perspective framework is then proposed for travel time-based functionality loss assessment of CRASs under localized events. Taking CRASs in China as an application, results demonstrate that (1) the functionality loss is highly sensitive to the event location, and events in areas with more population, higher GDP, and larger transport systems tend to cause higher functionality loss; (2) the impact coverage of events has a limited influence on functionality loss, as critical areas identified under circle-shaped impact coverage keep consistent with those under administrative district-based coverage; (3) the functionality loss and critical areas vary largely with the time span of localized events. The findings provide valuable insights in devising mitigation strategies for CRASs against various extreme events.

Infrastructure systems are indispensable to modern societies, underpinning vital services that sustain everyday life. Nevertheless, these systems are intricately interdependent and frequently vulnerable to various hazards, often manifesting as localized disruptions. This study eschews hazard-specific modeling in favor of a worst-case scenario approach to identify the potential for the most severe damage scenario that could precipitate the greatest loss of performance across these interconnected systems. Traditional identification methods primarily assess post-disaster performance based on the operational states of individual components. However, many authoritative reports emphasize the affected population as a significant consequence of such hazards. Addressing this discrepancy, this study proposes a population-based method to identify critical areas whose component disruptions could impact the largest number of people. This model also facilitates the exploration of new types of logical interdependencies, captured through a population-based objective function, thus providing a complementary tool for the disaster management of urban areas. Applied to Tianjin Eco City, the results demonstrate the dynamic nature of disaster consequences, varying with the intensity of disruptions. The study further elucidates the implications of considering interdependencies within road transport systems and offers a comparative analysis of different modeling approaches for attackers and operators.

Hurricanes have caused power outages and blackouts, affecting millions of customers and inducing severe social and economic impacts. The impacts of hurricane-caused blackouts may worsen due to increased heat extremes and possibly increased hurricanes under climate change. We apply hurricane and heatwave projections with power outage and recovery process analysis to investigate how the emerging hurricane-blackout-heatwave compound hazard may vary in a changing climate, for Harris County in Texas (including major part of Houston City) as an example. We find that, under the high-emissions scenario RCP8.5, the expected percent of customers experiencing at least one longer-than-5-day hurricane-induced power outage in a 20-year period would increase significantly from 14% at the end of the 20th century to 44% at the end of the 21st century in Harris County. The expected percent of customers who may experience at least one longer-than-5-day heatwave without power (to provide air conditioning) would increase alarmingly, from 0.8% to 15.5%. These increases of risk may be largely avoided if the climate is well controlled under the stringent mitigation scenario RCP2.6. We also reveal that a moderate enhancement of critical sectors of the distribution network can significantly improve the resilience of the entire power grid and mitigate the risk of the future compound hazard. Together these findings suggest that, in addition to climate mitigation, climate adaptation actions are urgently needed to improve the resilience of coastal power systems.

The impacts of flash flooding on road transportation, such as travel time delays, are crucial concerns by different stakeholders including neighborhoods and governments at all levels, which calls for a multi-scale flooding impact assessment. Although existing methods have such a capacity, they are not effectively validated, limiting our understanding of the impacts for successful flooding management. This paper develops a multi-scale flooding impact assessment framework and a validation method applied to road transportation in Wuhan, China. Using real-world traffic data in a historical flooding event, the validation results show the capability of the framework in predicting the post-flooding travel time of trips across different scales. The framework is further used to predict travel time delays in flooding at neighborhood, district, and city scales during peak-traffic periods. Results show that while flooding primarily impacts commuting times in central city areas, the integration of traffic accidents in flooding expands the impact to districts and communities farther away from the city center. At the city scale, our analysis indicates that the travel time to workplaces is mostly increased, compared to the increase in travel time to critical facilities. The proposed framework is also adopted to prioritize the treatment of flood-prone sites. These results show the potential of the framework to help flooding response and mitigation.

The efficacy of urban rail transit in reducing the impact of road congestion is the subject of debate in policy and academic communities. Despite huge challenges arising from multifaceted factors such as data limitation, the recently released annual average congestion data enables us to empirically analyze the effect of urban rail transits in terms of their contribution to the average accessibility under road congestion. This paper presents a framework to evaluate the accessibility improvement benefit of urban rail transit under road congestion from a mean-field perspective. A cross-regional analysis in 43 Chinese cities demonstrates that the rail transit networks enhance the potential accessibility and contour accessibility by up to 5.07% and 12.09%, respectively. Besides, the accessibility improvement benefit largely depends on the road congestion level, rail network size, and layout of rail network. Furthermore, simulations on randomly generated rail transit networks indicate that the population coverage and proximity to roads are two structural determinants on the performance of rail transit. The proposed framework, together with the empirical findings, provides insights for city authorities to plan urban rail networks.