To effectively coordinate the response to a flood disaster, decision-makers have to prioritise areas that are in most urgent need of assistance. This prioritisation often has to be carried out under time pressure and on the basis of incomplete information, creating a high cognitive load for decision-makers. Methods that integrate Bayesian networks into GIS to draw spatial inference can inform this prioritisation process. However, existing approaches are not equipped to address the time pressure and unclear information-scape that is typical for a flood disaster. In this work, we present a novel spatial inference method for area prioritisation that is designed to address these time and information constraints. The core of this method is a GIS-informed Bayesian network, integrated into an expected loss framework, that can be set up during the preparation phase. The method can then quickly provide area prioritisation recommendations for disaster relief, which has the potential to support decisions-makers during the response phase. In this way, our method provides a means of shifting some of the most time-consuming aspects of the decision-making process from the time-critical disaster response phase to the less critical preparation phase. To illustrate how our method can support rapid and transparent area prioritisation, we present a case study of an extreme flood scenario in Cologne, Germany.

Using novel data and artificial intelligence (AI) technologies in crisis resilience and management is increasingly prominent. AI technologies have broad applications, from detecting damages to prioritizing assistance, and have increasingly supported human decision-making. Understanding how AI amplifies or diminishes specific values and how responsible AI practices and governance can mitigate harmful outcomes and protect vulnerable populations is critical. This study presents a responsible AI roadmap embedded in the Crisis Information Management Circle. Through three focus groups with participants from diverse organizations and sectors and a literature review, we develop six propositions addressing important challenges and considerations in crisis resilience and management. Our roadmap covers a broad spectrum of interwoven challenges and considerations on collecting, analyzing, sharing, and using information. We discuss principles including equity, fairness, explainability, transparency, accountability, privacy, security, inter-organizational coordination, and public engagement. Through examining issues around AI systems for crisis management, we dissect the inherent complexities of information management, governance, and decision-making in crises and highlight the urgency of responsible AI research and practice. The ideas presented in this paper are among the first attempts to establish a roadmap for actors, including researchers, governments, and practitioners, to address important considerations for responsible AI in crisis resilience and management.

Migration is among the most uncertain and contested topics for policymaking. The increasing number of migrants and refugees globally necessitates effective planning and management, particularly in addressing infrastructure needs such as access to healthcare. While efforts to accom- modate a surge of refugees prioritise primary needs, improving structural access to essential infrastructure becomes imperative over time. However, the path-dependent nature of the expansion of refugee settlements poses challenges for infrastructure development. Existing facility location models for infrastructure planning overlook the interplay of infrastructure growth and human behaviour. This chapter presents a study on the interplay between the settling preferences of refugees (behaviour) and the location of healthcare facilities as essential infrastructure. We develop a data-based approach that combines an agent-based model representing decision beha- viour with facility location optimisation models for infrastructure planning. Through a case study of Cox's Bazar, Bangladesh, home to over 1 million Rohingya refugees, we demonstrate the implications of different optimisa- tion approaches and thereby explore how and in how far digital tools influence policymaking on one of the most contested and uncertain topics in the current policy landscape. Our findings underscore the importance of integrating uncertainty about human behaviour in infrastructure decisions.

High Impact Low Probability events (HILPs), often referred to as outliers, are becoming more important in disaster management because they are linked to complex risks and tipping points in interconnected systems. Recent events, such as the cascading effects of the coronavirus pandemic, rising uncertainties from global geopolitical instability, and successive and concurrent extremes driven by climate change, underscore the limitations of relying solely on severe but plausible scenarios for risk practitioners and policymakers. Despite the critical need to integrate HILPs into risk assessment models and emergency preparedness, the field is fragmented, with inconsistent definitions and methodologies. We present a perspective developed under the HORIZON AGILE project (AGnostic risk management for high Impact Low probability Events), which introduces two comprehensive definitions of HILPs and a taxonomy designed to enhance risk assessment, resilience analysis, and crisis management. We provide a validated scientific definition for the academic community and an operational definition tailored for practitioners and stakeholders. Additionally, our taxonomy offers a structured framework to address outlier events that often fall below traditional risk thresholds, ensuring that low-probability, high-impact scenarios with cascading and concurrent dynamics are effectively integrated into risk registers, legislation, and standards development. This study shows how this approach improves methods like stress testing and scenario modelling, especially for the loss of critical services. This empowers policymakers, practitioners, and stakeholders to include more scenarios in their strategies, enhancing resilience and preparedness.

Introduction: The 2021 European floods in Germany, Belgium, and the Netherlands significantly impacted healthcare. With climate change increasing flood risks, healthcare preparedness is essential. Floods affect healthcare directly and indirectly by disrupting patient access, damaging infrastructure and impeding care continuity. Our interdisciplinary research in the Netherlands systematically assesses flood impacts on healthcare, optimises disaster preparedness, patient logistics, and continuity and explores crisis governance, incorporating lessons from coronavirus disease-2019 (COVID-19). Methods: Our multi-sited, interdisciplinary project titled “Pandemic lessons for flood disaster preparedness” includes literature reviews on: (i) the (in) direct impacts of floods on healthcare, (ii) disaster decision-making strategies and (iii) patient logistics during crises. Empirically, ethnographic methods (interviews, focus groups, document analyses, and observations) will: (a) assess hospital flood preparedness, (b) explore decision-making and crisis management strategies and (c) analyse the dynamics of health system governance during floods. Data from these sources and flood scenarios will inform models on healthcare impacts and decision-making, culminating in a simulation game for research and training. Discussion: This study offers a comprehensive, interdisciplinary approach to understanding and improving healthcare system preparedness for floods. By integrating diverse fields such as healthcare governance, disaster risk management, logistics and hydraulic engineering, we provide a unique lens on resilience. A key strength is the incorporation of lessons from the COVID-19 pandemic, allowing us to draw parallels between pandemic response and flood preparedness. In addition, our simulation game serves as a robust tool for translating knowledge into practice. However, the study’s reliance on collaboration with busy healthcare and disaster response professionals may limit engagement. Moreover, the absence of direct public and patient involvement in the research design, though partially mitigated by engaging representative organizations, presents a potential limitation. Lastly, the challenge of obtaining real-time data from flood events could introduce recall bias, but triangulation of various data sources aims to address this issue. Despite these challenges, the study’s integration of long-term data from recent floods and focus on healthcare-specific crisis governance provides valuable insights for improving disaster preparedness.

In emergencies, high stake decisions often have to be made under time pressure and strain. In order to support such decisions, information from various sources needs to be collected and processed rapidly. The information available tends to be temporally and spatially variable, uncertain, and sometimes conflicting, leading to potential biases in decisions. Currently, there is a lack of systematic approaches for information processing and situation assessment which meet the particular demands of emergency situations. To address this gap, we present a Bayesian network-based method called ERIMap that is tailored to the complex information-scape during emergencies. The method enables the systematic and rapid processing of heterogeneous and potentially uncertain observations and draws inferences about key variables of an emergency. It thereby reduces complexity and cognitive load for decision makers. The output of the ERIMap method is a dynamically evolving and spatially resolved map of beliefs about key variables of an emergency that is updated each time a new observation becomes available. The method is illustrated in a case study in which an emergency response is triggered by an accident causing a gas leakage on a chemical plant site.

Urban areas are dynamic systems, in which different infrastructural, social and economic subsystems continuously co-evolve. As such, disruptions in one system can propagate to another. However, open challenges remain in (i) assessing the long-term implications of change for resilience and (ii) understanding how resilience propagates throughout urban systems over time. Despite the increasing reliance on data in smart cities, few studies empirically investigate long-term urban co-evolution using data-driven methods, leading to a gap in urban resilience assessments. This paper presents an approach that combines Getis-ord Gi* statistical and correlation analyses to investigate how cities recover from crises and adapt by analysing how the spatial patterns of urban characteristics and their relationships changed over time. We illustrate our approach through a study on Helsinki’s road infrastructure, socioeconomic system and built-up area from 1991 to 2016, a period marked by a major socioeconomic crisis. By analysing this case study, we provide insights into the co-evolution over more than two decades, thereby addressing the lack of longitudinal studies on urban resilience.

Increasingly, our cities are confronted with crises. Fuelled by climate change and a loss of biodiversity, increasing inequalities and fragmentation, challenges range from social unrest and outbursts of violence to heatwaves, torrential rainfall, or epidemics. As crises require rapid interventions that overwhelm human decision-making capacity, AI has been portrayed as a potential avenue to support or even automate decision-making. In this paper, I analyse the specific challenges of AI in urban crisis management as an example and test case for many super wicked decision problems. These super wicked problems are characterised by a coincidence of great complexity and urgency. I will argue that from this combination, specific challenges arise that are only partially covered in the current guidelines and standards around trustworthy or human-centered AI. By following a decision-centric perspective, I argue that to solve urgent crisis problems, the context, capacities, and networks need to be addressed. AI for crisis response needs to follow dedicated design principles that ensure (i) human control in complex social networks, where many humans interact with AI; (ii) principled design that considers core principles of crisis response such as solidarity and humanity; (iii) designing for the most vulnerable. As such this paper is meant to inspire researchers, AI developers and practitioners in the space of AI for (urban) crisis response – and other urgent and complex problems that urban planners are confronted with.