Advancing Flood Management Strategies: A Review of Agent-Based Models in Flood Risk Assessment

Flooding is one of the most destructive natural disasters worldwide, causing significant socio-economic losses, disruption of critical infrastructure, and loss of lives. The increasing frequency and intensity of floods due to climate change and rapid urbanisation have underscored the need for advanced flood management strategies [1]. While traditional flood risk assessment methods primarily focus on deterministic approaches to predict flood extents and impacts, they often overlook the dynamic interplay between human behaviour and flood dynamics [2,3]. This limitation prevents the development of effective flood management strategies that reflect real-world complexities [4]. This review identifies key research gaps, such as the limited exploration of cascading failures in critical infrastructure and the need for multi-agent collaboration in large-scale flood scenarios. It also outlines opportunities for future development, including the use of synthetic population generation and participatory modelling to enhance the realism and applicability of ABMs.Agent-based models (ABM) have emerged as a transformative tool in addressing these gaps, offering a bottom-up approach to simulating individual and collective behaviours during flood events. By representing individuals, groups, or entities as autonomous agents with distinct decision-making rules, ABMs provide valuable insights into how human behaviors influence, and are influenced by, flood risks and interventions. Recent advancements have enhanced the utility of ABMs, particularly their integration with real-time data, which are sources that enable the dynamic simulation of human mobility and interactions under varying flood conditions. Additionally, the coupling of ABMs with hydrological and flood-forecasting models has created comprehensive frameworks for evaluating proactive and reactive flood management strategies. Despite these advancements, challenges remain in the broader adoption of ABMs. Computational complexity, the need for extensive data to calibrate and validate models, and the difficulty of capturing long-term behavioural adaptations are significant hurdles. Furthermore, there is a growing need for the integration of machine learning and cloud computing methods to improve the scalability, accuracy, and predictive power of ABMs. By providing a detailed evaluation of current methodologies, challenges, and future directions, this study underscores the transformative potential of ABMs in advancing adaptive and resilient flood management strategies. The findings are particularly relevant for policymakers, urban planners, and emergency responders seeking to design targeted, effective interventions that reduce flood impacts and improve community resilience.References[1] Ferdowsi, A., Piadeh, F., Behzadian, K., Mousavi, S., Ehteram, M. (2024). Urban Water Infrastructure: A Critical Review on Climate Change Impacts and Adaptation Strategies. Urban Climate, 58, p.102132.[2] Girottoa, C., Piadeh, F., Bakhtiari, V., Behzadian, K., Chen, A., Campos, L., Zolgharni, M. (2024). A Critical Review of Digital Technology Innovations for Early Warning of Water-Related Disease Outbreaks Associated with Climatic Hazards, International journal of disaster risk reduction, 100, p.104151.[3] Anshuka, A., Ogtrop, F., Sanderson, D., Leao, S.Z. (2022). A systematic review of agent-based model for flood risk management and assessment using the ODD protocol. Natural Hazards, 112(3), pp.2739-2771.[4] Zhuo, L., Han, D. (2020). Agent-based modelling and flood risk management: a compendious literature review. Journal of Hydrology, 585, p.124755.