Zylfije Tahiri, PhDc
South East European University, Faculty of Languages, Cultures, and Communications, Department of English Language Studies, Tetovo, North Macedonia
E-mail: zt32329@seeu.edu.mk
ORCiD iD: https://orcid.org/0009-0009-8646-3789
Abstract
Disaster risk management (DRM) is increasingly confronted with the complexities of contemporary hazards, exacerbated by climate change, urbanization, and interconnected infrastructures. This systematic review investigates the transformative impact of artificial intelligence (AI) on predictive modeling, early warning systems, and decision-support frameworks within DRM. By synthesizing peer-reviewed literature, policy frameworks, and regional comparative studies, the analysis demonstrates that AI enhances anticipatory governance through probabilistic hazard forecasting, anomaly detection, and real-time scenario simulations. Machine learning, deep learning, and natural language processing significantly improve the accuracy of predictions for floods, wildfires, landslides, and seismic events. Moreover, AI-driven early warning systems minimize latency and streamline emergency response efforts. Nonetheless, there are notable structural, ethical, and institutional challenges, including data scarcity, algorithmic opacity, and uneven integration across agencies. Comparative evaluations reveal that while Balkan countries are conceptually aligned with European DRM standards, they lag in technological implementation. This study emphasises that the successful integration of AI into DRM relies on human-centred governance, transparency, and ongoing capacity-building, ultimately providing strategic pathways for resilient, data-driven disaster preparedness.
Keywords: Artificial Intelligence, Disaster Risk Management, Early Warning Systems, Predictive Analytics, Decision Support
- Introduction
Disasters constitute complex socio-ecological phenomena arising from the interaction between natural hazards, exposure, vulnerability, and limited coping capacity, according to the United Nations Office for Disaster Risk Reduction (UNDRR, 2019), disaster risk is defined as the potential loss of life, injury, or destroyed and damaged assets that could occur to a system, society, or community within a specific period. Disaster Risk Management (DRM), therefore, refers to the systematic process of using administrative directives, organizational capacities, and operational skills to implement strategies, policies, and improved coping capacities to lessen the adverse impacts of hazards and the possibility of disaster (UNDRR, 2019). Contemporary DRM has progressively shifted from reactive emergency response toward anticipatory governance, risk prevention, and resilience-building, as articulated in the Sendai Framework for Disaster Risk Reduction 2015–2030. This paradigm emphasizes risk understanding, risk reduction, and preparedness rather than post-disaster recovery alone (UNDRR, 2019). However, the accelerating complexity of global risk landscapes driven by climate change, urbanization, environmental degradation, and interconnected infrastructures has exposed structural limitations in conventional risk modeling approaches (Cutter, Kevin, & Christopher, 2014).