Integrating Artificial Intelligence, IoT, and GIS for Smart Disaster Risk Management: A Systematic Review

Abstract

Disasters triggered by hydro-meteorological, geophysical, and climatic hazards are increasing in frequency and severity, straining the capacity of conventional disaster risk management (DRM) systems that rely on manual data collection and delayed decision-making [1], [5]. Recent advances in Artificial Intelligence (AI), the Internet of Things (IoT), and Geographic Information Systems (GIS) have opened new possibilities for building smart, data-driven, and anticipatory disaster management systems [3], [14]. This paper presents a systematic review of the literature published mainly between 2020 and 2026 that examines how AI, IoT, and GIS are being integrated to support hazard prediction, real-time monitoring, spatial risk assessment, early warning, and post-disaster response. Fifty-seven peer-reviewed and indexed sources were synthesised following a structured review protocol. The review identifies five major integration themes: (i) sensor-driven early warning networks, (ii) geospatial machine-learning hazard susceptibility mapping, (iii) UAV and remote-sensing based damage assessment, (iv) social-media and big-data situational awareness, and (v) digital-twin-enabled smart-city resilience platforms. The paper further develops a conceptual AI-IoT-GIS integration architecture, presents mathematical formulations commonly used for spatial risk indexing, sensor network reliability, and machine-learning performance evaluation, and illustrates these formulations with worked numerical examples, tables, and charts. The review concludes that while integration of these three technologies significantly improves prediction accuracy and response times, challenges remain around data interoperability, energy-constrained sensor networks, algorithmic bias, and the digital divide affecting adoption in low-resource regions [2], [15].