An Analytical Examination of the Role of Artificial Intelligence and Geographic Information Systems in the Development of Intelligent Urban Disaster Management Systems
An Analytical Examination of the Role of Artificial Intelligence and Geographic Information Systems in the Development of Intelligent Urban Disaster Management Systems
Mohammad Jalali1 Komeil Jafari Tirtashi2 Azizolla Ghoolipour3
1) Department of Computer Engineering, Technical and Vocational University, Tehran, Iran Email:
2) Department of Civil Engineering, Technical and Vocational University, Tehran, Iran Email:
3) Department of Electrical Engineering, Technical and Vocational University, Tehran, Iran Email:
IOI Code
:
ABFG-ACZGAE
Publication :
12th.international Congress on Civil Engineering, Architecture, Art and Urban Development(12icsau.ir)
Abstract :
The expansion of urbanization, the increasing complexity of infrastructure, and the growing exposure of cities to natural and human-induced hazards have increased the need for developing intelligent disaster management systems. Geographic Information Systems (GIS) are considered an important infrastructure for spatial data management and disaster analysis; however, they face limitations when dealing with the volume, velocity, heterogeneity, and dynamic nature of emerging urban data. In contrast, advances in artificial intelligence, machine learning, deep learning, GeoAI, remote sensing, digital twins, and large language models have enabled the development of predictive, adaptive, and decision-support systems. This review-analytical study examines studies published primarily between 2024 and 2026 and classifies them based on technology, data type, artificial intelligence method, disaster application, and decision-making level. The findings indicate that the integration of AI and GIS has significant applications in flood prediction, vulnerability assessment, damage detection, early warning, risk analysis, and decision support. Furthermore, digital twins enable the integration of spatial data, sensors, analytical models, and urban infrastructure, while large language models (LLMs) facilitate more natural interaction with data and decision-support systems. Nevertheless, challenges such as explainability, uncertainty, transferability across cities, online validation, privacy, and security remain. The most significant identified gap is the disconnect between analytical models and operational decision-support systems. Finally, the proposed architecture integrates AI, GIS, IoT, remote sensing, social data, digital twins, and a decision-support layer to provide a closed-loop cycle encompassing perception, prediction, simulation, decision-making, and feedback.
Keywords : artificial intelligence gis geoai urban disaster management large language models digital twin