AI And MI in Disaster Prediction and Management for Urban Resilience
Year: 2026
Type: conference-paper
Abstract: Despite the embedded trends of accelerated urbanization and uncertainties arising from environmental changes blamed on climate change, urban resilience is an emerging field of concern among policy makers, urban planners, and disaster management professionals. Article Title: Application of Machine Learning & Artificial Intelligence in Zero Disaster Cities & Disaster Resilience. This article takes a look into the ongoing progress to leverage the latest technologies of machine learning (ML) and artificial intelligence (AI) that are creating radical transformation in urban disaster preparedness, response and recovery, leading to a paradigm shift in urban resilience. AI and ML technologies can also facilitate the mining of huge amounts of constantly evolving and real time data produced by satellite imagery, sensor networks, weather data, social media feeds and historical disaster records. Such technology can uncover marled patterns and trends that human analysts often miss. ML can be trained to predict the probability, timeframe and possible location of natural disasters such as earthquakes, floods, wildfires and hurricanes with a high accuracy level using pertinent training data. Deep learning models can analyze satellite and drone imagery to monitor shifting land use, water levels, or deforestation — all signals of potential future environmental risks. AI and ML can analyze real-time data to assist in data about different disasters — air, seas, and ground — and enhance situational awareness and decision-making, perform risk assessment, and optimize resource allocation. With the help of simulation models driven by AI, for instance, they can generate a range of different, possible disaster scenarios which would enable cities to develop more reliable contingency plans. Natural language processing and chatbot applications, for instance, have been used to inform the public in real-time during emergencies, while machine learning models can be used behind the scenes to optimise evacuation routes, resource allocation and rescue operations. The use of predictive analytics enabled (or made possible) through AI and ML, among other things, helped shift from a reactive to proactive of disruption strategies, which is one of the key advantages of AI and ML for disaster resilience.
Institution: Fergana State Technical University
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7203683115