Key points are not available for this paper at this time.
In post-disaster scenarios, government and non-governmental organizations (NGOs) work to coordinate disaster relief efforts efficiently, aiming to meet humanitarian needs promptly. Utilizing available data such as satellite imagery, sensor data, and social media, alongside data mining and big data analytics, can significantly enhance disaster management efforts. However, timely access to these often fragmented and incomplete data presents challenges. Artificial Intelligence (AI) techniques, including machine learning, deep learning, and natural language processing, offer solutions by enabling faster response times for government agencies, NGOs, and other authorities. This research examines 72 studies on the application of AI in various stages of disaster management, including the simulation, detection, prediction, and handling of post-disaster situations. AI-based systems improve the issuance of early warnings, streamline risk communication, optimize relief logistics, and support evacuation plans while also aiding in the decision-making processes for the issuance of building permits and grants. The analysis highlights the transformative potential of AI across all disaster management phases, from preparedness and response to prevention/mitigation and recovery, and identifies future challenges in this domain. In conclusion, the study underscores the power of AI not only in forecasting the occurrence and impact zones of disasters but also in identifying the most vulnerable communities and assessing the viability of disaster response strategies.
Gupta et al. (Fri,) studied this question.