Title: "The effectiveness of artificial intelligence in creating algorithms for human action in emergency situations"
Authors: Akhmadbek Jalilov
Volume: 9
Issue: 9
Pages: 134-139
Publication Date: 2025/09/28
Abstract:
This article is devoted to the analysis of the effectiveness of artificial intelligence (AI) technologies in the formation and optimization of human action algorithms in emergency situations. In emergency situations such as earthquakes, fires, floods and mass casualty situations, human decision-making is limited by stress, uncertainty and resource shortage. Therefore, the use of AI algorithms is becoming increasingly important in evacuation, triage and rescue operations. The article extensively covers the capabilities of machine learning, deep learning, reinforcement learning, multi-agent simulation (MAS) and computer vision technologies.The research methodology includes simulating evacuation and rescue scenarios, comparing SI algorithms with traditional human decision-making during triage, and assessing resource efficiency. The results show that SI-based models can reduce evacuation time by 30-40%, increase decision-making accuracy, and reduce resource waste. The discussion section provides a comprehensive analysis of the advantages and limitations of human and AI decision-making, ethical and legal issues, and the challenges of adapting AI systems to real-world conditions. Finally, recommendations are made on the prospects for the widespread implementation of AI technologies in emergency management and future research directions.