Development of an Automated Prototype Platform for Evacuation of Human Behavior Data Collection and Storage Using Unreal Engine

Urbanization has increased the scale and structural complexity of buildings, thereby reducing evacuees’ chances of survival during fire emergencies. Therefore, analyzing evacuee behaviors is essential for establishing efficient evacuation strategies. However, previous studies were limited to collecting simplified behavioral and decision-making data, owing to restrictions in experimental conditions. To overcome these limitations, we developed a prototype platform for collecting and storing human behavior data based on evacuation-influencing factors. The platform incorporates a realistic virtual environment, a systematic classification of human behaviors, and automated functions for collecting and storing partial behavioral factor data, and its effectiveness was validated through a small-scale virtual environment experiment. Although the platform currently captures only a limited set of evacuation-related behavioral factors, it represents a meaningful step toward an integrated system for automated human behavior data collection and storage in evacuation analysis. Future work will expand and integrate evacuation-influencing factors to improve the completeness and applicability of the platform.