An ontology-based approach to dynamic indoor fire emergency evacuation path planning with BIM integration

Abstract

The collection and updating of critical fire information remain key challenges in indoor fire emergency evacuation. This paper presents an innovative approach that integrates an Indoor Fire Emergency Evacuation Ontology (IFEEont) with Building Information Model (BIM). IFEEont is structured into five major categories, encapsulating key elements of fire evacuation. Unlike existing ontologies, it fully utilizes reasoning capabilities to infer and update node accessibility, a crucial factor in fire escape planning. A multi-sensor semantic model collects fire scene data, while ontology-based reasoning enables dynamic updates. Fire simulations conducted in an office building validate the effectiveness of the proposed method. Case study results demonstrate that the method not only captures critical fire information but also employs ontology-based inference to update node accessibility within the indoor geometric network model (GNM), effectively supporting escape route planning. Beyond serving as a fire knowledge base, the proposed ontology dynamically models fire information. This integrated approach—combining ontology, sensors, and BIM-based information updates—continuously optimizes evacuation strategies, enhances the accuracy and responsiveness of escape routes.