LADM-IndoorGML for exploring user movements in evacuation exercise

Abstract

The users' movements in the indoor environments differ based on the condition of the environments. During an indoor emergency, an efficient evacuation is required to help the users to move to the safe areas. Many types of incidents could impact the movements of users and this requires studying the behavior of the people during the evacuation. The reaction of the users to the incidents could affect the evacuation procedures and that could lead to several types of injuries or death. Each user understands and perceives the indoor environment differently and this plays a critical role in the evacuation. Furthermore, the users of the indoor environments have different rights to access the indoor spaces, which affects the movements of the users during an incident. This paper aims to support the evacuation of a building (educational building) in a crisis by using the integrated model of LADM-IndoorGML and the representation of the 3D model of the building. This research is presenting the initial assessment based on real world application. To reflect evacuation cases, we extended the conceptual model of LADM-IndoorGML to define the access rights for users of indoor environments during crisis. An evacuation exercise has been held at the Faculty of Applied Science at TU Delft to study the access rights during an incident. During the evacuation, Wi-Fi data has been collected for the users of the building for further analysis. A 3D model has been built for the Faculty of Applied Science to analyze the movement of the users. The collected data of the Wi-Fi access points have been structured and imported into the freeware database PostgreSQL/PostGIS. Furthermore, the geometry of 3D model was used to visualize the users’ movements as individuals and groups of users according to their connection to Wi-Fi access. Appropriate visualization has been created using QGIS. This paper demonstrates the entire process of analysis and visualization of users’ movements based on the Wi-Fi logs by using the extended LADM-IndoorGML. The outcome of the research has showed that the results for individual users and group users attached to the same access point differs. The study has also exhibited the importance of the time resolution on Monitoring the movements of a single user or group of users. The completed study clearly demonstrates that with the proposed extension, the integrated LADM-IndoorGML model is able to support the decision-making process during an incident in educational building.