Crowd simulation-based knowledge mining supporting building evacuation design. (August 2018)

• Record Type: Journal Article
• Authors: Boje, Calin; Li, Haijiang
• Is Part Of: Advanced engineering informatics. Volume 37(2018)
• Page Start: 103
• Page End: 118
• Publication Date: 2018-08
• Subjects: Knowledge mining -- Crowd simulation (CS) -- Ontology -- Evacuation design -- Building Information Modelling (BIM) -- Industry Foundation Classes (IFC); Computer-aided engineering -- Periodicals; Engineering -- Data processing -- Periodicals; 620.00285
• DOI: 10.1016/j.aei.2018.05.002
• Languages: English
• ISSNs: 1474-0346

Abstract

Abstract: Assessing building evacuation performance designs in emergency situations requires complex scenarios which need to be prepared and analysed using crowd simulation tools, requiring significant manual input. With current procedures, every design iteration requires several simulation scenarios, leading to a complicated and time-consuming process. This study aims to investigate the level of integration between digital building models and crowd simulation, within the scope of design automation. A methodology is presented in which existing ontology tools facilitate knowledge representation and mining throughout the process. Several information models are used to integrate, automate and provide feedback to the design decision-making processes. The proposed concept thus reduces the effort required to create valid simulation scenarios by applying represented knowledge, and provides feedback based on results and design objectives. To apply and test the methodology a system was developed, which is introduced here. The context of building performance during evacuation scenarios is considered, but additional design perspectives can be included. The system development section expands on the essential theoretical concepts required and the case study section shows a practical implementation of the system.