Computer simulations vs. building guidance to enhance evacuation performance of buildings during emergency events Aysu Sagun a,⇑ , Dino Bouchlaghem b , Chimay J. Anumba c a Anglia Ruskin University, Cambridge School of Art, CB1 1PT, UK b Loughborough University, Department of Civil and Building Engineering, LE11 3TU, UK c The Pennsylvania State University, Department of Architectural Engineering, PA 16802, USA article info Article history: Received 26 March 2009 Received in revised form 15 March 2010 Accepted 1 December 2010 Available online 23 December 2010 Keywords: Building simulation Building evacuation Crowd modelling Building guidance Emergency events abstract Computer technologies can play an important role in the establishment of dynamic building information by introducing predictive modelling where behaviours of structures or groups of people can be simulated and observed. This way they can facilitate the design of the built environment to cope with emergency events. Modelling and simulation applications can be particularly useful at pre-planning, predicting possible damage, training responders, raising public awareness, and performance evaluation for reconstruction. They can be used for the development of virtual scenarios that include aspects of rescue operations, social behaviour of building occupants, and basic design requirements to test the current building codes and regulations. Within this context, the contribution of crowd simulation to improving the design of the built environment and guidelines is highlighted in this paper. Current building guidance for emergencies are summarised and the methodology developed to use crowd modelling to define design information associated with exit preferences of people during evacuations is explained. The results of the case studies underlined that there is a difference between the assumptions used for static information in current building guidance. Ó 2010 Elsevier B.V. All rights reserved. 1. Introduction The complexity of city life and the increase in the number of crowded public spaces such as entertainment, sports or transportation venues underline the need to identify risks and predict possible hazards to enhance the design of the built environment for increased safety and security. Several projects have explored the management of extreme events, especially during the response stage using emerging information, communication and computer technologies. One of these studies was a research project conducted at Loughborough University, entitled ‘‘ICT Enabled Solutions for Safety and Security Risks in the Built Environment’’. This study was focused on the deployment of Information and Communication Technologies (ICT) in disaster management and mitigation to improve safety and security within the built environment during the pre, during and post phases of disaster events. The research identified safety-aware building design as an important area in need of further investigation [1–3] and highlighted the need for the investigation of the potential application of computer simulations in crowd modelling to improve safety in building design. Most of the crowd simulation research deal with the development of simulation tools for the analysis of the evacuation process [4]. However, there is also a need to investigate the use of computer simulations to identify dynamic information required for building design guidance. Design guidance (codes, standards, guidelines, etc.) is necessary to identify and apply the building and environmental requirements however their sole application does not provide sufficient designs. Each building complex has specific charac- teristics that produce unique and dynamic information based on the relationship between people and building, functions, 1569-190X/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.simpat.2010.12.001 ⇑ Corresponding author. Tel.: +44 (0)845 196 2515. E-mail address: Aysu.Sagun-Kentel@anglia.ac.uk (A. Sagun). Simulation Modelling Practice and Theory 19 (2011) 1007–1019 Contents lists available at ScienceDirect Simulation Modelling Practice and Theory journal homepage: www.elsevier.com/locate/simpat