Hybrid Spatial Modelling, from Modelling People in the Vicinity to Urban Scale Simulation of Pedestrians

Abstract

Buildings, structures, public venues and urban areas can typically accommodate large numbers of people and behaviours. Consequently, it is a challenge for authorities to demonstrate that these heavily populated structures can be evacuated quickly and efficiently in emergencies such as fires, flash floods amongst others. Computational evacuation models are increasingly being used as decision support tools for evaluating evacuation efficiency. These models typically use either macroscopic or microscopic approaches for simulating pedestrian navigation. However, these approaches represent a trade-off between accuracy and scalability. In this work, we present a hybrid evacuation model, combining both macroscopic and microscopic modelling techniques, which can be scaled to larger structures, while maintaining accuracy. This is demonstrated in this work by applying the model to a building, rail tunnel station and large urban area.