Abstract

• We investigate evacuation simulation in flooded outdoor built environments. • We compare a generic and a custom simulation model based on a microscopic approach. • We set a generic simulator up to reproduce flood-related behaviors. • Simulators are applied to an idealized literature-based case study, including comparisons with real-world data. • Results seem to encourage the proposed generic simulator setup. Floods are among the most destructive sudden-onset disasters affecting worldwide communities and society. Pedestrians can be forced to evacuate affected areas thus being exposed to multiple risks. Outdoor built environment flood risks analyses should be performed through rapid, easy, and sustainable tools to speed up and support risk assessment and mitigations. Custom evacuation simulators have been developed, but are generally used in research, are not user-friendly, and need high-level training. On the contrary, generic (e.g. commercial) software tools seem to be more suitable for low-trained technicians but should be modified to include human behaviors effects, especially considering the evacuation, when people's peculiar choices depend on interactions with floodwaters and built environment layout/composing elements. This work provides preliminary setups of a generic software tool to perform quick and sustainable assessments of pedestrians’ flood safety in outdoor spaces, using an easy-to-apply no-code modification approach to include flood peculiar behaviors. Simulation outputs of the setup-based generic software are compared with a custom simulator relying on the same modelling approach, and with real-world observations, using an idealized literature-based outdoor scenario. Results provide the best setup of the generic software to reliably represent evacuation phenomena, thus encouraging its future application also by local authorities.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call