Viral transmission in pedestrian crowds: Coupling an open-source code assessing the risks of airborne contagion with diverse pedestrian dynamics models

TL;DR

This paper presents a model-based approach to estimate airborne viral transmission risks in pedestrian crowds by coupling pedestrian simulations with an open-source transmission algorithm, analyzing the impact of different crowd dynamics.

Contribution

It introduces a method combining CFD-derived viral concentration maps with pedestrian models to assess infection risks and evaluates the robustness of predictions across various crowd behaviors.

Findings

Transmission risk predictions are stable across different pedestrian models under similar flow conditions.

Major crowd behavior changes, like jams, significantly increase transmission risk estimates.

The open-source code facilitates accessible and adaptable risk assessment in crowd scenarios.

Abstract

We study viral transmission in crowds via the short-ranged airborne pathway using a purely model-based approach. Our goal is two-pronged. Firstly, we illustrate with a concrete and pedagogical case study how to estimate the risks of new viral infections by coupling pedestrian simulations with the transmission algorithm that we recently released as open-source code. The algorithm hinges on pre-computed viral concentration maps derived from computational fluid dynamics (CFD) simulations. Secondly, we investigate to what extent the transmission risk predictions depend on the pedestrian dynamics model in use. For the simple bidirectional flow under consideration, the predictions are found to be surprisingly stable across initial conditions and models, despite the different microscopic arrangements of the simulated crowd, as long as the crowd evolves in a qualitatively similarly way. On the other hand, when major changes are observed in the crowd's behaviour, notably whenever a jam occurs at the centre of the channel, the estimated risks surge drastically.