The immense edifice of Fluid Mechanics is partly built on the virtually impalpable contribution of dimensionless numbers, which are instrumental to transpose the full-scale reality into small-scale experiments conducted e.g. in wind tunnels, but also to determine which approximate model is most suitable to describe a phenomenon. It had long been suggested that fluid flows and crowd flows bear similarities, but only very recently have researchers come up with suitable dimensionless numbers to characterise the latter.
On this basis, using a gamut of empirical and experimental datasets, they have classified the heteroclite variety of crowd fluids into more homogeneous groups of situations, governed by similar processes and within which the crowd arranges according to similar `rules’.
This has opened the door to a perturbative handling of pedestrian dynamics, inspired by the perturbative approaches applied near phase transitions in condensed-matter physics. These approaches are instructive regarding the global behaviour of a system, even when one only has blurred information at the microscopic scale, which is quite valuable for entities as complex as pedestrians. In short, it is easier to arrange crowd flows in separate compartments than it is to fit the pedestrian behaviour in a box.
These findings have just been published in the journal PNAS Nexus.
