Pedestrian flows through a narrow doorway: Effect of individual behaviours on the global flow and microscopic dynamics

The authors study the dynamics of pedestrian flows through a narrow doorway by means of controlled experiments. The influence of the pedestrians’ behaviours is investigated by prescribing a selfish attitude to a fraction c<sub>s</sub> of the participants, while the others behave politely. Thanks to an original setup enabling the re-injection of egressed participants into the room, the analysis is conducted in a (macroscopically) quasi-stationary regime. The authors find that, as c<sub>s</sub> is increased, the flow rate J rises, interpolating between published values for egresses in normal conditions and measurements for competitive evacuations. The dependence of several flow properties on the pedestrian density ρ at the door, independently of c<sub>s</sub>, suggests that macroscopically the behavioural aspects could be subsumed under the density, at least in the authors' specific settings with limited crowd pressure. In particular, under these conditions, J grows monotonically with ρ up to “close-packing” (ρ ≈ 9 m -² ). The flow is then characterised microscopically. Among other quantities, the time lapses between successive escapes, the pedestrians’ waiting times in front of the door, and their angles of incidence are analysed statistically. In a nutshell, the authors' main results show that the flow is orderly for polite crowds, with narrowly distributed time lapses between egresses, while for larger cs the flow gets disorderly and vanishing time lapses emerge. For all cs, the authors find an alternation between short and long time lapses, which they ascribe to a generalised zipper effect. The average waiting time in the exit zone increases with its occupancy. The disorder in the flow and the pressure felt by participants are also assessed.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01635636
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Apr 25 2017 5:00PM