Citation

Teknomo K, Fernandez P. Safety Sci. 2012; 50(5): 1228-1236.

Copyright

(Copyright © 2012, Elsevier Publishing)

DOI

10.1016/j.ssci.2011.12.025

PMID

unavailable

Abstract

A standard evacuation map posted in a room shows the location of the current room and the path to the nearest exit. If the number of occupants in the building is only small, then the shortest paths to the exits may enable all the occupants to evacuate in minimum possible time. Previous studies have shown, however, that the shortest path configuration does not lead to minimum egress time for large crowds in public facilities such as those in a school, theater or gym.

We extend the previous studies by determining the minimum egress time for different crowd sizes on a fixed network graph. We apply optimization search on a mesoscopic multi-agent pedestrian simulation that employs the concept of Route-Choice Self-Organization (RCSO). We show that the egress time gap between the shortest path configuration and RCSO configuration increases very quickly with respect to crowd size. Thus, for crowded pedestrian facilities, there may be a need to revise the standard evacuation map so that evacuation behavior that approximates the RCSO becomes possible, and this can lead to a much better egress time.