NIST Time|NIST Home|About NIST|Contact NIST

HomeAll Years:AuthorKeywordTitle2005-2010:AuthorKeywordTitle

Federal Investigation of the Evacuation of the World Trade Center on September 11, 2001.

pdf icon Federal Investigation of the Evacuation of the World Trade Center on September 11, 2001. (4027 K)
Averill, J. D.; Mileti, D.; Peacock, R. D.; Kuligowski, E. D.; Groner, N.; Proulx, G.; Reneke, P. A.; Nelson, H. E.

Pedestrian and Evacuation Dynamics 2005. 3rd International Conference. Proceedings. September 28-30, 2005, Springer-Verlag, New York, NY, Waldau, N.; Gattermann, P.; Knoflacher, H.; Schreckenberg, M., Editor(s), 1-12 p., 2007.


evacuation; people movement; World Trade Center; investigations; occupants; occupant behavior; human behavior; egress; stairwell; elevators (lifts); mehtodology; statistics; high rise buildings; simulation; damage; human response; human performance


This paper presents the findings of the NIST World Trade Center Investigation describing the occupant evacuation of WTC 1 and WTC 2 on September 11, 2001. The egress system, including stairwells and elevators, is described along with the evacuation procedures. The population in WTC 1 and WTC 2 on September 11, 2001 at 8:46 a.m. is enumerated and described, where the background of the population was relevant to the subsequent evacuation, including training, experience, mobility status, among others. The progress of the evacuation of both towers is described in a quasi-chronological manner. A decedent analysis explores where occupants were located when each tower was attacked. Multiple regression models were built to explore the sources of evacuation initiation delay (why people did not immediately start to leave the building), as well as stairwell evacuation time (how long the average occupant spent in the stairwells per floor). Issues identified as contributing to either slowing or aiding the evacuation process were explored. Egress simulations provided context for estimating how long WTC 1 and WTC 2 would have taken to evacuate with different populations, using three different models, and subject to different assumptions of damage to the building.