Citation

Kostreva MM, Lancaster LC. Fire Technol. 1998; 34(3): 227-246.

Affiliation

Clemson University, South Carolina

Copyright

(Copyright © 1998, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1023/A:1015345923210

PMID

unavailable

Abstract

Within the framework of HAZARD I, fire egress analysis is performed using the EXITT program. Yet another way to analyze fire egress employs newly developed multiple objective dynamic programming. This paper compares these two approaches by applying them to a model fire of moderate power in a residential building. The findings demonstrate that multiple objective dynamic programming can compute all the paths EXITT finds, but EXITT can't find all the paths multiple objective programming can. Some trade-offs inherent in choosing among the computed egress paths are discussed, and the features of the two fire egress methods are contrasted.