A Review of Building Evacuation Models - NIST Virtual Library

A.12 EXITT
Developer: B.M. Levin, NBS, U.S.
Purpose of the model: The purpose of this model is to simulate occupant decisions and actions
in fire emergencies in small residential buildings 63, 64 . The decision rules used by the model
were designed to resemble decisions made by occupants during a fire emergency. These
decision rules are based on:
• Judgment by the author
• Case studies of residential fires
• A limited number of controlled experiments
Availability to the public for use: This model is available for public use through the NFPA.
Modeling method: This is a behavioral model.
Structure of model: This is a coarse network system. The building is made up of nodes used to
represent rooms, exits, and secondary locations within a room, and the arcs are the distances
between the nodes.
Perspective of model and occupant: The model views occupants as individuals by assigning
each individual characteristics as well as tracking their movements throughout the simulation.
The occupant characteristics input into EXITT are age, sex, normal travel speed, whether or not
the occupant needs assistance during the evacuation, whether or not the person is asleep, room
location, and difficulty of waking up, if the person is sleeping.
The occupants also have an individual view of the building, due to their choice in exit path. The
occupants’ moves throughout the building are based on a shortest path algorithm included in
EXITT. During each action of the occupant, the route taken to the destination is via the shortest
path. This algorithm assigns penalties to certain paths due to heavy smoke or having to leave via
windows. In certain circumstances, the occupant is left to choose the exit with the lowest
number of penalties or demerits. Demerits work in the following way: each meter traveled is
assigned 1 demerit, leaving through a window is assigned 100 demerits, and traveling through
“bad” smoke is given 200 demerits. In some situations, all routes can become blocked, which
will leave occupants trapped in the residence.
Occupant behavior: Rule-based or conditional behavior. One way that occupants make
decisions is based on the optical density of the smoke in the upper layer using the equation for
psychological impact of smoke, S (equivalent to the equation used in EXIT89).
S = 2*OD*(D/H) where OD is the optical density of the upper layer, D is the depth of the upper
layer, and H is the height of the room. The following decision rules are incorporated into the
model:
• Occupants do not move to a node where S>0.5 (or into a room where S>0.4) unless the (H-D)
is at least 1.2 meters (the occupant can crawl)
• Occupants increase their travel speed by 30 % after they encounter smoke of S>0.1
A-45