The prediction of incident durations can facilitate incident managemen
t and support traveler decisions. This paper develops a procedure for
predicting incident durations. First, the causal and non-causal factor
s which influence incident durations are conceptualized. These include
operational characteristics such as response times and whether a heav
y wrecker was used, incident characteristics such as injuries and numb
er of vehicles involved and environmental conditions such as weather a
nd visibility. Specific hypotheses are tested by developing truncated
regression models of incident duration using data provided by the Illi
nois Department of Transportation (IDOT) on Chicago area freeways. The
n, a time sequential methodology is developed to predict the incident
durations as information about the incident is acquired in a Traffic O
perations Center or TOC. Initially, after an incident is detected, inf
ormation at a TOC is often acquired at a high rate, then information a
cquisition levels off and toward the end of an incident the acquired i
nformation may decay. Accordingly, the incident duration models grow i
n terms of their explanatory variables at first, then they are sustain
ed during the middle stages and begin shrinking toward the end when in
formation starts decaying. Finally, the implications of this predictio
n methodology are discussed.