01152276

Component

In the last two decades, there have been relatively large number of studies attempting to measure and quantify incident frequency, duration, and delay for different traffic and incident management conditions. This paper focuses on some open research questions on incident duration estimation and related traffic delay analysis by using quantitative comparisons of nationwide study results and three New Jersey incident datasets. An immediate conclusion of the study is that there is no universally accepted model to estimate incident durations. Incident durations (both response and clearance times) are observed to vary considerably across years, state and facility. Thus, it is recommended that each state collect incident/accident duration and delay data in addition to frequency data that is being currently collected in all of the States. This will ensure better understanding of the impacts of non-recurrent congestion by providing decision makers with more robust information to solve ever increasing congestion problem. Finally, using the simple deterministic queuing model combined with a probabilistic representation of clearance and response times, it was shown that clearance times, with larger mean and variance, contribute more to the total delay compared with response times. In that respect, incident management strategies aiming to reduce clearance times may have larger impact on reducing the total delay. This is another way of demonstrating the usefulness of having detailed incident/accident data for making more informed future decisions.

01147878

10-3527

English

20p

2010

Transportation Research Board 89th Annual Meeting

DVD

Figures (1) ; References (34) ; Tables (10)

Crash data; Emergency response time; Highway operations; Incident management; Time duration; Traffic delays; Traffic incidents

New Jersey

Highways; Operations and Traffic Management; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2010 Paper #10-3527

TRIS, TRB

Jan 25 2010 11:47AM