Article ID: | iaor2005317 |
Country: | United States |
Volume: | 38 |
Issue: | 1 |
Start Page Number: | 86 |
End Page Number: | 96 |
Publication Date: | Feb 2004 |
Journal: | Transportation Science |
Authors: | Black John A., Guo Xiaoping, Dunne Michael C. |
Keywords: | Transportation (Pedestrian) |
The phenomenon of the bunching of urban traffic is considered together with a mathematical model for the associated problem of the delay to pedestrians crossing a traffic stream. The approach used is also applicable to estimating delay to a single vehicle at a priority intersection and to merging vehicular traffic. Many pedestrian delay models include the assumption that successive vehicular headways are independent and identically distributed random variables. Such models may not be appropriate where vehicles form bunches – as occurs, for example, near traffic signals. Modeling of pedestrian delay is considered in two parts: the delay to pedestrians arriving during a vehicular bunch, and delay to those arriving during the random flow between bunches. These two parts are considered in combination to obtain overall delay. Three field surveys were conducted in metropolitan Sydney. Useable data were obtained on 4,350 headways and 252 crossing pedestrians on major arterial roads. The field data were used to assess the validity and accuracy of the proposed model. They support the theoretical model and suggest that, under certain conditions, it provides a more accurate estimation of pedestrian delays downstream of traffic signals than do some existing models, such as Adams' model.