Modeling Variability of Urban Travel Times by Analyzing Delay Distribution for Multiple Signalized Intersections
Fangfang Zheng, Henk J. van Zuylen
Abstract
Fangfang Zheng, Henk J. van Zuylen
Abstract
It is widely accepted that travel times in urban road networks can vary greatly within a short period as well as from period to period in a single day. One important cause for this variability in urban travel times is the delay that drivers experience at intersections, particularly signalized intersections. This delay can be due to the time a driver needs to wait for a red light or the time needed for a queue of waiting vehicles at an intersection to dissolve, or both. As a result, the ability to capture the characteristics of delay at signalized intersections is a critical, yet challenging, step in estimating and predicting urban link travel times. The uncertainties in these experienced delays are caused by traffic conditions and traffic control at the intersections. In this paper a delay distribution model is developed for an urban trip over a route with two consecutive fixed-time controlled intersections. The proposed model captures the stochasticity in the arrival and departure process at the first intersection. An example illustrates how these stochastic factors influence the realized delay distribution for the two consecutive signalized intersections. Delay distributions are analyzed for different levels of mismatch of the two controlled intersections (from well coordinated to badly coordinated) under different degrees of saturation. Finally, the predicted delay distributions are validated with the VISSIM simulation model and show that the proposed (analytical) model can represent the VISSIM simulation data.
OpenAlex reports 27 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
It is widely accepted that travel times in urban road networks can vary greatly within a short period as well as from period to period in a single day. One important cause for this variability in urban travel times is the delay that drivers experience at intersections, particularly signalized intersections. This delay can be due to the time a driver needs to wait for a red light or the time needed for a queue of waiting vehicles at an intersection to dissolve, or both. As a result, the ability to capture the characteristics of delay at signalized intersections is a critical, yet challenging, step in estimating and predicting urban link travel times. The uncertainties in these experienced delays are caused by traffic conditions and traffic control at the intersections. In this paper a delay distribution model is developed for an urban trip over a route with two consecutive fixed-time controlled intersections. The proposed model captures the stochasticity in the arrival and departure process at the first intersection. An example illustrates how these stochastic factors influence the realized delay distribution for the two consecutive signalized intersections. Delay distributions are analyzed for different levels of mismatch of the two controlled intersections (from well coordinated to badly coordinated) under different degrees of saturation. Finally, the predicted delay distributions are validated with the VISSIM simulation model and show that the proposed (analytical) model can represent the VISSIM simulation data.
Key concepts: VisSim, Intersection (aeronautics), Queue, Transport engineering, Traffic simulation, Retard, Computer science, Travel time