Article ID: | iaor19881097 |
Country: | United States |
Volume: | 22 |
Issue: | 2 |
Start Page Number: | 154 |
End Page Number: | 168 |
Publication Date: | Jun 1988 |
Journal: | Journal of Advanced Transportation |
Authors: | Rossi Thomas F., Ceder Avishai |
Keywords: | simulation, queues: theory, networks |
One of the most common measures of signalized intersection operation is the amount of delay a vehicle incurs while passing through the intersection. Traditional models for estimating vehicle delay at intersections generally assume fixed signal timing and uniform arrival rates for vehicles approaching the intersection. One would expect that highly variable arrival rates would result in much longer delays than uniform arrival rates of the same average magnitude. Furthermore, one might expect that signal timing that is adjusted according to traffic volume would result in lower delay signal when variations in flow warrant such adjustable timing. This paper attempts to test several hypotheses concerning the effects of variable traffic arrival rates and adjusted signal timing through the use of simulation. The simulation results corroborate the hypothesis concerning the effect of varying arrival rates. As the variance of the arrival rate over time increases, the average delay per vehicle also increases. Signal timing adjustments based on traffic appear to decrease delay when flow rates vary greatly. As flow variations stabilize, the benefits of signal adjustments tend to diminish.