TITLE:
Roundabout Performance Analysis Using Connected Vehicle Data
AUTHORS:
Enrique Saldivar-Carranza, Jijo K. Mathew, Howell Li, Darcy M. Bullock
KEYWORDS:
Roundabout, Performance Measure, Connected Vehicle, Big Data
JOURNAL NAME:
Journal of Transportation Technologies,
Vol.12 No.1,
December
17,
2021
ABSTRACT: There are over 8000 roundabouts in the United States. The current
techniques for assessing their performance require field counts to provide
inputs to analysis or simulation models. These techniques are labor-intensive
and do not scale well. This paper presents a methodology to use connected
vehicle (CV) trajectory data to estimate delay and level of service for
roundabout approaches by adapting the Purdue Probe Diagram used for traffic
signal analytics. By linear referencing vehicle trajectories with a particular
movement based on the location and time they exit a roundabout, delay can be
calculated. The scalability is demonstrated by applying these techniques to
assess over 100 roundabouts in Carmel, IN during the weekday afternoon peak
period in July 2021. Over 264,000 trajectories and 3,600,000 GPS points were
analyzed to rank over 300 roundabout approaches by delay and summarize in Pareto-sorted
graphics and maps. The paper concludes by discussing how these techniques can also be used to analyze queue lengths and origin-destination characteristics at roundabouts. The
methodology presented in this study can be used by any agency that wants to
assess the performance of all roundabouts in their system.