Citation

Wu N, Brilon W. Transp. Res. Rec. 2018; 2672(15): 9-22.

Copyright

(Copyright © 2018, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.1177/0361198118788171

PMID

unavailable

Abstract

The conventional method of analysis for capacity of roundabouts is based on gap-acceptance methods or on empirical regression or a combination of both. The influence of pedestrians is modeled by reduction factors with a questionable empirical background. Thus, the current methods are based on an incoherent mix of approaches. Moreover, they do not account for the interaction between the different elements. The paper presents a new model which treats the whole intersection as one entity. Here all the conflict points where different streams (vehicles and non-motorized road users) intersect within the roundabout are identified. Each conflict point is treated as one queuing system with a simplified queuing mechanism. In addition, the interactions between the consecutive queuing systems are taken into account according to the theory of chains of queues where the distance (= storage area) between the conflict points becomes important. The paper explains the development of the model, presents the mathematical derivations of ready-to-use capacity equations plus parameter calibration by existing data, and demonstrates its real-world application. The advantage of the technique is that all conflicts--both between vehicle streams and with pedestrians at entries and exits--are treated by the same congruent methods. In addition, the interactions between consecutive arms of the roundabout are modeled. The method can also model limited priority (e.g., for pedestrians at crosswalks) for all conflict points. Finally, the conflicts that are decisive for the performance of the whole intersection are identified. The method can be easily implemented into computer software.

Language: en