Citation

Oh H, Sisiopiku V. Transp. Res. Circular 2000; (E-C018): 459-470.

Copyright

(Copyright © 2000, U.S. National Academy of Sciences Transportation Research Board)

DOI

unavailable

PMID

unavailable

Abstract

The increase of the popularity of modern roundabouts in urban settings creates the need to assess their impact on pedestrian delays and capacities. This paper presents an analytical approach for calculation of pedestrian crossing capacity and pedestrian delay at roundabouts. Comparisons between pedestrian capacity and delays at roundabouts and signalized intersections are also provided. The pedestrian crossing capacity is conceptualized with the help of the maximum pedestrian crossing rate (MPCR). The MPCR is based on a probabilistic gap acceptance model. The model is made possible with the following practical assumptions: (a) priority cannot be given to the pedestrian at the crosswalk of roundabouts and (b) arrival rate of approaching vehicles follows the probabilistic distribution. The MPCR at signalized intersections is calculated based on the existing pedestrian capacity model of the "Highway Capacity Manual" considering signal phases and timing. As MPCRs for both intersection alternatives are proposed, the resulting delays can be calculated and compared. Queuing theory principles under M/M/1 (random arrivals/random departures/one server) and service on vehicles in the order of arrival (first in first out, FIFO) conditions are employed for this purpose. Capacity and delay are compared for a set of experimental conditions. The results indicate that roundabouts provide more capacity than signalized intersections for approaching traffic of up to 1000 veh/h/direction. The comparison of pedestrian delays confirms the intuitive concept that roundabouts can provide lower pedestrian delay under light vehicle and pedestrian traffic. In this paper, the MPCR model for calculation of pedestrian capacity and delay at roundabouts has been suggested and experimental results from model application are presented and discussed. These models, concepts, and results themselves may be useful to evaluate performance of roundabouts with respect to pedestrians. Employment of alternative arrival distributions is expected to enhance applicability, accuracy, and diversity of this model and is recommended for future studies.