Citation

Krogscheepers J, Roebuck C. Transp. Res. Circular 2000; (E-C018): 446-458.

Copyright

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

DOI

unavailable

PMID

unavailable

Abstract

The analysis of traffic flow at roundabouts has often been simplified by treating the roundabout as a series of isolated T-intersections with no interaction between the entering traffic streams from the various approaches. More recently, the effects of significant differences in approach volumes have been taken into account--as may be found in the latest version of the SIDRA software package. Such differences in approach volumes--or imbalances as they are termed in this paper--can have a significant effect on delay and capacity estimates for any given total volume of traffic entering a roundabout. The SIDRA package endeavors to take into account the effect of one approach volume dominating the other approaches, but it does not appear to be directly responsive to the actual location of the dominant approach relative to any other approach being analyzed. However, it was found in this research--which used the simulation program TRACSIM--that the delay on any given approach appears to be sensitive to a change in the balance of the circulating flows. For example, if most of the flow circulating past an approach originates from the first approach upstream then the average delay tends to be high. However, if the same volume of circulating traffic had originated from the second approach upstream then the average delay would tend to be low. Consequently, the "origin-destination" as well as the magnitude of the various approach volumes can be expected to affect the performance of a roundabout. The lack of appropriate field data limits the result of this study to a comparison between simulated and analytical models.