Evaluation of Roundabout Capacity Models: An Empirical Case Study

Ren, Liang; Qu, Xiaobo; Guan, Hong; Easa, Said M.; Oh, Erwin · 2016 · OpenAlex-citations

DOI: 10.1061/(asce)te.1943-5436.0000878

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study evaluates the accuracy of various roundabout entry capacity models by comparing analytical and empirical methods against field data. The research addresses the limitation of existing models, such as the Highway Capacity Manual (HCM) 2000, German Highway Capacity Manual (GHCM), and SIDRA, which often underestimate capacity by ignoring the impact of exiting vehicles. The authors propose and validate a New Roundabout Capacity (NRC) model that incorporates the proportion of exiting vehicles with indicators on, arguing that these vehicles create additional entry opportunities for waiting traffic. Data were collected from nine single-lane roundabouts in Gold Coast, Australia, using video recordings during peak hours. The researchers calibrated critical gaps (ranging from 4.32 to 4.82 seconds) and follow-up times (2.35 to 2.75 seconds) for each site using graphical and maximum likelihood methods. To establish a baseline for actual capacity, the study employed the concept of At-Capacity Conflicting Headway (ACCH), measuring the number of vehicles entering during specific headways in the circulating flow. This empirical data was then compared against estimates from the NRC model, HCM 2000, HCM 2010, GHCM, SIDRA, a linear regression model, and a VISSIM simulation. The results demonstrate that the NRC model significantly outperforms the other models in accuracy. While all models tended to underestimate capacity, the NRC model exhibited the smallest range of relative error (-1.07% to -5.74%) and the lowest root-mean-square deviation (47.68). In contrast, the GHCM model showed the largest errors (-8.95% to -21.26% relative error; 146.82 RMSD), and the VISSIM simulation performed poorly due to inaccurate gap acceptance simulation. The study found that while all models produce similar estimates under low and medium traffic volumes, discrepancies widen at high volumes. Specifically, the NRC model accounts for the fact that at high traffic conditions, a high proportion of exiting vehicles warrants more entry opportunities, effectively offsetting the negative impact of high circulating volumes. The significance of this work lies in validating the importance of including exiting vehicle behavior in capacity estimation. The findings suggest that traditional models fail to capture the self-regulating nature of roundabouts where drivers utilize gaps created by exiting traffic. By incorporating the proportion of exiting vehicles, the NRC model provides a more precise tool for transport agencies to evaluate roundabout performance, particularly in high-volume scenarios. This contributes to more accurate infrastructure planning and highlights the need for capacity models that reflect real-world driver gap acceptance behaviors beyond simple conflicting flow metrics.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• traffic density