Comparison of Procedures to Estimate Critical Headways at Roundabouts

Vasconcelos, António Luís Pimentel; Seco, Álvaro Jorge Maia; Silva, Ana Maria César Bastos · 1970 · OpenAlex-citations

DOI: 10.7307/ptt.v25i1.1246

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 addresses the limitations of the empirical Transportation Research Laboratory (TRL) method, which is the standard for roundabout capacity analysis in Portugal but lacks explanatory power for innovative layouts. To enable lane-by-lane analysis using gap-acceptance theory, specifically Hagring’s generalization of Tanner’s formula, the authors sought to locally calibrate critical headways and follow-up times. The research was motivated by the recognition that driver behavior varies by country and that existing parameters often rely on indirect fitting rather than direct observation of gap-acceptance decisions. The methodology involved collecting gap-acceptance data at six roundabouts in Coimbra and Viseu, Portugal, including single-lane and two-lane configurations. Video recordings were processed using a semi-automatic tool (LUT|VP2) to record vehicle arrival and entry times, as well as conflicting vehicle passages. Data were filtered to include only observations with at least one rejected lag. The authors applied five distinct estimation procedures to the collected data: Siegloch, Raff, Maximum Likelihood (ML), Wu, and Logit. The Siegloch method required saturated conditions and varied based on move-up time thresholds (4s or 6s), while the other methods utilized accepted and rejected gap distributions or binary choice models. The results revealed significant variations in estimated critical headways depending on the method used. Siegloch estimates were highly sensitive to the move-up time threshold, with the 6-second threshold yielding results more consistent with other methods but potentially including perturbed maneuvers. The Wu and Maximum Likelihood methods produced very similar estimates. At multilane roundabouts, critical headways were generally smaller for right-lane entries, though exceptions occurred where traffic patterns were uniform. Notably, the Maximum Likelihood method yielded considerably higher estimates at one site due to low opposing flows limiting data on short-gap acceptance. Comparisons with reference values from other countries indicated significant differences, suggesting distinct driving styles in Portugal. The study concludes that locally calibrated, country-specific parameters are essential for accurate capacity calculations, as universal values fail to account for regional driving behaviors. The findings highlight that the choice of estimation procedure significantly impacts capacity estimates, necessitating careful selection based on data availability and traffic conditions. The research supports the adoption of gap-acceptance models over empirical regression methods for analyzing non-conventional roundabout geometries, provided that parameters are derived from direct, local observations of driver gap-acceptance decisions.

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.

• gap acceptance