Traffic Behavior at Freeway Bottlenecks

Coifman, Benjamin; Kim, Seoungbum · 2014 · ROSA P / NEXTRANS Center (U.S.)

archive: archived pipeline: cataloged verified

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

Summary

This study investigates traffic behavior at freeway on-ramp bottlenecks, challenging conventional assumptions regarding bottleneck formation, capacity drop, and the fundamental relationship (FR). The research is motivated by the need to identify systematic biases in empirical studies that likely distort the understanding of bottleneck capacity. Specifically, the authors examine how driver relaxation—the process where drivers entering a freeway or following an entering vehicle temporarily tolerate shorter headways before returning to their preferred speed-spacing relationship—affects traffic flow measurements. The researchers employed a simulation-based approach using Newell’s lower-order car-following model, modified to include a driver relaxation factor for "affected vehicles" (those entering or immediately behind an entering vehicle). The simulation modeled a one-lane freeway with a free speed of 60 mph and a roadway capacity (RCap) of 2,200 vehicles per hour (vph). The study tested nine combinations of mainline and ramp demands, including scenarios where combined demand exceeded RCap. The model tracked vehicle trajectories to simulate conventional detector measurements, analyzing flow, density, and speed over time and space. The findings reveal that traffic flow becomes "supersaturated" in any sample containing an affected vehicle with a truncated headway, resulting in flows higher than the underlying FR predicts. During the initial "loading period" of bottleneck activation, which can last several minutes, flows exceed RCap while speeds remain near free speed. This creates a state where conventional point bottleneck models erroneously indicate the bottleneck is inactive because no queuing is visible upstream of the ramp. The actual bottleneck process occurs over an extended distance, with queues initially forming downstream and eventually receding upstream. Consequently, flow does not drop "from capacity" upon activation but rather drops "to capacity" from a state of supersaturation. This mechanism shifts the empirically observed flow-density relationship above the theoretical curve, potentially leading to overestimations of bottleneck capacity. The significance of this work lies in its implications for traffic flow theory and control strategies, such as traffic-responsive ramp metering. The study demonstrates that the bottleneck process extends over a distance exceeding one mile, contradicting the simple point bottleneck model. The authors conclude that failing to account for driver relaxation leads to inaccurate capacity estimates and delayed detection of bottleneck activation. They recommend that future empirical studies and data collection efforts account for these extended spatial dynamics and develop higher-resolution data sources to better understand the nuances of bottleneck mechanisms, which are critical for alleviating congestion and increasing freeway throughput.

Key finding

Supersaturated flows caused by driver relaxation exceed roadway capacity during the initial bottleneck activation period, leading to overestimation of bottleneck capacity and delayed detection of bottleneck activity.

Methodology

simulator

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. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

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

• traffic density