The Dynamic Merge: Using Traffic Volume Based Signing to Improve Workzone Throughput

Weaver, Starla; Arnold, Michelle; Gonzalez, Tracy; Balk, Stacy · 2019 · Crossref

DOI: 10.17077/drivingassessment.1718

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 investigates the efficacy of the "dynamic merge," a traffic control strategy designed to improve safety and efficiency at roadway work zones involving lane closures. Lane reductions often cause vehicle queuing, which increases the risk of queue-end collisions and aggressive driving behaviors. Traditional "early merge" signing encourages drivers to merge before the queue, enhancing safety but reducing roadway capacity and throughput, particularly during heavy traffic. Conversely, "late merge" signing encourages drivers to use both lanes until the closure point, maximizing throughput but potentially compromising safety during light traffic when vehicles merge at high speeds. The dynamic merge adapts signing based on real-time traffic volume: employing early merge messaging for light traffic and late merge messaging for heavy traffic. However, prior research could not definitively determine whether behavioral changes were caused by the signage or simply by the traffic conditions themselves. This study aimed to isolate the effect of merge messaging from traffic volume to validate the specific benefits of dynamic merge systems. The researchers conducted a driving simulator experiment with 120 licensed drivers from the Washington, DC metropolitan area. The study utilized a 2x2 between-subjects design manipulating two independent variables: merge environment (early vs. late) and traffic volume (light vs. heavy). The simulator featured a motion base and a 200-degree screen. Participants drove a simulated four-lane highway with a 65 mph speed limit, encountering an advance warning area with static signs and three portable message signs displaying either early or late merge instructions. Traffic volume was controlled via microscopic traffic flow software, with light traffic defined as 5.82 vehicles per minute and heavy traffic as 8.92 vehicles per minute. Dependent variables included merge location, vehicle speed, time to clear the advance warning area, and self-reported stress levels. After excluding participants who merged before viewing the warning area, data from 78 participants were analyzed using generalized estimating equation models. The results demonstrated that merge environment significantly influenced driver behavior independent of traffic volume. Drivers in the early merge condition merged significantly earlier (mean 8.31 miles) than those in the late merge condition (mean 8.83 miles), confirming that the signage directly dictated merge location rather than traffic density alone. While traffic volume affected driving speed and overall travel time, a significant interaction was found regarding throughput. Under heavy traffic conditions, participants in the late merge condition cleared the advance warning area significantly faster than those in the early merge condition. Under light traffic, there was no significant difference in clearance time between the two merge environments. Stress levels remained low across all conditions and were not influenced by either variable. The findings confirm that dynamic merge messaging effectively influences driver merging behavior and supports the use of traffic-volume-based signing to optimize work zone operations. By encouraging late merging during heavy traffic, the strategy maximizes road capacity and improves throughput without the negative effects associated with early merging in congested conditions. The study validates that the benefits of dynamic merge systems are attributable to the messaging itself, providing evidence for transportation agencies to implement these adaptive strategies to enhance both safety and efficiency during lane closures.

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.

• work zones
• traffic density