Towards linking driving complexity to crash risk

Zurlinden, Hendrik; Baruah, Anita; Gaffney, John · 2020 · Crossref

DOI: 10.33492/jrs-d-19-00070

archive: archived pipeline: cataloged verified

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

Summary

This paper investigates the relationship between complex traffic flow phenomena on urban motorways and increased crash risk, motivated by the rising number of fatal and serious injury crashes in high-density environments. While it is well-established that high vehicle density increases crash risk at a macro level, the microscopic mechanisms driving this risk remain unclear. The authors aim to bridge this gap by exploring how unstable or congested traffic flows trigger cognitive overload and human error, particularly during critical manoeuvres like lane changes. The study seeks to inform future road design and operational practices by identifying specific triggers of complexity and proposing methods to mitigate them. The research methodology combines a comprehensive literature review with empirical data analysis from the Metropolitan Melbourne motorway network. The authors utilized permanent camera installations and modern infrastructure-based detection technologies to measure individual vehicle behaviours, such as braking, speeding, and lane changes. These micro-level events were analyzed to understand their role as "events of exposure"—elementary units that generate opportunities for accidents. The study contrasts traditional exposure metrics, such as Vehicle Kilometres Travelled (VKT), with these event-based metrics to better predict crash occurrences under varying traffic conditions. Key findings indicate that unstable or congested flow creates low-speed, high-density clusters, such as nucleations and shockwaves, which propagate upstream against the direction of travel. These phenomena introduce "surprise elements" that sharply increase cognitive workload and reduce the freedom to perform necessary manoeuvres, thereby increasing the likelihood of human error. The authors demonstrate that traditional VKT-based crash rates can overestimate risk in lower-density scenarios because they fail to account for the specific critical manoeuvres associated with shockwaves and lane changes. Instead, linking crashes to countable events like harsh braking or narrow-gap lane changes provides a more robust and accurate understanding of crash causality. The significance of this work lies in its proposal for a new framework for understanding crash rates based on "events of exposure." By establishing robust relationships between specific driving events and crash outcomes, road operators can better identify high-risk locations and operational states. The paper concludes that appropriate planning and real-time traffic control, such as coordinated ramp metering, can mitigate the complexity characterized by high vehicle density. Furthermore, modern detection technologies offer the potential to analyze individual vehicle manoeuvres in real-time, enabling more precise interventions to reduce crashes and improve safety on saturated urban motorways.

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-nvidia on 2026-08-10; verification: verified.

Topics

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

• traffic density
• incidence prevalence
• naturalistic crash near crash
• pre crash contributing factors
• crash typology
• induced exposure

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: crash risk outcomes
• Methodological Resource: dataset resource
• Theoretical Contribution: theory or model