Studying the Safety Impact of Autonomous Vehicles Using Simulation-Based Surrogate Safety Measures

Morando, Mark Mario; Tian, Qingyun; Truong, Long T.; Vu, Hai L. · 2018 · OpenAlex-citations

DOI: 10.1155/2018/6135183

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 safety impacts of autonomous vehicles (AVs) using simulation-based surrogate safety measures, addressing the lack of empirical research on AV safety benefits in mixed traffic environments. While AVs are expected to reduce crashes by eliminating human error, their actual safety performance, particularly at varying penetration rates and in complex network settings like intersections and roundabouts, remains underexplored. The authors aim to quantify these impacts by modeling Level 4 automation behaviors and analyzing potential conflicts rather than relying on rare crash data. The methodology employs the VISSIM traffic microsimulation platform combined with the Surrogate Safety Assessment Model (SSAM). Human-driven vehicle (HV) behaviors were modeled using default Wiedemann 99 car-following parameters, while AV behaviors were simulated using two parameter sets (AV-1 and AV-2) that reflect more assertive driving, such as shorter headways, reduced standstill distances, and faster reaction times. Two case studies were analyzed: a signalized intersection in Melbourne, Australia, and a roundabout in Schenectady, New York. Scenarios varied AV penetration rates from 0% (base case) to 100% in 25% increments. Safety was assessed by counting potential conflicts identified through Time to Collision (TTC) and Post-Encroachment Time (PET) thresholds, with specific sensitivity analyses for AV-AV conflict thresholds. Results indicate that AVs significantly improve safety, particularly at high penetration rates. For the signalized intersection, AV penetration rates between 50% and 100% reduced the total number of conflicts by 20% to 65% compared to the base case, with statistical significance at $p < 0.05$. In the roundabout case study, a 100% AV penetration rate reduced conflicts by 29% to 64%, also statistically significant. The study found that mixed traffic (e.g., 50% AV penetration) often resulted in the highest number of HV-AV conflicts, suggesting that safety benefits are maximized when AVs dominate the traffic mix. Additionally, AVs traveling with shorter headways to increase capacity did not compromise safety; instead, they reduced both conflicts and travel delay. The roundabout scenario showed slightly higher safety benefits at 100% penetration compared to the signalized intersection. The findings suggest that high AV penetration rates are necessary to realize significant safety improvements, as mixed traffic environments can initially increase conflict frequencies between human and automated drivers. The study confirms that AVs can safely operate with shorter headways, enhancing road capacity without increasing risk. These results imply that infrastructure designs, such as roundabouts, may offer superior safety outcomes in fully automated environments. The research provides critical evidence for policymakers and urban planners regarding the conditions under which AVs deliver their promised safety benefits.

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.

• adas effectiveness

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