Crash Avoidance Technology Evaluation Using Real-World Crash Data

Flannagan, Carol A.; Leslie, Andrew · 2020 · ROSA P / United States. Department of Transportation. National Highway Traffic Safety Administration

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 evaluates the real-world field performance of various crash avoidance technologies using linked vehicle safety content and police-reported crash data. The research addresses the challenge of measuring the safety benefits of rapidly emerging automotive systems in a timely manner to inform regulatory decisions and New Car Assessment Programs (NCAPs). Previous studies relying on insurance claims often lacked specific crash circumstance data, leading to mixed conclusions. This project aimed to provide precise effectiveness estimates by linking specific safety equipment to relevant crash types while controlling for confounding variables. The researchers collaborated with General Motors (GM) to obtain VIN-linked safety content data for 1.2 million vehicles from model years 2013–2015. This data was linked to crash records from 13 U.S. states that provide 17-character VINs, resulting in an analysis dataset of 35,401 vehicles. The study employed a quasi-induced exposure method, comparing "system-relevant" crashes (e.g., rear-end striking for forward systems) to a control crash type (rear-end struck) that reflects driving exposure but is not influenced by the safety systems. Logistic regression models were used to estimate effectiveness, controlling for covariates such as driver age, gender, speed limit, alcohol/drug presence, fatigue, weather, road surface conditions, and vehicle model. Analyses were restricted to systems with sufficient statistical power to detect at least a 25% reduction in crashes. The results demonstrate significant effectiveness for systems involving automatic vehicle control compared to alert-only systems. For frontal collisions, Forward Collision Alert (FCA) reduced rear-end striking crashes by 16%, while Front Automatic Braking (FAB) reduced them by 45%. For lateral crashes, Lane Keep Assist (LKA) reduced lane departure crashes by 30%, whereas Lane Departure Warning (LDW) alone showed a non-significant 3% reduction. Lane Change Alert (packaged with Side Blind Zone Alert) reduced lane-change crashes by 32%, outperforming Side Blind Zone Alert alone (8%, non-significant). For backing crashes, Rear Automatic Braking (RAB) was 83% effective, Rear Cross-Traffic Alert (RCTA) was 56% effective, and Rear Park Assist (RPA) was 50% effective. Notably, the analysis revealed an interaction with driver age for backing systems; Rear Vision Camera (RVC) showed a non-significant 38% disbenefit for drivers aged 65 and older, suggesting potential differences in technology usage among older drivers. The study concludes that systems providing automatic vehicle control, whether brief (like LKA) or sustained (like FAB and RAB), yield substantially greater crash avoidance benefits than alert-only counterparts. This is attributed to the fact that control systems do not rely solely on timely driver response. The findings highlight the necessity of large, multi-manufacturer datasets linked to state crash databases to accurately assess new safety features. The authors recommend ongoing efforts to collect and combine such data to support NHTSA and global NCAP evaluations, ensuring that safety assessments keep pace with technological advancements in the automotive industry.

Key finding

Automatic braking systems produced substantially greater crash reductions (45 percent for frontal and 83 percent for backing crashes) compared to alert-only systems, which showed lower or non-significant effectiveness.

Methodology

dataset

Sample size: 35401

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_with_issues.

Topics

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

• adas effectiveness
• incidence prevalence
• naturalistic crash near crash
• induced exposure
• automated enforcement cameras
• crash typology

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).

• Applied Guidance: countermeasure evaluation
• Empirical Findings: crash risk outcomes
• Methodological Resource: dataset resource