Naturalistic Driving Database Development and Analysis of Crash and near-Crash Traffic Events in Honolulu
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 addresses the limitations of traditional post-crash traffic safety analysis, which often relies on small sample sizes and retrospective data that fails to capture pre-crash conditions. To overcome these challenges, the researchers utilized Naturalistic Driving Data (NDD) to analyze crash and near-crash events in Honolulu, Hawaii. The primary motivation was to identify specific factors contributing to traffic conflicts by leveraging the higher frequency of near-crash events as a surrogate for actual crashes, thereby enabling more robust statistical analysis of driver behavior, environmental conditions, and road characteristics. The methodology involved a partnership between the University of Hawaii at Manoa and Charley’s Taxi and Limousine (CTL). Dashboard cameras and sensors from the Samsara system were installed in 233 taxi vans on Oahu. Data collection occurred over seven months, from fall 2019 to spring 2020, though it was halted early due to the COVID-19 pandemic. The study focused on events triggered by driving maneuvers causing longitudinal acceleration of 0.5g or higher. The resulting database comprised 402 harsh events: 398 near-crashes and four crashes. Researchers coded variables related to road geometry, environmental conditions, driver demographics, and vehicle features. Analysis techniques included deriving basic statistics, investigating correlations via Chi-square tests, and applying stepwise linear regression models to determine causal relationships for the most frequent event types. The findings revealed distinct patterns in event frequency and contributing factors. Nearly 18% of events occurred on Thursdays, while 82.8% happened on straight road segments. The most common event type was near rear-end collisions involving the taxi van and the vehicle ahead, followed by lane-changing events and incidents involving pedestrians. Regression analysis identified mobile phone use as a significant positive factor increasing the risk of both near rear-end events on freeways and lane-changing near-crashes. Conversely, light traffic conditions and uninterrupted flow facilities reduced the risk of near rear-end events. Vehicles lacking automated braking systems or lane-occupancy warnings exhibited a higher risk of lane-changing near-crashes. Additionally, the presence of pedestrian crossings significantly increased the likelihood of pedestrian-related events, while wider expressways correlated with a higher risk of near rear-end incidents. The significance of this research lies in its demonstration of how commercial fleet monitoring systems can provide valuable, real-world data for traffic safety analysis. The study confirms that near-crash data can effectively identify risk factors such as distraction and infrastructure design flaws that are difficult to isolate in traditional crash databases. The findings offer actionable insights for driver coaching programs and potential policy adjustments, such as addressing the risks associated with mobile phone use and the benefits of automated driver aids. The authors conclude that future research should expand to non-professional drivers to better represent the general driving population, as the current study was limited to a professional taxi fleet.
Key finding
Mobile phone use and the absence of automated driver aids significantly increase the risk of near-crash events, particularly for rear-end and lane-changing incidents on freeways.
Methodology
naturalistic
Sample size: 233
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).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 2026-06-11 |
| verify | partial | — | — | — | 2 | 2026-06-10 |
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.
- naturalistic crash near crash
- incidence prevalence
- pre crash contributing factors
- crash typology
- urban rural setting
- causation analyses
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, observational prevalence
- Methodological Resource: dataset resource