Crashes and Injuries on Rural Roads in Alaska – Toward a Better Understanding of Rural Safety Issues Through Linked Data and Environmental Factors Task B: Environmental and Geometric Safety Factors of Rural Crashes

Vasudevan, Vinod · 2025 · ROSA P / University of Alaska Fairbanks. Center for Safety Equity in Transportation (CSET)

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 investigates the environmental and geometric factors contributing to crashes and injuries on rural roads in Alaska, addressing the unique safety challenges posed by the state’s extreme weather, seasonal daylight variations, and scattered rural population. Motivated by the limitations of traditional crash data analysis in capturing these complex risks, the research aims to integrate linked data, including solar position metrics, to better delineate safety issues specific to Alaskan conditions. The work was conducted by the Center for Safety Equity in Transportation at the University of Alaska Fairbanks, utilizing crash data from the Alaska Department of Transportation and Public Facilities for the period 2009–2012. The methodology involved dividing crashes into urban and rural categories based on municipality status and analyzing them using Python programming. The study compared crash distributions by hour, season, traffic control type, and crash type. A key component of the analysis was correlating crash occurrences with sun elevation angles, calculated using latitude and longitude data. Additionally, the researchers employed machine learning techniques, specifically CART decision trees, to identify risk factors for injury severity and specific crash types, such as collisions with animals or non-collisions. The findings reveal that while urban and rural areas share similar hourly crash patterns, rural areas experience a slightly higher proportion of crashes during winter months, with over 60% of all crashes occurring in winter despite lower traffic volumes compared to summer. Rural crashes are predominantly single-vehicle events, such as run-off-roadway or moose-related incidents, whereas urban crashes are more frequently multi-vehicle angle or rear-end collisions. A strong correlation was identified between crash frequency and sun elevation; crashes peaked when the sun was at a critical low elevation of 0 to 15 degrees, particularly during periods of shorter daylight. At these critical elevations, a higher proportion of crashes occurred at uncontrolled locations, likely due to glare impairing visibility. Machine learning models identified Road Surface, Time of Day, and First Sequence Event as the top predictors for injury. For non-collision and animal-involved crashes, Lighting, Driver Age, and Road Surface were the best predictors, with non-collisions most likely occurring in dark conditions with wet, icy, or snowy roads. The significance of this research lies in its demonstration that environmental factors, particularly sun glare and road surface conditions, are critical determinants of crash risk in Alaska. The study highlights that rural and urban areas face distinct safety challenges, with rural areas more susceptible to single-vehicle crashes influenced by wildlife and slippery surfaces, while urban areas are more affected by sun glare at uncontrolled intersections. These insights suggest that targeted safety interventions, such as improved lighting and reflective signage, are necessary to address the specific geometric and environmental risks inherent to Alaskan roadways.

Key finding

Crash frequency in Alaska is highly correlated with low sun elevation angles (0–15 degrees), and machine learning models identified road surface, time of day, and first sequence event as the top predictors for injury severity.

Methodology

dataset

Sample size: 81836

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.

Topics

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

• urban rural setting
• incidence prevalence
• demographic disparities
• roadway lighting effects
• pre crash contributing factors
• vru 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).

• Empirical Findings: crash risk outcomes