Development of Safety Screening Tool for High Risk Rural Roads in South Dakota
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 challenge of identifying high-risk locations on rural highways in South Dakota, where fatal and severe crashes are sparsely distributed and traditional "hot spot" analysis is ineffective. With an average of 120 fatal crashes annually across 83,744 miles of highway, the low crash density (0.0015 crashes per mile) makes reactive, location-specific treatments inefficient and cost-prohibitive. The research aims to develop a proactive, system-wide safety screening tool that identifies boundaries for safety improvements without relying on predefined roadway segments, thereby capturing risks that span multiple segments or are obscured by statistical noise. To achieve this, the authors developed an Empirical Bayes (EB) based sliding window technique within a Geographic Information Systems (GIS) framework. The methodology utilizes crash prediction models, specifically Safety Performance Functions (SPFs), calibrated for South Dakota’s rural environment. The EB method was selected to account for regression-to-the-mean, a statistical phenomenon where sites with unusually high crash counts in one year often revert to average levels in subsequent years, leading to inaccurate risk assessments. By combining predicted crash frequencies from SPFs with historical crash data, the EB method provides a more robust estimate of long-term safety performance. The sliding window approach allows for spatial analysis that is not constrained by fixed segment boundaries, enabling the identification of high-risk areas that may extend across multiple predefined segments. The study implemented these methodologies through the development of the South Dakota GIS Highway Safety Review (GIS-HSR) Tools. This tool integrates data from the South Dakota Department of Transportation and the Department of Public Safety, processing variables such as Annual Average Daily Traffic (AADT), segment length, and crash history. The tool generates safety metrics, including EB crash rates and excessive crash indices, for both fixed segments and sliding windows. The application of the tool demonstrated its ability to identify high-risk locations that traditional methods might miss, such as areas where a short, dangerous segment influences the safety profile of a longer stretch. The results showed that the EB-based sliding window method provided a more accurate and comprehensive view of rural highway safety performance compared to observational crash analysis alone. The significance of this work lies in its provision of a practical, data-driven tool for transportation practitioners to prioritize safety investments in rural areas. By shifting from a reactive approach based on historical crash clusters to a proactive approach based on predictive modeling and spatial analysis, the GIS-HSR Tools enable more effective deployment of system-wide safety measures, such as rumble strips or median barriers. The tool’s design, requiring minimal user input and leveraging existing GIS infrastructure, facilitates widespread adoption. While tuned for South Dakota’s specific conditions, the general architecture is applicable to other regions facing similar challenges with sparse crash data, offering a scalable solution for improving rural highway safety nationwide.
Key finding
The empirical Bayes sliding window technique successfully identified high-risk rural highway locations by removing dependence on predefined segments and accounting for regression-to-the-mean, providing a more accurate safety assessment than traditional hot-spot analysis.
Methodology
modeling
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.
- incidence prevalence
- comparative international
- naturalistic crash near crash
- induced exposure
- urban rural setting
- demographic disparities
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