Estimation of Safety Performance Functions (SPF) at signalized intersections in Medellín Colombia
DOI: 10.15446/dyna.v87n214.83880
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 addresses the need for context-specific Safety Performance Functions (SPFs) to estimate crash frequency at signalized intersections in Medellín, Colombia. Motivated by high traffic fatality rates in Latin America and the limitations of applying generic models like the Highway Safety Manual (HSM) to local conditions, the authors aim to quantify safety risks using local data. The research focuses on evaluating how geometric features and traffic exposure influence crash severity, specifically distinguishing between Property Damage Only (PDO) crashes and Fatal or Injury (FI) crashes. The methodology utilizes crash records from 2012 to 2016 provided by the Medellín Secretariat of Mobility, alongside geometric and traffic volume data. After preprocessing and cleaning, the dataset included 447 signalized intersections (88 T-junctions and 359 four-leg intersections) with 9,974 total crashes. The authors employed a Negative Binomial (NB) regression model to estimate the SPFs, chosen to account for the over-dispersion inherent in crash count data, which violates the assumptions of Poisson regression. Exposure was defined as the product of the Annual Average Daily Traffic (AADT) on major and minor approaches. Variables analyzed included intersection type, lane counts, median presence, crosswalks, flow direction (one-way vs. two-way), and right-turn lanes. Statistical analysis was performed using R, with model fit assessed via Likelihood Ratio Tests and goodness-of-fit measures. The results indicate that exposure, geometry, and traffic volume are statistically significant predictors for both PDO and FI crashes. For PDO crashes, T-intersections, the presence of medians, crosswalks, and one-way approaches were found to reduce crash frequency. Specifically, one-way approaches showed the strongest safety benefit due to reduced conflicting traffic volumes. Conversely, an increase in the number of lanes increased PDO crashes. For FI crashes, T-intersections also reduced crash counts by a factor of 0.553, attributed to simpler geometry and lower approach speeds. However, two-way approaches increased FI crash counts by 1.30, and the presence of right-turn lanes significantly increased the likelihood of FI crashes due to conflicts with pedestrians. The NB models demonstrated significant over-dispersion parameters, confirming the appropriateness of the chosen statistical method. The significance of this study lies in providing calibrated, local SPFs that improve the reliability of safety evaluations in Medellín compared to generic international models. The findings highlight specific design features that enhance safety, such as one-way approaches and T-intersections, while identifying right-turn maneuvers as a critical risk factor for severe crashes. These insights support targeted safety audits and infrastructure improvements, such as enforcing pedestrian priority at right-turn locations. The study contributes to the broader field of road safety by demonstrating the efficacy of context-specific modeling in developing countries, aiding in the reduction of crash frequency and severity through evidence-based design criteria.
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | DOAJ | — | — | 1 | 2026-06-25 |
| archive | success | unpaywall | — | — | 1 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-25 |
| chunk | success | chunk | — | — | 1 | 2026-06-25 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-25 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-25 |
| verify | success | — | — | — | 1 | 2026-06-26 |
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.
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