Patterns of rainy weather crashes: Applying rules mining

Das, Subasish; Dutta, Anandi K; Sun, Xiaoduan · 2019 · Journal of Transportation Safety & Security

DOI: 10.1080/19439962.2019.1572681

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 investigates the specific patterns and contributing factors associated with traffic crashes occurring during rainy weather in Louisiana. Motivated by the state’s high precipitation levels and the significant proportion of fatal crashes occurring in rain (approximately 25%), the research aims to identify actionable safety countermeasures. While previous studies have examined crash frequency and severity using parametric models, this research employs association rules mining, a nonparametric data mining technique, to uncover complex relationships between roadway environment, driver characteristics, and crash outcomes without predefined assumptions about variable dependencies. The methodology utilized eight years of police-reported crash data (2004–2011) from the Louisiana Department of Transportation and Development. After filtering for rainy weather conditions and removing records with missing or questionable data, the final dataset comprised 58,288 crash records. The researchers selected key variables including roadway alignment, lighting conditions, driver age and gender, crash severity, and collision type. Using the Apriori algorithm in R, they generated association rules with a minimum support threshold of 1% and a minimum confidence of 60%. The strength of the discovered rules was evaluated primarily based on lift values to identify significant dependencies between antecedent conditions and consequent crash outcomes. The results indicate that single-vehicle run-off-road crashes are the predominant crash type during rainy weather. These crashes are strongly associated with grade-curve aligned roadways, curved roadways, and roadways lacking streetlights at night. Specifically, the absence of street lighting in dark conditions showed a high lift value for single-vehicle crashes, suggesting a strong dependency. Additionally, no-injury and sideswipe crashes were found to be significant in number. The analysis revealed that moderate injuries are dominant in single-vehicle crashes, while poor illumination is associated with straight, level-aligned roadways. Drivers aged 15–44 were identified as particularly vulnerable to run-off crashes when roadway conditions included poor illumination and curves. The significance of this study lies in its application of association rules mining to traffic safety, a methodological approach not previously used for rainfall-related crash analysis. By identifying specific combinations of environmental and human factors that lead to crashes, the findings provide targeted insights for safety practitioners. The results suggest that infrastructure improvements, such as enhanced lighting on curved roadways, and targeted safety interventions for younger drivers could effectively reduce the risk and severity of crashes during rainy conditions, supporting Louisiana’s goal of achieving zero traffic fatalities.

Key finding

Single-vehicle run-off-road crashes are the predominant crash type during rainy weather and are significantly associated with curved roadways and the absence of streetlights at night.

Methodology

dataset

Sample size: 58288

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 author_sweep_intake on 2026-05-28.

StageOutcomeToolModelPromptAttemptsCompleted
discover success author_sweep 2 2026-05-28
archive success canonical_url 7 2026-06-06
extract success cached 3 2026-06-10
clean success clean 1 2026-06-04
chunk success chunk 1 2026-06-04
embed success embed Qwen/Qwen3-Embedding-8B 1 2026-06-04
enrich success 1 2026-05-28
promote success 1 2026-06-04
summarize success llm qwen3.6-27b-prismaquant summ-v5 2 2026-06-10
tag success vector_similarity 15 2026-06-11
verify success 2 2026-06-10

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.

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).