Investigation on the wrong way driving crash patterns using multiple correspondence analysis

Das, Subasish · 2017 · Accident Analysis & Prevention

DOI: 10.1016/j.aap.2017.11.016

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 persistent traffic safety issue of wrong-way driving (WWD) crashes, which, despite their low frequency, result in disproportionately high rates of fatalities and severe injuries. The authors argue that traditional statistical methods, such as logistic regression, often fail to adequately control for the impact of all contributing variables due to the rarity of WWD events, leading to potential bias. To overcome this limitation, the research employs Multiple Correspondence Analysis (MCA), a distribution-free, dimensionality reduction technique that identifies associations among multiple categorical variables without requiring prior assumptions or distinguishing between explanatory and response variables. The analysis utilized five years (2010–2014) of crash data from Louisiana. From an initial pool of 800,000 crashes, the researchers identified and confirmed 1,203 WWD incidents involving at-fault drivers. The dataset was constructed by merging crash, roadway geometry, and vehicle databases, resulting in a matrix of 1,203 records across 24 categorical variables. These variables included driver characteristics (age, gender, impairment, license state), vehicle attributes, temporal factors, and geometric/environmental conditions such as lighting, signage, and roadway type. The application of MCA generated a proximity map of variable categories, revealing sixteen significant clusters that define specific WWD crash scenarios. Key findings indicate strong associations between WWD crashes and specific factors: older drivers, non-local drivers (those with out-of-state licenses), and impaired drivers. Environmental and geometric factors also played critical roles, with crashes frequently occurring on roadways lacking physical separations, those with inadequate signage and markings, and those with higher posted speeds. Additionally, darkness and the absence of roadway lighting were identified as significant contributing conditions. The study noted that while WWD crashes constitute a small fraction of total crashes, their fatality rate is significantly higher than that of other crash types. The significance of this research lies in its demonstration of MCA as an effective tool for analyzing rare crash events, offering a clearer view of complex variable interactions than conventional parametric methods. By identifying distinct clusters of risk factors, the study provides actionable insights for transportation authorities. The findings support the implementation of targeted countermeasures, such as improved signage, enhanced lighting, and physical barriers on high-risk roadways, as well as monitoring programs for non-local and older drivers. This approach aids in developing more effective safety strategies to mitigate the severe outcomes associated with wrong-way driving.

Key finding

Multiple correspondence analysis of Louisiana crash data revealed sixteen significant clusters of contributing factors, notably linking wrong-way driving crashes with older drivers, unlit roadways at night, and inadequate physical separations or signage.

Methodology

dataset

Sample size: 1203

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 2026-05-27.

StageOutcomeToolModelPromptAttemptsCompleted
discover success author_sweep 2 2026-05-27
archive success canonical_url 7 2026-06-06
extract success cached 3 2026-06-10
clean success clean 1 2026-06-07
chunk success chunk 1 2026-06-07
embed success embed Qwen/Qwen3-Embedding-8B 1 2026-06-07
enrich failed 5 2026-07-02
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).