Applying Bayesian data mining to measure the effect of vehicular defects on crash severity

Das, Subasish; Dutta, Anandi; Geedipally, Srinivas Reddy · 2019 · Journal of Transportation Safety & Security

DOI: 10.1080/19439962.2019.1658674

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 relationship between vehicular defects, vehicle characteristics, and crash severity to inform roadway safety policies, particularly regarding vehicle inspection mandates. Motivated by the National Motor Vehicle Crash Causation Survey’s finding that 2% of crashes are caused by vehicle defects and Louisiana’s disproportionately high rate of defect-related fatalities (3%), the authors aimed to identify specific risk subgroups. The research addresses a gap in understanding how factors like vehicle age, type, and specific defect types influence injury outcomes, providing evidence relevant to states reconsidering annual or biennial safety inspections. The researchers analyzed seven years (2010–2016) of police-reported traffic crash data from Louisiana, filtering for 24,185 crashes where vehicle defect was the primary contributing factor. They employed the Empirical Bayes Geometric Mean (EBGM) method, a Bayesian data mining technique, to analyze the association between vehicle characteristics (type, defect, and age) and crash severity. Unlike conventional statistical models that require assumptions or threshold settings, EBGM identifies significant associations in large, sparse contingency tables by applying Bayesian shrinkage corrections to relative reporting ratios. This approach allows for the discovery of risk subgroups without imposing prior assumptions on the data structure. The findings reveal that defective brakes, tire failures, worn tires, defective suspension, and engine failure account for approximately 81% of defect-related crashes in Louisiana. Vehicle age is significantly associated with severe injury crashes, with vehicles older than seven years showing strong associations with defect-related incidents. Motorcycles are over-represented in crashes involving injuries, particularly those involving tire issues, defective steering, or engine failure. Specific high-risk combinations identified include worn tires on SUVs and cars aged 7–20 years, which are linked to fatal crashes, and defective brakes on motorcycles aged 3–10 years, which are linked to incapacitating injuries. Additionally, vehicles older than 20 years were associated with defective rear lights and steering. The study concludes that Bayesian data mining effectively identifies nuanced risk subgroups that traditional methods might overlook, offering valuable insights for safety stakeholders. The results suggest that older vehicles and specific defect types, particularly tires and brakes, pose significant safety risks. These findings support the importance of periodic vehicle inspections and can help policymakers prioritize safety targets and legislative changes. By highlighting the correlation between vehicle aging and defect severity, the research provides empirical evidence to guide interventions aimed at reducing defect-related crashes and fatalities.

Key finding

Vehicle age is significantly associated with severe injury crashes, with worn tires and defective brakes identified as the most over-represented defect categories contributing to high-severity outcomes.

Methodology

dataset

Sample size: 24185

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 2026-06-07
chunk success chunk 1 2026-06-07
embed success embed Qwen/Qwen3-Embedding-8B 1 2026-06-07
enrich success semantic_scholar 4 2026-06-15
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).