Predicting Drivers' Crash Risk Based-on Previous Crash History

Das, Subasish · 2013 · KTH Publication Database DiVA (KTH Royal Institute of Technology)

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 investigates the characteristics of crash-prone drivers and evaluates the predictive power of past crash history for future crash risk. Motivated by the need to target safety education and enforcement programs more effectively, the authors aim to identify high-risk drivers who disproportionately contribute to highway crashes. The research addresses the gap in existing literature regarding detailed driver characteristics and the specific relationship between historical crash frequency and future probability of involvement. The analysis utilizes seven years of crash data (2004–2010) from Louisiana, focusing on at-fault drivers with valid driver identification records. Approximately 10% of records were excluded due to missing driver IDs. The researchers analyzed crash frequencies, severity, and driver demographics, while developing conditional probability matrices to estimate the likelihood of crashes in the seventh year based on the number of crashes in the preceding six years. Additionally, the study examined time gaps between crashes and contributing factors such as distraction, violation type, and crash geometry. The results demonstrate that a small subset of drivers is responsible for a large proportion of crashes; specifically, 5% of licensed drivers accounted for 35% of all crashes over the seven-year period. Crash risk is strongly correlated with history: drivers with no crashes in the previous six years had less than a 4% probability of crashing in the seventh year, whereas those with nine or more prior crashes had a probability exceeding 30%. Drivers with multiple crashes were predominantly male (71% for those with five or more crashes) and aged 20–40. These drivers also exhibited higher injury rates and shorter intervals between crashes compared to uniform distribution expectations. Contributing factors for repeat offenders included careless operation, distraction (particularly among younger drivers), and rear-end collisions. The study concludes that crash-prone drivers present a significant adverse effect on roadway safety and should be targeted for specialized interventions. The authors recommend that motor vehicle agencies implement license review programs that mandate safety education or license suspension for drivers with multiple crashes in short time periods. These programs should focus on distracted driving and careless operation, as the analysis suggests that subsequent crashes for these drivers are likely to be severe or fatal. The findings provide empirical evidence for shifting safety strategies from general public education to targeted regulation of high-risk individuals.

Key finding

A small subset of drivers with extensive crash histories accounts for a disproportionate share of total crashes, and their probability of future crashes increases significantly with the number of prior incidents.

Methodology

dataset

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 1 2026-06-04
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 skipped 4 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 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.

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).