Identifying relevant patterns between injury crashes and road safety inspection deficiencies

• Published in: Journal of safety research Vol. 93; pp. 99 - 134
• Main Authors: Gutiérrez-Rodríguez, Rubén, Rojí, Eduardo, Cuadrado, Jesús
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.07.2025
• Subjects: Accidents, Traffic - prevention & control; Accidents, Traffic - statistics & numerical data; Adult; Association rules; Bicycling - injuries; Crash severity; Environment Design; Female; Humans; Male; Motorcycles - statistics & numerical data; Pedestrians - statistics & numerical data; Road safety; Safety; Safety countermeasures; Spain - epidemiology; Traffic crash; Wounds and Injuries - epidemiology; Wounds and Injuries - prevention & control; Spain; DFB; CBA; KP; CSV; LIC; Crash severity; RSI; PRM; Association rules; AADT; DGT; CV; Traffic crash; GV/EJ; AADTT; LHS; Safety countermeasures; ESI; Road safety; ARM; CSM; RHS; EC
• ISSN: 0022-4375; 1879-1247; 1879-1247
• DOI: 10.1016/j.jsr.2025.02.014

•ARM method reveals hidden relationships in crash data for effective countermeasures.•Strong link found between vulnerable users and crash severity.•The systematization of road safety inspections enables scalable and efficient safety improvements.•Recommendations based on road safety inspections ensure practical applicability. Introduction: Crashes are the result of a complex interaction of many factors. However, the severity of injuries is often influenced by road conditions. Method: Data mining techniques, in particular Association Rule Mining (ARM), help to uncover non-trivial patterns in crash data, providing essential information for effective countermeasures to improve road safety and reduce crash severity. Using ARM, this study analyzes the relationship between crash severity and factors such as road deficiencies identified during safety inspections (Element Subject to Improvement or ESI) on two roads in Spain. Results: The primary findings reveal a strong link between vulnerable road users (pedestrians, cyclists, and motorcyclists) and crash severity. These results align with the Strategic Plan of the Basque Country, which aims to reduce severe road injuries by 50% by 2030, identifying the need for specific interventions to protect this type of user. Pedestrian casualty crashes are linked to inadequate restraint systems, while bicycle-related crashes are associated with glare from road orientation, inadequate shoulders, and inappropriate barriers. Unchanneled access points on roads also pose risks for cyclists. For motorcycles, factors such as inadequate speed limitations, poor shoulder design, and obsolete or inadequate restraint systems are strongly associated with crash severity. Poor signage, illegible road markings, water accumulation, and inadequate transport stops further contribute to increased crash severity. Conclusions and practical applications: Based on these findings, key countermeasures should focus on protecting vulnerable road users. Recommendations include installing pedestrian-friendly barriers and clear pedestrian pathways, widening shoulders, implementing anti-glare barriers for cyclists, and adding motorcycle-friendly restraint systems. Dynamic speed controls and better road signage are critical for reducing risks to motorcyclists. Other important measures include improving drainage systems to prevent water accumulation and redesigning public transport stops for safer bus merging areas. These targeted interventions can significantly reduce crash severity and support broader road safety goals.