This paper presents a consolidated overview of random forest-based methods for road traffic safety research, mainly focusing on recent methodological developments tailored to particular needs in three typical crash data modelling tasks: crash frequency modelling, crash injury severity modelling, and countermeasure evaluation. Along the way, basic ideas behind the traditional random forests and how they fit with crash data analytics are first described. Newer forest-based alternatives are then detailed, together with selected empirical applications for illustration. Random forests have achieved significant success in this field due to their intuitive algorithmic ideas and competitive predictive performance, and we believe that a comprehensive summary of their new features would be helpful for road traffic safety researchers and practitioners to better exploit this strand of tools. Lastly, future research directions are outlined.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:
More from IATSS Research