Highlights Foursquare data was used to uncover different zones based on its venues’ composition. A K-Means model was trained to group zones with similar venues’ composition. Five clusters were identified in Medellín according its venues’ composition. Analysis performed to each cluster based on its traffic collisions trends was made. Proposed methodology is useful to develop located policies to reduce traffic collisions.

    Abstract Introduction Road traffic collisions are one of the leading causes of deaths and long-term disability in the world, particularly in low- and middle-income countries. In Medellín (Colombia), road traffic collisions have increased in recent years, which has led to the need of developing optimal strategies to reduce these events. Methodology This study presents a systemic diagnosis of the trends in road traffic collisions in 219 territorial administrative zones of Medellín. To achieve this, we used a K-Means clustering model to group all the zones in Medellín into different clusters using the composition of their venues using Foursquare data, then we associated the different types of road traffic collisions of each cluster to analyse trends in road traffic collisions in each cluster. Results Results show that crashes are the most common road traffic collision for all clusters, while zones surrounded by landscapes and parks have more run overs than fallen occupants. In addition, places with food trucks, gyms, bars and stores have more fallen occupants and other types of road traffic collisions than run overs and they are also the most dangerous zones of Medellín in terms of road traffic collisions per area. Finally, run overs and fallen occupants generally have a major severity than crashes, and these road traffic collisions are common in places with leisure venues and zones with food trucks, gyms, bars and stores. Conclusions This diagnosis provides important insights to public agencies in the prioritisation and evaluation of strategies to reduce road traffic collisions in Medellín, based on the differences of clusters.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Modelling road traffic collisions using clustered zones based on Foursquare data in Medellín



    Publication date :

    2021-04-27


    Size :

    7 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English




    Modelling the effects of road pricing on traffic using ERP traffic data

    Xie, Litian / Olszewski, Piotr | Elsevier | 2011


    Evaluation of Pollutants Dispersion in an Urban Traffic Scenario in Medellín

    Gómez-Pérez, Cesar Augusto / Espinosa, Jairo | TIBKAT | 2019




    Spatial analysis methods of road traffic collisions

    Loo, Becky P. Y. / Anderson, Tessa Kate | TIBKAT | 2016