Geoinformation method for identifying and analyzing areas of concentration of traffic accidents

DOI: 10.35595/2414-9179-2024-2-30-445-455

View or download the article (Rus)

About the Authors

Lev A. Izrailev

Southern Federal University,
Shevchenko str., 2, Taganrog, 347900, Russia,
E-mail: izrailev@sfedu.ru

Larisa V. Gordienko

Southern Federal University,
Shevchenko str., 2, Taganrog, 347900, Russia,
E-mail: lgordienko@sfedu.ru

Abstract

According to the World Health Organization, about 1.25 million people die each year in car accidents, and 30–50 million people get seriously injured. In the Russian Federation, this figure has averaged about 17 000 deaths and 201 000 injuries per year over the past 10 years. At the same time, the number of cars on the roads is only increasing every year. In these conditions, the task of reducing the number of road traffic accidents and the severity of their consequences remains relevant. This paper proposes a method for identifying and analyzing areas where traffic accidents are concentrated, taking into account the spatial reference of the event and study of the territory. Areas with homogeneous spatial characteristics were combined into clusters and then analyzed in order to better understand the nature of the occurrence of accident concentration areas. Understanding the reasons for the emergence of areas with high accident rates would contribute to more accurate forecasting for these areas. This will also make it possible to carry out more rational measures to reduce hazards in these areas. As a result, this will reduce the number of transport accidents and their damage. The work uses various density-based clustering algorithms: DBSCAN, HDBSCAN, OPTICS. Each of these algorithms is tested in two variants of the search distance and the minimum number of objects per cluster. For accident analysis, it was proposed to use ArcGIS Pro geographic information system (GIS) software and accident location data. GIS ArcGIS Pro has powerful functions for spatial analysis, visualization, and spatial data processing. The paper also concludes which density-based clustering algorithm would be best suited for traffic accident analysis.

Keywords

traffic accident, cluster analysis, spatial analysis, GIS

References

  1. Al-Haji G. Traffic safety in developing countries—new approaches in technology transfer by using distance education technique. Master’s thesis, LITH-ITN-2001:156-SE, Linkoping University, Sweden, 2001.
  2. Arjun D. Clustering techniques and their applications: a review. American journal of advanced computing, 2020. V. 1. No. 4. P. 1–6. DOI: 10.15864/ajac.1404.
  3. Gershtein A.M., Terekhov A.N. Identification of high-risk areas on Massachusetts roads in 2013–2018. Computer tools in education Journal, 2021. No. 1 (19). P. 46–58 (in Russian). DOI: 10.32603/2071-2340-2021-1-46-58.
  4. Eliseev M.E., Galkina E.D., Mazunova L.N. Criterion for identifying accident concentration areas when developing an intelligent information transport system. Proceedings of NNSTU n.a. R.E. Alekseev, 2021. No. 1. P. 14–23 (in Russian). DOI: 10.46960/1816-210X_2021_1_14.
  5. Nisha, Kaur P.J. A survey of clustering techniques and algorithms. 2015 2nd International conference on computing for sustainable global development (INDIACom), 2015. P. 304–307.
  6. Prasad V., Srikanth T. A survey on clustering algorithms and their constraints. International journal of intelligent systems and applications in engineering, 2023. No. 11 (6s). P. 165–179.
  7. Smeed R.J. Some statistical aspects of road safety research. Journal of the royal statistical society. Series A (general), 1949. V. 112. No. 1. P. 1–34.
  8. Soloviev A.A., Zharkikh J.I., Krasnoperov R.I., Nikolov B.P., Agayan S.M. GIS-oriented solutions for advanced clustering analysis of geoscience data using ArcGIS platform. Russian journal of Earth sciences, 2016. V. 16. DOI: 10.2205/2016ES000587.
  9. Haoxiang Xu. Research on clustering algorithms in data mining. 2022 3rd International conference on big data, artificial intelligence and internet of things engineering (ICBAIE), 2022. P. 652–655. DOI: 10.1109/ICBAIE56435.2022.9985831.

For citation: Izrailev L.A., Gordienko L.V. Geoinformation method for identifying and analyzing areas of concentration of traffic accidents. InterCarto. InterGIS. GI support of sustainable development of territories: Proceedings of the International conference. Moscow: MSU, Faculty of Geography, 2024. V. 30. Part 2. P. 445–455. DOI: 10.35595/2414-9179-2024-2-30-445-455 (in Russian)