Original scientific paper
https://doi.org/10.7307/ptt.v38i2.1070
The Use of Neural Networks to Predict Frequency of Road Accidents in Poland and Slovenia
Piotr GORZELAŃCZYK
orcid.org/0000-0001-9662-400X
; Department of Transport, Stanislaw Staszic State University of Applied Sciences, Piła, Poland
*
Marina ZANNE
orcid.org/0000-0001-7288-670X
; Faculty of Maritime Studies and Transport, University of Ljubljana, Portorož, Slovenia
* Corresponding author.
Abstract
Despite a general decline in recent years, road accidents are still a major problem worldwide, including in Poland and Slovenia. Even if there is no denying that the COVID-19 epidemic has had an impact on accident rates, the statistics nonetheless show that urgent action is required to further lower their frequency. The purpose of this study is to project how many traffic accidents will occur in Slovenia and Poland between 2024 and 2030. This was accomplished by analysing historical data on yearly accident incidence from Eurostat and the Polish Police. To create the predicted numbers, certain neural network models were used, utilising these datasets. The results indicate that the number of road accidents is likely to stabilise soon. This forecast is influenced by a number of factors, including the continued increase in car ownership and the ongoing development of road infrastructure, including the construction of new motorways and roads. It is important to remember that the accuracy of the forecast is susceptible to inherent limitations due to the random sampling of the data used to test, validate and train the models.
Keywords
traffic accident; pandemic; forecasting; neural networks; Poland; Slovenia
Hrčak ID:
344878
URI
Publication date:
24.2.2026.
Visits: 171 *