Forecasting the number of road accidents in Poland in 2024–2030 – a comparative analysis of classical methods, neural networks, and forecasts generated by AI systems

The article presents a comparative analysis of forecasts of the number of road accidents in Poland in 2024–2030, developed using classical methods, artificial neural networks, and artificial intelligence (AI) tools. Based on historical data from 1990 to 2023, the accuracy, stability, and reliability of the forecasts were assessed using the mean absolute percentage error (MAPE) as the main evaluation metric. The results show that neural networks and exponential smoothing methods are characterized by the highest accuracy and stability of forecasts, while high-degree polynomial regression models and forecasts automatically generated by AI are characterized by greater variability and lower reliability. The forecasts indicate that the number of road accidents in Poland in the analyzed horizon is likely to remain stable at around 20,000–22,000 per year. The study emphasizes the importance of controlled modeling and conscious selection of forecasting methods in predicting road safety trends.

Authors

Institutions

Publication Details

Journal
Journal of Civil Engineering and Transport
Published
2026-11-16
DOI
https://doi.org/10.24136/tren.2026.009
Primary Topic
Forecasting Techniques and Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Forecasting the number of road accidents in Poland in 2024–2030 – a comparative analysis of classical methods, neural networks, and forecasts generated by AI systems

Piotr Gorzelańczyk
3 links
Journal of Civil Engineering and Transport
Forecasting Techniques and Applications
article

Forecasting the number of road accidents in Poland in 2024–2030 – a comparative analysis of classical methods, neural networks, and forecasts generated by AI systems

Piotr Gorzelańczyk
article en

Abstract

The article presents a comparative analysis of forecasts of the number of road accidents in Poland in 2024–2030, developed using classical methods, artificial neural networks, and artificial intelligence (AI) tools. Based on historical data from 1990 to 2023, the accuracy, stability, and reliability of the forecasts were assessed using the mean absolute percentage error (MAPE) as the main evaluation metric. The results show that neural networks and exponential smoothing methods are characterized by the highest accuracy and stability of forecasts, while high-degree polynomial regression models and forecasts automatically generated by AI are characterized by greater variability and lower reliability. The forecasts indicate that the number of road accidents in Poland in the analyzed horizon is likely to remain stable at around 20,000–22,000 per year. The study emphasizes the importance of controlled modeling and conscious selection of forecasting methods in predicting road safety trends.

Journal of Civil Engineering and TransportVol. 8(3)
Stanisław Staszic University of Applied Sciences in Piła (PL)
Industry, innovation and infrastructure, Sustainable cities and communities
Forecasting Techniques and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon