Lidé na fakultě
Ing. Igor Gavrilov

2026, 2026 Smart City Symposium Prague (SCSP), IEEE eXplore), p. 1-3), ISBN 979-8-3315-8260-9, ISSN 2831-5618

Accurate identification of hazardous road segments is essential for improving road safety and prioritising preventive interventions. This study proposes a neural network-based framework for predicting hazardous road segments by leveraging detailed geometrical properties of roads, including lane number, lane width, and curvature. The proposed framework was applied to the D1 highway in the Czech Republic, using geospatial data and reported road accidents from 2024. The road geometry was divided into 10 m sectors, from which sequences of 300 m were generated to form input for a Long Short-Term Memory (LSTM) network. The model was trained using a weighted binary cross-entropy loss function to address class imbalance, where hazardous segments represented only 29.2 % of samples. The trained model demonstrated strong predictive performance, achieving a weighted F1 score of 0.895 and a macro F1 score of 0.906, which effectively distinguished between hazardous and non-hazardous highway segments. The results demonstrate that integrating detailed road attributes within an LSTM-based framework can effectively capture spatial dependencies and non-linear relationships associated with accident occurrence. The framework provides a scalable approach for identifying high-risk segments across various road types, offering valuable support for data-driven decision-making in transportation safety management.

Stať ve sborníku z prestižní konf. (Scopus)

2026, European Transport Research Review, 18 (1), p. 1-13), ISSN 1867-0717

This study investigates the forecasting of daily road traffic accidents using time series models, through a controlled comparative evaluation of baseline SARIMA and Hybrid SARIMA-LTSM frameworks. We employ a Seasonal Autoregressive Integrated Moving Average (SARIMA) model to capture linear and seasonal patterns in accident data from 2010 to 2023. To address potential non-linearities, we extend the approach by modelling SARIMA residuals using Recurrent Neural Networks (RNNs), including bidirectional RNN (BiRNN) and Long Short-Term Memory (LSTM) variants, to develop a hybrid SARIMA-RNN framework. The hybrid model integrates SARIMA forecasts with RNNbased residual predictions as a methodological extension whose empirical performance is rigorously evaluated against the baseline. Within residual-domain model selection, the unidirectional LSTM achieved the lowest RMSE (0.184) and MAPE (1.0966%) among tested architectures, confirming its effectiveness in capturing temporal dependencies in the log-residual space. A hybrid SARIMA-LSTM model was also evaluated to enhance residual prediction; however, the one-step-ahead rolling-origin strategy with a 7-day sliding window did not improve accuracy relative to the baseline SARIMA, yielding only minor increases in RMSE and MAPE.

Článek v periodiku excerpovaném SCI Expanded

2026, International Symposium on Applied Geoinformatics 2026, Ostrava, VŠB - Technical University of Ostrava), p. 1-12), ISBN 978-80-248-4869-3

Road safety studies often rely on predefined road segments of varying lengths, which can introduce spatial inconsistency and bias in statistical modelling. This paper proposes a GISbased framework that uses dynamically generated fixed-length road segments before and after each accident location to evaluate the influence of road geometry and temporal factors on multi-vehicle crash occurrence. The approach ensures a spatially consistent and localised assessment of highway conditions at each crash site. The methodology is demonstrated using detailed police-reported accident data from the D1 highway in the Czech Republic, covering January to December 2025, combined with a national road network dataset. For each accident, local geometric attributes, including road width, number of lanes, and curvature, were derived from a 300 m dynamically extracted highway segment centred on the projected crash location. A binomial generalised linear model with a logit link function was applied to estimate the probability of multi-vehicle accidents as a function of these geometric factors and temporal variables, including time of day, month, and day of week. Results show that wider road segments and daytime periods, particularly morning and afternoon hours, significantly increase the likelihood of multi-vehicle crashes, whereas higher lane counts and weekend periods are associated with lower risk. This framework provides a transferable, spatially robust tool for localised road safety assessment and infrastructure planning.

Stať ve sborníku z lokální konf. cizojazyčně

2025, Konference GIS Ostrava 2025, Ostrava, Vysoká škola báňská - Technická univerzita Ostrava), p. 1-12), ISBN 978-80-248-4795-5, ISSN 1213-239X

Tento článek se zaměřuje na statistickou analýzu vzdálenosti mezi místy dopravních nehod na dálnicích první a druhé třídy v České republice. Analyzovaná data zahrnují široké spektrum dopravních nehod, které se staly na těchto komunikacích v období 2007-2024. Cílem je identifikovat oblasti se zvýšenou koncentrací incidentů, které vyžadují další bezpečnostní opatření. Využití lokalizace nehod v GIS poskytuje přehled o prostorovém rozmístění nehod a pomáhá odhalit segmenty silnic, kde je pravděpodobnost nehody vyšší. Klasifikace typu silniční komunikace na základě vektorových dat z OpenStreetMap přispívá k rozlišení mezi dálnicemi první a druhé třídy. V analýze se statisticky hodnotí vzdálenosti mezi místy nehod. Výsledky studie slouží jako podklad pro tvorbu preventivních opatření zaměřených na snížení rizika nehod na klíčových úsecích české dálniční sítě.

Stať ve sborníku z lokální konf. česky

2024, Neural Network World, 34 (6), p. 317-328), ISSN 1210-0552

This study analyzes the impact of COVID-19 restrictions on road accident trends in the Czech Republic from 2015 to 2024, utilizing a comprehensive dataset of over one million recorded accidents. The research highlights a significant decline in accident rates during the strict lockdown periods, correlating reduced mobility with fewer traffic incidents. As restrictions eased, accident rates rose again, revealing seasonal variations and regional disparities, particularly in urban areas like Prague. Findings suggest that the pandemic has reshaped commuting patterns and could influence future traffic management strategies. The analysis underscores the need for targeted policies to enhance road safety, especially during high-risk seasons.

Článek v periodiku excerpovaném SCI Expanded