Within the activities of planning, organizing, or regulating traffic, a considerable amount of various traffic data is actively used. For responsible data-based decision-making, the quality of the utilized data is crucial. However, defining the specifics of this quality and establishing procedures for its control is often ambiguous. The primary objective of the research was to review methods for ensuring or managing data quality within the transportation environment. Building on the collected information, a practical data governance approach was delineated to be applied to specific transportation data. Simultaneously, data from detectors and floating vehicles on the highway network in the Czech Republic were analyzed for the same reason. The results show that the chosen data governance approaches are applicable to transportation data, even in the task of evaluating data quality in the detail of individual dimensions. In the context of semantic data quality control, it is necessary to explore additional suitable tools, such as the use of statistical models, to reliably distinguish real errors or otherwise reduced data quality from genuinely truthful values, such as traffic conditions.
Floating car data (FCD) is a beneficial consequence of the modern telematics technologies in the logistics and automotive industry, and can thus be an additional source of transportation data about current traffic flows on the road network. There is a state-provided free source of this data available in the Czech Republic, which is relatively new and gradually being introduced for use by the public and private sectors. This data source appears to be very promising for nationwide coverage; however, due to its novelty, it is necessary and advisable to investigate its qualitative parameters and assess whether it actually adequately characterizes vehicle traffic flows. The present paper deals with the characteristics and quality of processed FCD outputs before and during the covid-19 pandemic, for the possibility of capturing and comparing significant changes in the behaviour of road users. The paper describes the impact of the associated pandemic measures on traffic in the country in general, which can be detected during FCD processing, as well as the impact on the quality of FCDs as such. It also tests the hypothesis that although the quality of the data has been reduced as a result of the pandemic, the data still have sufficient predictive value. The paper describes the specifics of the data in the spatial and temporal domain analyzed and makes recommendations for future work for use in models and predictions.
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 $\mathbf{F 1}$ 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.