Pedestrian-involved road traffic accidents represent a serious challenge for traffic safety and require a comprehensive analysis of the interactions within the driver–vehicle–road–environment system. The objective of this study is to develop a methodology for risk assessment in road traffic accidents involving pedestrians based on the analysis of real court cases and dynamic modeling of vehicle motion. A database of 105 court cases was analyzed, enabling the identification of the main factors influencing the occurrence of pedestrian-related accidents. Based on this analysis, a system of 31 linguistic variables was developed to characterize driver behavior, vehicle technical characteristics, and road environment conditions. These variables were integrated into a mathematical model for quantitative risk assessment that enables the evaluation of the relative influence of different groups of factors on accident probability. In addition, a dynamic model of vehicle motion was developed to analyze the influence of driver reaction time, vehicle speed, and road surface conditions on the possibility of avoiding a collision. The results of the numerical analysis demonstrate that even minimal delays in hazard perception and driver reaction significantly increase the probability of pedestrian-related accidents. These findings highlight the importance of early hazard detection and automated emergency braking systems. The proposed methodology provides a framework for integrating intelligent driver assistance systems and automated braking control aimed at improving the safety of vulnerable road users.

This This study presents a comparative analysis of two methodologies for predicting the risk of pedestrian-involved traffic accidents. The methodologies examined include one based on proportional risk distribution and another employing the machine learning algorithm Random Forest. The primary objective is to evaluate the applicability and accuracy of the proposed approaches through the processing and analysis of data derived from real-world cases of concluded legal proceedings. Linguistic variables, defined as risk factors, are classified and quantified based on expert evaluations. The results include interpolation models and graphical representations illustrating the risk severity according to the two methodologies. The analysis demonstrates that both methodologies are applicable for risk assessment, with the Random Forest algorithm providing superior accuracy and reliability in processing complex and heterogeneous data. Furthermore, a correlation analysis confirms a statistically significant linear relationship between the results of the two approaches. Visualization of the results through various graphical tools supports an objective comparison of the methodologies and their application in transport safety analysis.

Pedestrian-involved road traffic accidents remain a major road safety concern and require a comprehensive analysis of the interactions within the “driver–vehicle–road–environment” system. Despite the substantial body of research in this field, most existing models rely on predefined parameters and experimental data, while the considerable information potential contained in the reasoning underlying judicial decisions concerning real-world road traffic accidents remains underutilized. This study proposes a novel approach in which the reasoning underlying judicial decisions is considered a structured source of technical information. Through a systematic analysis of the documented factual circumstances, technical expert reports, witness testimonies, and judicial findings, quantitative and qualitative characteristics of the transport system are extracted. This information is subsequently transformed into a system of linguistic variables describing driver behavior, vehicle technical characteristics, and road-environment conditions. The use of a symmetrically structured system of variables enables a consistent description of different vehicle–pedestrian interaction scenarios, regardless of the direction of motion and their relative positions. The geometric symmetry of the vehicle with respect to its longitudinal axis, together with the symmetrical representation of the traffic corridor, provides a unified coordinate framework for dynamic modeling and quantitative risk assessment. The proposed methodology was developed through an analysis of the reasoning underlying 105 judicial decisions concerning real-world pedestrian-involved road traffic accidents. As a result, 31 linguistic variables were identified and combined into an information vector describing each individual case. The analysis of the empirical distribution of the risk assessments shows mean values of 55.42% for driver-related factors, 49.29% for vehicle-related factors, and 35.77% for factors characterizing the road and environment. The results obtained indicate the dominant influence of the human factor on the formation of the integrated risk.

This article presents a developed information-based approach for analyzing the factors contributing to pedestrian road traffic accidents (RTAs) through the application of advanced machine learning methods. The approach encompasses the identification and classification of key risk factors, the structuring of linguistic variables, and the quantitative assessment of their impact on risk. Machine learning algorithms are employed to model complex dependencies, alongside statistical and regression analyses to explore the causal relationships among the components of the transportation system-driver, vehicle, road, and environment. At the core of this approach lies the systematic utilization of empirical data and expert evaluations, enabling accurate risk predictions and assessments. The developed methodology facilitates integrated safety management and the creation of effective preventive strategies.