Multicriteria method for assessing the road traffic safety level on a street network section

Introduction. Despite the general decrease in recent years, the accident rate in the Russian Federation remains at a sufficiently high level. Mortality as a result of road traffic accidents in our country is also at a fairly high level. Meanwhile, the occurrence of a Road Traffic Accident (RTA), as a phenomenon, depends on the presence of factor or on the combination of factors that exceed their normal state, thereby initiating a dangerous chain of events leading to an accident. In this regard, assessing the influence of various factors on the accident rate, their identification, and prevention is a relevant scientific task.

Materials and methods. This article outlines the main principles for the development of an Intelligent Informational System based on the integrated consideration of significance of an interconnected array of criteria and sub-criteria, obtained by normalized expert and experimental evaluation of the main final importance of subsystem elements that constitute the common system at various levels of forming a comprehensive assessment of road safety on a separate section of the street network. The object of assessment is a traffic accident as a phenomenon on a street network section, originated from a set of predictors (final elements) of the assessment, which at different evaluation levels make sub-criteria, criteria, and subsystems of the general assessment system, one or more of which has a deviation from the normal state, which leads to an emergency situation and the occurrence of a traffic accident as a phenomenon.

Conclusions. Multicriteria method for assessing road safety on a road network section using final predictors for evaluating individual subsystems within the Driver-Vehicle-Road-Environment (DVRE) system has been developed. The ranking and interinfluence of these predictors are implemented in the form of a software product that includes computational methods based on fuzzy inference rules. The assessment of the system state predictors is carried out exclusively by expert and experimental methods at locations of concentrated road traffic accidents on the street network section. The specified method is universal, employs several levels for evaluating various influencing factors, and can be supplemented with additional estimating predictors if necessary.

Research scope/Potential for further use of the scientific work results. The proposed method can serve as a basis for developing an analytical database on the state of the transport environment in our country in terms of its safety, as well as act as a foundation for the further transformation of the Driver-Vehicle-Road-Environment (DVRE) complex into more advanced systemic forms.

Practical significance. The research results can be used in a comprehensive assessment of road traffic safety to identify factors influencing the occurrence of dangerous situations, for conducting preventive measures to eliminate these factors as causes of accidents.

Originality. For the first time, a comprehensive approach has been applied to assess road traffic safety on the street network section. This approach uses the decomposition of the interacting subsystems of the Driver-Vehicle-Road-Environment (DVRE) system into evaluation predictors whose state is determined not by predictive or statistic methods, but by the expert and experimental research with the establishment of criteria that exceed the norm.

1. Rubtsova M.V. Road safety in Russia: problems and ways to solve them. Road safety. 2021; 3: 4-7. (in Russ.)

2. Pirogov Ya.E., Evtyukov S.S., Terentyev A.V. The concept of an information and analytical platform in the road transport system of a metropolis. Bulletin of Civil Engineers. 2022; 5 (94): 123-129. (in Russ.)

3. Novikov A.N., Eremin S.V., Lomakin D.O. Assessment of the level of road safety at the regional level. World of transport and technological machines. 2020; 3 (70): 72-79. (in Russ.) Https://doi.org/10.33979/2073-7432-2020-70-3-72-79

4. Loparev E.A. , Sergunova A.S. Influence of external factors on the level of road accidents associated with pedestrian collisions. Road safety. 2022; 1. 28-34 s. (in Russ.)

5. Petrov A.I., , Evtyukov S.A. Specificity of the use of an information-entropy approach in the organization and management of urban road safety systems. Bulletin of Civil Engineers. 2022; 4 (105): 80-88. (in Russ.)

6. Podoprigora N.V. Structure and functioning of the Driver-Car-Road-External Environment system. Bulletin of Civil Engineers. 2022; №2 (91): 154-159. (in Russ.)

7. Novikov I.A., Lazarev D.A., Ziborova E.I. Zhikharev A.G. Improvement of road transport expertise in the field of road safety by introducing a methodology for the integrated determination of the components of the mechanism of road accidents. World of transport and technological machines. 2024; 1-3 (84): 71-81. (in Russ.) Https://doi.org/10.33979/2073-7432-2024-1-3(84)-71-81

8. Zelikov V.A., Shatalov E.V., Yernapeshyan E.N. [et al.] Traffic Safety Management in the field of motor transport services for the population. International Research Journal. 2015; 8 (39). (in Russ.)

9. Evtyukov S.S., Vasiliev Ya.V., Golubev D.S. Accident prediction model based on adaptive neuro-fuzzy inference. World of transport and technological machines. 2025; № 2-2 (89): 126-133. (in Russ.) Https://doi.org/10.33979/2073-7432-2025-2-2(89)-126-133

10. Gavrilenko A.A. On the prospects for using neural networks in road safety analytics. Modern science. 2024; № 4. S. 8-11. (in Russ.)

11. Anikeev E.A. Application of fuzzy logic for traffic control. Actual directions of scientific research of the XXI century: theory and practice.2014; Volume 2 No. 4 h. 3: 315-318. (in Russ.)

12. Klyushin A.Yu., Kuznetsov V.N., Kucheryavykh V.V. Tasks of building an intelligent information system for road safety management. Software products and systems. 2010; 4. (in Russ.)

13. Mosin K.K. Using automated machine learning techniques to classify road accidents/Intelligent technologies in transport. 2023; 2. (in Russ.)

14. Xueqiu, Wang., Huanbing Gao, Zemeng Jia, Zijian Li BL-YOLOv8: An Improved Road Defect Detection Model Based on YOLOv8. Sensors 2023, 23, 8361. https://doi.org/10.3390/s23208361

15. Lazarev D.A., Grinyakin R.V., Strekalov D.P., Shatalov E.V. Application of digital algorithms in optimizing the road accident investigation process. Voronezh Scientific and Technical Bulletin No. 4 (46) December 2023; (in Russ.) Https://doi.org/10.34220/2311-8873-2023-107-119

16. Lazarev D.A., Ivashchuk O.D., Beketov Ya.M., Novikov I.A. Intelligent system for assessing the level of road safety on a section of the road network. Truck. 2025; 8: 44-53. (in Russ.) Https://doi.org/10.36652/1684-1298-2025-8-44-53

17. Ivashchuk O.D., Beketov Ya.M., Novikov I.A., Lazarev D.A. Intelligent system for assessing the level of road safety on the road network section: certificate of state registration of the computer program No. 2025660188; right holder FSAEI HE BelGNIU; State registration date: 22.04.2025. (in Russ.)