This paper explores the application of artificial intelligence to improve traffic safety. By creating indicators for traffic safety monitoring and using advanced artificial intelligence techniques, can be identified to warn of traffic accidents and improve the safety of all traffic participants. The paper aims to contribute contribute to the understanding of the potential of artificial intelligence in solving challenges in the field of traffic safety, but also to present the practical application of artificial intelligence in the field of creating models for improving traffic safety.
The focus of this research paper is the analysis of the safety of vulnerable categories of traffic participants in Macedonia. Through the realization of the case study, in the period from 2021 to 2023, the research methodology is used to identify and analyze the main challenges and risks to which pedestrians, cyclists and motorcyclists are exposed. The research process includes the analysis of data on traffic accidents obtained from the investigative procedure in the state prosecutor's offices in Macedonia. The obtained results and conclusions provide a detailed overview of the problems and challenges related to traffic accidents with vulnerable categories of participants, injuries, infrastructure, etc. Based on the obtained analytical results, recommendations were given for improving safety and traffic culture, which would have significant potential for improving the safety of vulnerable traffic participants and for implementing effective measures and innovative solutions for their protection.
This paper presents an analysis of the ejection distance of a cyclist's body during a collision with a passenger vehicle using the PC-Crash 9.0 software package. In the preparation of this paper, deductive, quantitative, and comparative methods were used. The comparative method was utilized to compare the results of traditional mathematical relations with those obtained via computer simulation. The data obtained was statistically processed and conclusions were drawn based on it. The key segment of the analysis is the dynamics of the cyclist's body ejection. The research variables in the study are factors that influence the dynamics of the collision, including the collision angle, the speed of the vehicle and bicycle, the cyclist's mass, as well as other physical characteristics of the participants in the traffic accidents. The determination of the cyclist's ejection distance was conducted using the PC-Crash 9.0 software package based on 1,620 cases of collisions between passenger vehicles and bicycles. The results showed a clear correlation between the vehicle's speed and the cyclist's ejection distance during the collision.
U ovom radu je razvijen i validiran analitički model za proračun daljinu odbačaja mopeda prilikom bočnog sudara sa automobilom. Istraživanje temeljimo na 504 simulacijskih scenarija saobraćajnih nezgoda dobijenih primenom softverskog paketa PC Crash 9.0. koje obuhvačaju releventne brzinske opsege oba vozila. Za svaku od 4 grupe brzina mopeda izvršena je uporedna analiza između polinomskih regresionih modela od trećeg do sedmog stepena, eksponencijalnih i stepenih, kako bi se odabrao model sa največom preciznošću odnosno najmanjom greškom. Na osnovu odabranih optimalnih modela, njihovi kojeficijenti su izražene kao funkcije brzine mopeda i aproksimirane što rezultira analitičkom funkcijom. Konačni model je testiran korišćenjem pomoču nekoliko fundamentalnih statističkih indikatora (MAE, RMSE, R2, t i F test), postižuči srednju relativnu grešku od 3,95%, a sva individualna odstupanja su u granici prahvatljive tačnosti.