Analysis of the applicability of methods for predicting and assessing accidents at conflict sites in the transportation network and prospects for their development

Abstract

Accident rate is one of the most significant losses in road traffic, as it affects each of the road users. Therefore, in order to reduce its level, it is necessary to develop a modern method of accident prediction, which would allow to accurately estimate the number of accidents and the severity of consequences not only by experimental data and survey results, but also by modelling at the decision-making stage. Such a method is the method of conflict situations, but to date it is characterized by low accuracy of prediction, not applicable for practical use, as well as the severity of obtaining (collection and analysis) of raw data and lack of automation of the decision-making process to select the optimal measure by the method of conflict situations. In the article the analysis of existing methods of accident forecasting, as sessment of their applicability at different stages of decision-making, as well as further prospects and directions of improvement of the method of conflict situations for the purpose of its applicability in the practice of traffic management for the purpose of optimizing its options and modes of regulation are determined.