Proactive Big Data Analysis for Traffic Accident Prediction1

The article, the authors presented a system for proactive monitoring and forecasting of the risks of road traffic accidents, depending on the influence of external factors. To solve the problem, a method for analysis and predictive modeling of changes in the road transport infrastructure has been developed to predict the risks of occurrence and development of destructive events under the influence of external factors. The purpose is to determine, assess and predict the dynamics of changes in factors that affect the likelihood of the occurrence of risks of accidents, depending on the current situation on the monitored road sections. For predictive risk analysis, information on the parameters of negative events and possible influencing factors obtained from various sources is presented in the form of a spectrum of time series. Comparative analysis of time series of event parameters and factors allows us to identify the causes of incidents and the correlation between factors and events. As factors of influence, meteorological conditions, parameters of auto-mobile and pedestrian traffic on road sections, the state of the road surface, characteristics of road sections, etc. are investigated. The monitoring system is implemented using a multi-agent approach, which involves the use of software agents on photoradar complexes for photo and video registration of road events and mobile communications. Agents solve a number of tasks of collecting, parsing, consolidating, analyzing and visualizing big sensory data.