Use of Fuzzy Neural Networks for a Short Term Forecasting of Traffic Flow Performance

The method for a short term forecasting of the traffic in the urban road network and of the average vehicle speed is suggested. The author's method is based on a regulatory approach to the calculation of the traffic capacity of the city road network. This method is completed with the methodology of forecasting the changes in the hourly traffic intensity. As the mathematical tool for the implementation of the forecasting methodology, the fuzzy neural networks are taken. It is suggested to make the forecast of short-term traffic intensity taking into account time of day, day of the week and season. On the basis of the data on the traffic capacity, the authors provide the relationships of an average speed change. The example of the calculation of transport flow performance is made in one of the motorways in the city of Volgograd.