ATHENA: A METHOD FOR SHORT-TERM INTER-URBAN MOTORWAY TRAFFIC FORECASTING

The need for accurate short-term traffic volume forecasts is well established. They are necessary in order to determine when extraordinary measures shall be needed in order to avoid traffic congestion and to minimize the duration of these measures. Here, a new short-term traffic volume forecasting method is presented, based on statistical analysis of traffic volume profiles. These profiles are consecutive values of the traffic volumes at a point divided by the total traffic volume for a given length of time. On the French toll motorway system, it is shown that the daily traffic volume profiles can be grouped into a small number of classes with members of the same class resembling each other more than members of different classes. Traffic forecasting is thus facilitated by forecasting for each class of traffic. A computer program (ATHENA) was developed and the method was tested during the summer of 1989 at three points on the French motorway system (Nimes, Orange and Beaune). The study shows conclusively that a traffic classification method, based on historical data, considerably improves accuracy of forecasting. (A) For the French abstract see IRRD 125463.