Different methods of traffic forecast based on real data

Different methods to forecast traffic are analysed and discussed. An elementary approach is to develop heuristics based on the statistical analysis of historical data. Daily traffic demand data from 350 inductive loops of the inner city of Duisburg over a period of 2 years served as input. The sets of data are organized into four basic classes and a matching process that assigns these sets into their class automatically is proposed. Furthermore, two models for short-term forecast are examined: the constant and the linear model. These are compared with a prediction based on heuristics. The results show that the constant model provides a good prediction for short horizons whereas the heuristics is better for longer times. The results can be improved with a model that combines the short- and long-term methods.