Freeway travel time forecast using artifical neural networks with cluster method
Ying Lee · International Conference on Information Fusion · 2009
This paper develops a novel travel time forecasting model using artificial neural network with cluster method. The core logic of the model is based on a functional relation between real-time traffic data as the input variables and actual travel time data as the output variable. Cluster method is employed to reduce the data features with fewer input variables while still preserving the original traffic characteristics. The forecasted travel time is then obtained by plugging in real-time traffic data into the functional relation. Our results show that the mean absolute percentage errors of the predicted travel time are mostly less than 22%, indicating a good forecasting performance. The proposed travel time forecasting model has shed some light on the practical applications in the intelligent transportation systems context.