Application of support vector machines in bus travel time prediction

Abstract

To obtain the real-time road traffic condition and dispatch and dispatch the vehicle, as well as plan the conveyance, a method based on support vector machines (SVM) was developed to predict bus travel time. Tests on No.23 bus in Dalian show that the prediction accuracy of SVM model is higher than that of history mean prediction (HMP) model and artificial neural network (ANN) model.