Modelling Traffic Flow on Motorways: a Hybrid Macroscopic Approach

The necessity for transportation research has been prompted by the exponential rise in vehicle use, research in the field has expanded and diversified accordingly. Despite substantial work having been carried out in Macroscopic modelling, the theory is far from complete [1]. The paper presents a hybrid model which combines the strengths of macroscopic models; the continuum method with forecasting models. The comparison presented herein is implemented using three different traffic flow forecasting methods: the artificial neural network, random walk and SARIMA. Each method is coupled with the Lighthill-Whitham and Richards continuum model [2,3]. The model is designed to function with speed and relative accuracy to allow for real-time applications. The results from numerical experiments using a case study (a junction in Hampshire, England) highlights the positive features of such a model in describing traffic dynamics.