A Machine Learning Based Approach for the Prediction of Road Traffic Flow on Urbanised Arterial Roads

Abstract

Urbanised arterial roads connect geographically important areas and are used for commuting and the movement of goods. Prediction of traffic flow on these roads is vital to aid in the mitigation of congestion. However, there is currently a lack of research in this area. In this work we have applied machine learning models to a real dataset for the prediction of road traffic congestion on urbanised arterial road. A comparative analysis was conducted on each machine learning model, examining the prediction accuracy and time-horizon sensitivity. Furthermore, we examined different input parameter settings (various classes of vehicles such as motorcycles, cars, vans, rigid goods lorries, articulated heavy goods vehicles (HGVs), and buses) to investigate how heterogeneous traffic flow can affect prediction. The experimental results show that the Artificial Neural Network Model outperforms other models at predicting short-term traffic flow on an urbanised arterial road based on the standard performance indicator: Root Mean Squared Error (RMSE). Additionally, it is found that different classes of vehicles can aid the improvement prediction.