Effective traffic flow prediction is an essential component of any proactive traffic control system and one of the pillars of an advanced traffic management system. Hence, the main objective of this study is to develop a general and flexible prediction model that simulates the freeway system as it responds to traffic state transition. To this end, a spatial–temporal-based model without the prior need for off-line parameter calibration is proposed for traffic flow prediction. The model predicts flow rates at a given freeway network based on contributions from their recent temporal profiles as well as the spatial contributions from neighbouring sites. An in-depth investigation of the variables pertinent to traffic flow prediction is conducted to examine the extent of the temporal and spatial contributions and their effects on prediction accuracy. The predictions of the proposed models have been compared with the field-measured flow rates, and show that the results predicted by the proposed model are quite close to the measurement values for all the investigated sites. The benefit from full-spatial contribution as opposed to no contribution at all (ARIMA model) is evident in all investigated sites, and the accuracy can be improved by 5% as an average. Compared with the linear regression model with only spatial contributions, the average prediction error of the proposed model is 9% lower. Through comparison, the result shows that the temporal input factor provides more accurate information to the flow rate prediction, but it can hardly track the flow rate drops during the congestion. During the flow rate drops, the spatial input factor would secure the prediction accuracy.
A spatial–temporal-based state space approach for freeway network traffic flow modelling and prediction
Transportmetrica A: Transport Science ; 11 , 7 ; 547-560
2015-08-09
14 pages
Article (Journal)
Electronic Resource
English
Short-Term Traffic Prediction considering Spatial-Temporal Characteristics of Freeway Flow
DOAJ | 2021
|Bayesian Network for Freeway Traffic State Prediction
Transportation Research Record | 2018
|Short-Term Freeway Traffic Flow Prediction: Bayesian Combined Neural Network Approach
British Library Online Contents | 2006
|