University of Twente Student Theses


Long-term prediction of vehicle speeds

Siersema, A. (2020) Long-term prediction of vehicle speeds.

[img] PDF
Abstract:Although the method is only evaluated on freeways, the method can also be applied on urban roads. Besides, the method works for all circumstances by only use of vehicle speed data. However, the prediction accuracy can be improved if it is known that irregular situations occurred. In this study, the prediction is only adapted if it is known that an accident happened. It is advised to investigate the impact of other irregular situations on the traffic conditions for better prediction accuracies. The long-term prediction method gives more accurate predictions for all time horizons longer than 30 minutes compared to other prediction methods. Adding realtime measurements of the location itself improves the accuracy of the prediction method short-term. However, real-time measurements of road segments downstream the location, do not improve the accuracy. The lack of success is most probably caused by the fact that congestion propagation is dependent on the location. Therefore it is advised to investigate better the impact of location characteristics on the congestion propagation to improve the accuracy of the short-term prediction.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Programme:Construction Management and Engineering MSc (60337)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page