Travel Time Prediction Using Empirical Mode Decomposition and Gray Theory: Example of National Central University Bus in Taiwan

Travel time information is generally nonlinear and nonstationary in a dynamic environment, and therefore no consistent tendency can be easily observed. This research developed a novel approach that combined the empirical decomposition method for speed data analysis and gray theory for travel time pr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Transportation research record 2012-01, Vol.2324 (1), p.11-19
Hauptverfasser: Chen, Huey-Kuo, Wu, Che-Jung
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Travel time information is generally nonlinear and nonstationary in a dynamic environment, and therefore no consistent tendency can be easily observed. This research developed a novel approach that combined the empirical decomposition method for speed data analysis and gray theory for travel time prediction to predict the arrival time at each stop along a bus route. In addition, sensitivity analysis was performed for different numbers of stops. With an average prediction error of less than 3.5%, the experiments showed that the proposed prediction approach, which employed both historical and real-time speed data collected from the geographic positioning system, outperformed Chou's approach, which used only historical speed data. The proposed prediction method could be readily incorporated into a cell phone–based information retrieval system that indicated bus position en route as well as its arrival times at all stops.
ISSN:0361-1981
2169-4052
DOI:10.3141/2324-02