A real time forecasting tool for dynamic travel time from clustered time series

Andres Ladino Lopez 1 Alain Y. Kibangou 1 Carlos Canudas de Wit 1 Hassen Fourati 1
1 NECS - Networked Controlled Systems
Inria Grenoble - Rhône-Alpes, GIPSA-DA - Département Automatique
Abstract : This paper addresses the problem of dynamic travel time (DT T) forecasting within highway traffic networks using speed measurements. Definitions, computational details and properties in the construction of DT T are provided. DT T is dynamically clustered using a K-means algorithm and then information on the level and the trend of the centroid of the clusters is used to devise a predictor computationally simple to be implemented. To take into account the lack of information in the cluster assignment for the new predicted values, a weighted average fusion based on a similarity measurement is proposed to combine the predictions of each model. The algorithm is deployed in a real time application and the performance is evaluated using real traffic data from the South Ring of the Grenoble city in France.
Complete list of metadatas

https://hal.inria.fr/hal-01521723
Contributor : Andres Ladino <>
Submitted on : Friday, May 12, 2017 - 4:45:15 PM
Last modification on : Thursday, August 22, 2019 - 11:32:02 AM
Long-term archiving on : Sunday, August 13, 2017 - 12:49:02 PM

File

TR-PartC-ToolForecastCluster.p...
Files produced by the author(s)

Identifiers

Collections

Citation

Andres Ladino Lopez, Alain Y. Kibangou, Carlos Canudas de Wit, Hassen Fourati. A real time forecasting tool for dynamic travel time from clustered time series. Transportation research. Part C, Emerging technologies, Elsevier, 2017, 80 (July), pp.216-238. ⟨10.1016/j.trc.2017.05.002⟩. ⟨hal-01521723⟩

Share

Metrics

Record views

775

Files downloads

550