This paper develops a KNN Model weighted by the importance of characteristic variables of the Random Forest, to predict the travel time between two adjacent signalized intersections. The importance of each characteristic variable is calculated by the Gini coefficient evaluation index based on the Random Forest model, and is weighted into the KNN model to predict the travel time. In the case study, the density and ifmpact of traffic lights are selected as characteristic variables due to their close relationship with travel time. The travel times are clustered by the DBSCAN algorithm to distinguish the number of stops affected by the traffic lights. Experimental results demonstrate that the proposed model provides an effective approach for urban travel time prediction and outperforms the considered competing methods. Combined with the Dijkstra’s Algorithm, the proposed model is applied to the road network to find the shortest travel time path.
Data-Driven Approach for Travel Time Prediction on Urban Road Sections and Its Application
20th COTA International Conference of Transportation Professionals ; 2020 ; Xi’an, China (Conference Cancelled)
CICTP 2020 ; 60-73
2020-12-09
Conference paper
Electronic Resource
English
Travel Time Prediction in Urban Road Networks
British Library Conference Proceedings | 1997
|Travel Time Prediction for Urban Arterial Road
British Library Conference Proceedings | 2003
|Travel time prediction for urban arterial road
IEEE | 2003
|Urban Road Travel Time Prediction Based on Taxi GPS Data
ASCE | 2013
|