The ability to predict traffic conditions accurately is of paramount importance in effective management of a highway network. A more accurate prediction will allow for better allocation of resources, which may reduce experienced travel times. This paper introduces a composite approach to the already popular nonparametric regression used in predicting traffic conditions. The composite approach performs a nearest neighbor search for each loop detector station using only data that are in proximity to the detector's position on the roadway. This method accommodates every detector station individually to minimize the forecast error on the entire roadway. A case study using data from the Next Generation Simulation program recorded on US Highway 101 demonstrates that the composite approach significantly mitigates forecast error and performs the forecast in a reasonable amount of computational time. The case study also shows the ability of the composite approach to predict the onset and propagation of traffic shock waves.


    Access

    Download

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Composite Nearest Neighbor Nonparametric Regression to Improve Traffic Prediction


    Additional title:

    Transportation Research Record: Journal of the Transportation Research Board


    Contributors:


    Publication date :

    2007-01-01




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English





    Hybrid Multi-metric K-Nearest Neighbor Regression for Traffic Flow Prediction

    Hong, Haikun / Huang, Wenhao / Xing, Xingxing et al. | IEEE | 2015


    Time-Aware Multivariate Nearest Neighbor Regression Methods for Traffic Flow Prediction

    Dell'Acqua, Pietro / Bellotti, Francesco / Berta, Riccardo et al. | IEEE | 2015