To predict the development trend of traffic accidents and improve the prediction accuracy of macro indicators of traffic accidents, three direct indicators of road traffic accidents were taken as output variables and six macro indicators were taken as input variables to establish the NARX model. The model was trained and fit based on China’s data from 2001 to 2016, the modeling results were used to predict three direct indicators of traffic accidents in 2017 and 2018. To validate the performance of the NARX model, MLR model, BPNN model and GRNN model were also used as comparative benchmarks. The models were compared by selecting mean square error (MSE), Theil IC (TIC), mean absolute error (MAE) and mean absolute percentage error (MAPE) as the error analysis indexes. The results show the NARX model accuracy is better than the other three contrast models in the aspect of traffic accident macro index prediction.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Macro Prediction Model of Road Traffic Accident Based on NARX Neural Network


    Beteiligte:
    Wang, Jianjun (Autor:in) / Cai, Lu (Autor:in)

    Kongress:

    21st COTA International Conference of Transportation Professionals ; 2021 ; Xi’an, China


    Erschienen in:

    CICTP 2021 ; 1008-1019


    Erscheinungsdatum :

    2021-12-14




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Road Traffic Accident Prediction Based on BP Neural Network

    Xing, Yan / Song, Wen-hao / Liu, Wei-dong et al. | TIBKAT | 2023


    Road Traffic Accident Prediction Based on BP Neural Network

    Xing, Yan / Song, Wen-hao / Liu, Wei-dong et al. | Springer Verlag | 2022


    The Prediction Model of Macro-Road Traffic Accident Basing on Radial Basis Function

    Song, C.J. / Li, Q.Y. | British Library Conference Proceedings | 2011


    Traffic Accident Prediction on Provincial Road Using Fuzzy Neural Network

    Zhao, Fei / Fu, Liping / Zhong, Ming et al. | TIBKAT | 2020