The prediction accuracy of airport passenger throughput plays an important role in airport planning and construction capacity. In this paper, we take Lhasa Gonggar Airport as the research object, use adaptive airport strategic planning ideas, and introduce exogenous features that are highly related to airport development. Based on these important features, we apply a multi-feature deep neural network model to predict the monthly airport passenger throughput. Experiments show that our multi-feature model outperforms traditional time series forecasting models (i.e., Exponential smoothing, ARIMA) in forecasting accuracy by a large margin. Through the feature analysis, further policy recommendations are put forward to promote the economic and social development of Tibet.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Airport Passenger Throughput Forecasting by Using Multi-features Long Short-Term Memory Neural Network


    Contributors:
    Zhu, Runze (author) / Yang, Lu (author) / Jiang, Ying (author)


    Publication date :

    2021-10-20


    Size :

    1869450 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English