Forecasting Model for the Scale of New-Built Airport´ Logistics Demand Based on the Back Propagation Artificial Neural Network

In order to forecast the scale of logistics demand for a new-built airport, economic indicators are used to forecast the scale of logistics demand and the measuring indicator of the scale of logistics demand is studied. The factor analysis and back propagation (BP) artificial neural network theory are applied to set up a model to forecast the scale of the logistics. The application of factor analysis is to reduce the number of indicators of the input layer in the BP artificial neural network, and to reduce complexity. Then a model is introduced to fit historical data of the scale of new-built airport logistics demand. Finally, a third-layer BP artificial neural network is constructed. This model is applied to predict the scale of the logistics demand in an example and the forecasting result shows that forecasting accuracy of the model is good. It also provides a new way of the logistics demand forecasting for a new-built airport.