Characteristics Analysis And Forecast Of Abnormal Departure Flight In Capital International Airport

With the civil aviation industry developing,the phenomenon of flights delay has been more and more severe.In this article,from a more microscopic point of view,the characteristics of the abnormal departure flights is studied,and base on which,establishes a forecast model of departure flight.The model is going to provide a reliable basis for the management of abnormal departure flight and emergency warning,and achieve the ultimate goal of improving the normal rate of departure flights.Firstly,delay characteristics of the abnormal departure flights is studied.The Beijing Capital Airport is selected as the research object,and based on the summary of existing flight normality statistical analysis method,a more detailed statistics method is put forward by using day,cycle,airlines,flight release direction and destination as new points of view,in order to re-examined the characteristics of the abnormal departure flights,and the correlation between influencing factors and the rate of abnormal departure flight.Specially,a normalized fairness indicator is established to measure the difference in departure flight delay.Secondly,based on the unbalanced distribution of departures flights,the existing statistical indexes of flight departure normality is studied comprehensively,then a index is established based on the directions of flight departure by using Clustering Methods,the feasibility of which is verified by time series forecast model.Lastly,according to delay characterizes of abnormal departure flights,such as the timing series characterize and so on,a time series forecasting model named Auto Regressive Moving Average Model(ARIMA)and the artificial neural network(BP)prediction model optimized by genetic algorithm is established respectively to predict flight delay.And the BP neural network model optimized by genetic algorithm showed high accuracy,as the result of prediction indicated.A statistical analysis method of abnormal departure flights from the point of the directions of flight departure is put forward,and a normalized index to measure the normality of departure flights and a index based on the statistics of the directions of flight departure is established.The indexes and the method could not only detail the flight normality statistical method,but also improve the accuracy of flight delay forecasting.