Prediction Model of Passenger Disturbance Behavior in Flight Delay in Terminal

Flight delays disposal is always a tricky problem in civil aviation. The prediction model of passenger disturbance is of great significance to improve civil aviation service level and spot management capability in flight delays. Taking Shenzhen airport as an example, the 2016-2017 flight delay data is analyzed. Based on the BP neural network algorithm, a prediction model is set up by using influence factors which include depth of delay, scheduled flight departure date, current moment, passenger density of gates and ground service company. According to this mode, weight of influence factors is calculated by training neural network. The Prediction Model of passenger disturbance in flight delay is established. The results show that the model prediction accuracy is over 90%, when the number of learning times is 50000. The prediction model is effective, by which the civil aviation staff can make more accurate decisions in large-scale flight delays in civil aviation.