Research On The Predictability Of The Flight Stopover Time

Flight delay has become a common phenomenon,there have been a lot of researches about how to avoid flight delays before.It is also very important to decrease the loss under the condition of having had delays.As for this issue,the reasonable arrangement of the stopover time can effectively reduce the loss caused by delay.In order to accurately predict the stopover time,first of all,it is necessary to evaluate the predictability of stopover time,and then predict the time of transit.According to these requirements,this paper analyzes the influencing factors of stopover time,and takes the Pudong airport as an example to predict the stopover time and evaluate its predictability.In this paper,the blocking time and off-blocking time of each flight were extracted from the CDM data of Pudong airport,and the time difference between the two was calculated and taken as the initial data of transit time.After that,the paper analyzed the distribution pattern of transit time from four aspects,including the flight type,the arrival time,the airport flying and landing flow and the number of flights in apron.The regression equation was used to analyze the correlation between the number of flights in apron and the stopover time of flight,and it was found that there is a linear relationship between them and the matching was comparatively satisfying.Once the influencing factors of transit time were analyzed,this study utilized the percentage of variance as the method of calculating predictability so as to analyze and evaluate the predictability of transit time based on the influencing factors of stopover time.The calculation showed that the type,arrival time,the airport flying and landing flow as well as the number of flights in apron flights had impacts on the value of predictability.On the basis of this,the predictability of transit time for flights with different arrival time was evaluated by a cloud model.Finally,the paper analysised therelationship between the predictability error of prediction.according to the influencing factors of transit time,this paper establishedthe relevant Bayesian network model and predicted the different situation for various types of flight arriving at different time.It was found that the error of prediction on stopover time is consistent with the predictability value of transit time.