Prediction of flight delay reason based on time node data of transit flight

Flight delays are the bottleneck restricting the service quality of the civil aviation industry. Only by understanding the mechanism of flight delays can we accurately develop countermeasures. This paper is oriented to ground service phase of transit flights wherein flight delays frequently occur. The time node data of flight ground service phase was collected and thoroughly cleaned. A variety of classification models are trained on the dataset after multi-source fusion to predict different reasons of flight delay. The results show that the reasons of delay can be effectively mined by using the time node data of transit flights. The research ideas and methods put forward in this paper can be used as a reference when the official judges and counts flight delay reasons every year.