Search

The prediction of flight delays is heavily investigated in the last few decades. However, there is a relatively low level of study on Random flight point delays in these important problems. In this paper, we present an influence factor model of random flight points by series analysis on actual airline data, which is combined with BN (Bayesian Network) and GMM-EM(Gaussian mixture model-expectation maximization algorithm) algorithm. The creation of the initial parameters is based on the analysis for the continuous flights fly over the same flight point. The test data is offered by some Air Traffic Management Bureau. And the test result clearly demonstrates the value of Bayesian Network for analyzing the system-level effects arising from micro-level causes.

We design a Bayesian network by series analysis on actual airline data. The creation of the initial parameters is based on the analysis for the continuous flights fly over the same point. The framework is constructed and process of delay prediction is analyzed. The method is test in data sets which is offered by the company. And it clearly demonstrates the value of BN for analyzing the system-level effects arising from micro-level causes