Flight Delay and Delay Propagation Analysis Based on Bayesian Network

The research in the aspect of flight delay and delay propagation is relatively short in domestic. A flight-delay and delay propagation model is established based on Bayesian network (BN). After trained the network by real data with Expectation-Maximization (EM) arithmetic, the influences from the Arrival-delay and the Flight-cancellation on the departure-delay are analyzed under different all kinds of states. The conclusion of study indicates that the propagation from arrival-delays to the departure-delays is very obvious, especially when the degree of arrival-delay was MEDIUM or HEAVY, the influence of propagation couldnpsilat be eliminated. The flight-cancellation is one efficient method to deal with the influence. Flight-cancellation is involved to build our model. After trained, the model proved that Flight-cancellation could be adopted when delay propagation happened seriously which can weaken the propagation in a certain extent and the weakening degree is related to the number of canceled flights.