| With the rapid development of the global civil aviation industry,the problem of flight delay is attracting more public attention.It is of great significance and value to study the evolution mechanism of air transport network,and to find the characteristics of flight delay propagation,which may lead to effective policies of delay controlling..In this paper,an evolving network model was improved based on the empirical analysis of air transportation network,as a result a preferential attachment mechanism of network growth was found and the preferential probability which related to route traffic was proposed.Simulation show that with the network expanding,the weight of each edge in network varies with the connecting of new nodes,and the reality characteristics of the constructed network such as the evolving quantity of airports as well as network structures were verified to be capable in applying to the air network.Furthermore,based on the evolving network,dynamic models of flight delay are improved by embedding network growth mechanism and route traffic growing.The propagation rule of flight delay and the probability of delay propagation are redefined on the basis of the epidemic model.Through the regression model,the delay propagation probabilities of different routes are analyzed,which related to factors that effect flight operations.An airport susceptible-infected-recovered(ASIR)model and a demand-driven model based on flight operation network(FSIS model,i.e.flight susceptible-infected-susceptible model)are established respectively.The simulation results show that route traffic,types of airports where delays are originated,and the level of airport turnaround services influence the propagation of delays in scale-free air transport networks.In addition,changes of network structure can also influence the delay propagation.The findings can help to improve airline and airport planning in order to achieve a higher level of resilience towards primary and propagated delays. |
