Analysis of Delay Recovery in Chinese Airports Network

Analysis of the flight delay recovery is crucial in the airports network operation. However, the analysis is very complex, as the effect of delays can be propagated through the network variously such that the recovery process depends on inherent characteristics of the network. In this paper, analyses of operation data are employed, and a simulation model is conducted to study the delay process with consideration of flight itineraries. Both the airports departure punctuality and the delay recovery indicator are employed to assess the influence of original delays considering different durations and start times. Our experiments are based on the Chinese historical flight data from July to September 2019. Results provide a comprehensive understanding of the delay recovery process with respect to flight itineraries. Surprisingly, the departure punctuality is non-monotonic process with multiple troughs in the considered recovery period, which implies that air traffic transport has the self-influence phenomenon. Sensitivity analysis implies that the phenomenon is highly correlated with characteristics of interruptions, which is an inherent quality of the structure of the flight schedule. Therefore, a well-organised flight schedule should be implemented in case delays relapse. This work can provide a broad prospect for the future flight schedule optimisation and improvement of airports network.