Exploring delay propagation in air transport using temporal networks

In air transportation, delays are typically categorised as primary, caused by factors like weather or technical issues, and reactionary, which result from dependencies on previously delayed flights. This knock-on effect resembles a contagion process, where delays propagate from flight to flight, leading to large-scale disruptions. Reactionary delays are estimated to account for nearly half of all delays, yet their dynamics remain poorly understood and are difficult to predict due to the complex interplay of operational factors. In this talk, I will show how temporal network analysis can help describe and model this propagation process. I will also discuss how the structure of these networks can be used to improve delay prediction through node embedding techniques.