Detecting delay propagation in regional air transport systems using convergent cross mapping and complex network theory

It has been widely recognized that delay propagation is a widespread phenomenon in air transport systems. Regional air transport systems are vulnerable to severe systemic delay propagation because of connected resources and shared airspace. This study presents a two-stage analytical framework to detect delay propagation in regional air transport systems, which is critical to support industrial applications in air transport operations. In the first stage, we propose a delay causality without the strong and independent causality (SIC) assumption to represent the delay propagation effect. A convergent cross mapping method is employed to obtain the delay causality. In the second stage, we build a delay causality graph based on the none SIC assumption (DCG-non-SIC). A complex network analysis of DCG-non-SIC is presented to enhance the understanding of delay propagation patterns. This study carries out a real-world case study and provides some interesting results. Robustness checks are conducted, which verifies that the results derived in the analytical framework are solid. The results show that there is extensive delay causality between airport pairs, and a transitive trait of delay causality is observed in regional air transport systems. In addition, the importance of airports in delay propagation is not exactly in line with the hub class or the airport size.