Weathering the Storm: How Network Design Shapes Airline Resilience in Extreme Weather

Abstract

Extreme weather events pose significant challenges to the reliability and efficiency of commercial air travel in the United States. As climate change increases the frequency and intensity of these events, it is essential to understand how airline networks are able to manage their responses and which structural features improve operational resilience. This thesis investigates the impact of Hurricane Helene (September 2024) on U.S. airline operations by modeling airline networks as graphs and analyzing changes in flight activity and connectivity. Using historic flight data from Flightradar24, the study compares airline networks during a control period and during Hurricane Helene, quantifying disruptions through changes in flight volume and network structure. Airline performance is evaluated using both standard centrality metrics and measures such as percent change in number of flights and network distance, a measure of the dissimilarity edge structure between two networks. Additionally, a series of ridge regressions are performed across different scales to assess how airline-specific structural characteristics, including number of destinations, number of routes, number of hubs, low-cost status, and partnership status, relate to observed changes in performance. The findings reveal that network disruptions were uneven across airlines, and that structural features such as redundancy, connectivity, and decentralized operations contributed to greater performance. Airlines with more routes and without extensive partnerships tended to perform better during Hurricane Helene. The study also finds that the effects of the hurricane propagated beyond directly impacted areas, influencing distant parts of the network. The results highlight the importance of network design in disruption management and support the use of network science as a practical tool for understanding and improving airline robustness and resilience. This case study provides a foundation for integrating empirical data with theoretical models to inform strategic planning in aviation systems.