Disruption and recovery of the US domestic airline networks during the COVID-19 pandemic

TL;DR

This study analyzes the day-to-day dynamics of US domestic airline networks during COVID-19, revealing structural changes, variability among airline types, and factors influencing recovery patterns.

Contribution

It introduces a temporal network analysis framework to quantify airline network changes and recovery during the pandemic, highlighting differences among airline categories.

Findings

Network structure remained altered despite node and edge recovery.

Full-service carriers were less flexible and suffered higher revenue losses.

Regional and low-cost carriers adapted their networks more drastically.

Abstract

The COVID-19 pandemic has had serious impacts on the airline industry. Ensuring that aviation policies in emergent situations both guarantee network connectivity and maintain competition among airlines is crucial in these circumstances. To this end, we aimed to understand the network dynamics of individual airlines. In this study, we quantitatively reveal the day-to-day dynamics of these US domestic airline networks, comprising 17 airlines, from January 2019 to December 2021. Specifically, we applied a framework for analyzing temporal networks, in which the network structure changes over time. We found that, first, even though the number of nodes and edges returned to pre-pandemic levels around July 2021, the structure of the entire US domestic airline network remained altered. We also found that the network dynamics varied significantly from airline to airline. Full-service carriers were less flexible in changing their network structure and suffered higher revenue losses. On the contrary, most regional carriers completely shifted to a new structure, which may have contributed to reducing their revenue losses. Low-cost carriers were characterized by more pronounced differences between airlines and drastically changed their network structure immediately after the declaration of a national emergency. Finally, we also examined the recovery process and found that the flights connecting airports that are more central and share more common neighbors, and those connecting airports with larger numbers of connections tend to recover earlier for most of the airlines.