Identification and Characterization for Disruptions in the U.S. National Airspace System (NAS)

TL;DR

This study uses clustering and anomaly detection to identify and characterize operational disruptions in the U.S. NAS, revealing patterns, severity, and trends, including an increase post-COVID and seasonal variations.

Contribution

It introduces combined clustering and anomaly detection methods to classify and analyze NAS disruptions with geographical and temporal patterns.

Findings

Disruptions are most frequent in summer and winter.

Two main disruption clusters account for less than 3% of days.

Disruption frequency has increased since COVID-19.

Abstract

Disruptions in the National Airspace System (NAS) lead to significant losses to air traffic system participants and raise public concerns. We apply two methods, cluster analysis and anomaly detection models, to identify operational disruptions with geographical patterns in the NAS since 2010. We identify four types and twelve categories of days of operations, distinguished according to air traffic system operational performance and geographical patterns of disruptions. Two clusters--NAS Disruption and East Super Disruption, accounting for 0.8% and 1.2% of the days respectively, represent the most disrupted days of operations in U.S. air traffic system. Another 16.5% of days feature less severe but still significant disruptions focused on certain regions of the NAS, while on the remaining 81.5% of days the NAS operates relatively smoothly. Anomaly detection results show good agreement with cluster results and further distinguish days in the same cluster by severity of disruptions. Results show an increasing trend in frequency of disruptions especially post-COVID. Additionally, disruptions happen most frequently in the summer and winter.