Characterizing Terminal Airspace Operational States and Detecting Airspace-Level Anomalies

Abstract

Global modernization efforts focus on increasing aviation system capacity and efficiency, while maintaining high levels of safety. To accomplish these objectives, new analysis methods are required that consider Air Traffic Management (ATM) system operations at both the flight level and the airspace level. With the expansion of ADS-B technology, open-source flight tracking data has become more readily available to enable larger-scale analyses of aircraft operations. Specifically, anomaly detection has been identified as being paramount. However, previous analyses of airspace-level operational states have not considered the observation of transitional (transitioning between two distinct airspace-level operational patterns) or anomalous operational states. Therefore, a method is proposed in which the time-series trajectory data of all aircraft operating within a terminal airspace during a specified time period is aggregated to generate a representation of the airspace-level operational states such that a recursive DBSCAN procedure to characterize airspace-level operational states as either nominal, transitional, or anomalous as well as to identify the distinct nominal operational patterns. This method is demonstrated on one year of ADS-B trajectory data for aircraft arriving at San Francisco International Airport (KSFO). Overall, visual inspection of results indicate the method’s promise in assisting ATM system operators, decision-makers, and planners in designing the implementation of new operational concepts.