Empirical exploration of air traffic and human dynamics in terminal airspaces

TL;DR

This study models and analyzes air traffic and human controller interactions in terminal airspaces using complex network tools, revealing fundamental traffic patterns, controller strategies, and chaotic behaviors to inform future automation.

Contribution

It introduces a multi-layer network framework and analytical metrics to empirically investigate traffic-human dynamics in terminal airspaces, filling a critical research gap.

Findings

Existence of fundamental and macroscopic fundamental diagrams at multiple levels.

Revealed adaptive meta-cognition strategies of air traffic controllers.

Identified chaos in conflict and human behavioral systems during congestion phases.

Abstract

Air traffic is widely known as a complex, task-critical techno-social system, with numerous interactions between airspace, procedures, aircraft and air traffic controllers. In order to develop and deploy high-level operational concepts and automation systems scientifically and effectively, it is essential to conduct an in-depth investigation on the intrinsic traffic-human dynamics and characteristics, which is not widely seen in the literature. To fill this gap, we propose a multi-layer network to model and analyze air traffic systems. A Route-based Airspace Network (RAN) and Flight Trajectory Network (FTN) encapsulate critical physical and operational characteristics; an Integrated Flow-Driven Network (IFDN) and Interrelated Conflict-Communication Network (ICCN) are formulated to represent air traffic flow transmissions and intervention from air traffic controllers, respectively. Furthermore, a set of analytical metrics including network variables, complex network attributes, controllers' cognitive complexity, and chaotic metrics are introduced and applied in a case study of Guangzhou terminal airspace. Empirical results show the existence of fundamental diagram and macroscopic fundamental diagram at the route, sector and terminal levels. Moreover, the dynamics and underlying mechanisms of "ATCOs-flow" interactions are revealed and interpreted by adaptive meta-cognition strategies based on network analysis of the ICCN. Finally, at the system level, chaos is identified in conflict system and human behavioral system when traffic switch to the semi-stable or congested phase. This study offers analytical tools for understanding the complex human-flow interactions at potentially a broad range of air traffic systems, and underpins future developments and automation of intelligent air traffic management systems.