Optimization-based training in ATM

Abstract

Air Traffic Management is responsible for guiding airplanes as efficiently and safely as possible at and between airports. A team of air traffic controllers is required to make good decisions at all times, even under high stress. The complexity of their tasks requires frequent and high-quality training to ensure constant high performance of the team. In this paper, we present work in progress on a novel training tool based on the Optimization-based Virtual Instructor. The tool we propose combines mathematical optimization with visualization, and is expected to improve the training quality while reducing the training cost. We discuss the new Virtual Instructor concept and introduce the necessary state-of-the-art advances needed for both visualization and mathematical optimization to make it work. Two early-stage visualization prototypes are presented. The paper concludes with a possible way forward in the development of the Virtual Instructor.