A hierarchical multi-agent coordinated decision-making for strategic slot allocation

Abstract

The Asia-Pacific (APAC) region is witnessing a surge in the demand for air traffic. Despite this growth, the absence of centralized Air Traffic Flow Management (ATFM) units presents a significant obstacle in executing tasks such as slot allocation to ensure demand and capacity balancing. This paper proposes a hierarchical multi-agent framework for automated demand and capacity balancing, resulting in slot allocation within the ATFM operations. Aligned with the multi-nodal ATFM (vATFM) concept, our proposal offers a coordinated approach to decision-making, aiming to streamline flight scheduling procedures. A heuristic algorithm is implemented based on the proposed framework to solve the demand-capacity balancing problem, using independent parallel decision-making to facilitate fairness among agents. Compared to a conventional peer-to-peer multi-agent design, the hierarchical approach results up to 100\% better in a typical traffic scenario of the APAC region and up to 400\% better in a doubled traffic scenario. Our findings demonstrate the superiority of the hierarchical multi-agent framework over traditional frameworks, offer a promising approach to address the strategic slot allocation challenges in the ATFM operations.