DYNAMIC GAME-BASED OPTIMISATION OF CLOUD RESOURCE SCHEDULING IN MACAU’S LOCAL AVIATION SECTOR

Efficient and fair scheduling has become increasingly critical in regional civil aviation systems, particularly in congested airspaces such as Macau. Existing scheduling approaches often overlook the strategic behaviour of stakeholders and fail to incorporate energy efficiency or real-time constraints. To address these gaps, this paper proposes a dynamic game-theoretic cloud scheduling model tailored for Macau’s civil aviation environment. The model captures the interactions among air traffic controllers (ATC), airport operation centres (AOC), and airline operators (AO) in a Stackelberg framework. A multi-objective optimisation algorithm based on a genetically-modified particle swarm optimisation (GMOPSO) is employed to balance flight delay, energy consumption, and fairness in cloud task scheduling. Simulation experiments using 2023 operational data from Macau International Airport show that our approach reduces average task completion time by 16.7%, improves Virtual Machine (VM) utilisation by 15%, and significantly enhances stakeholder fairness compared to conventional scheduling strategies.