Air traffic management (ATM) is a complex process that includes objectives such as minimising delays, costs, and environmental impacts, with safety being a top priority. Growth in air traffic over the past few decades has increased the complexity of this process. In European airspace, ATM is more crucial because of the high traffic density. A key aspect of this management strategy is the allocation of flight levels in the planning phase to avoid conflicts before takeoff. Therefore, a model is required for the allocation of flight levels in the planning phase to eliminate all conflicts and meet objectives such as minimising cost and emissions. In the absence of conflicts, planning delays in takeoff are avoided and in-flight corrective manoeuvres (such as changes in altitude or speed during flights) are reduced, reducing the workload of the controller. This study proposed a method based on a metaheuristic variable neighbourhood search (VNS) strategy on a multistart framework. Various experiments with simulated instances demonstrated that this method improves the results obtained using commercial software. In addition, tests based on real data (both in Europe and Spain) showed satisfactory results in terms of economics and sustainability.
