Aircraft landing planning under uncertain conditions

Aircraft Landing Planning is challenging because the inherently limited capacity of airport runways causes bottlenecks. This type of planning involves different stakeholders (e.g., airlines, air traffic services providers, airport authorities, and passengers) and faces various uncertainties (e.g., take-off time variability, or wind speeds). This study, conducted in collaboration with the European Organization for the Safety of Air Navigation (EUROCONTROL), proposes a mathematical formulation of the problem and a simulation framework that accounts for uncertainties. We also propose different solution methods: a descent and a tabu search, as well as a mechanism for guiding restarts, to diversify the search process. These methods provide, in our simulated environment, more effective and stable solutions than the popular first-come-first-served practice regarding three objective functions (namely, delay, fuel, and landing sequence stability), which are considered lexicographically. Indeed, the average delays and fuel costs are reduced by 50% and 10%, respectively, at the cost of a small number of landing-sequence modifications, as each flight is repositioned an average of 0.5 times. Moreover, the computations can be performed quickly, which is crucial because re-optimization needs to be done online when flight information is updated.