Flying safely by bilevel programming

Martina Cerulli (1, 2, 3) , Claudia d'Ambrosio (1, 2, 3) , Leo Liberti (1, 2, 3)

1 LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
2 CNRS - Centre National de la Recherche Scientifique
3 X - École polytechnique

Preventing aircraft from getting too close to each other is an essential element of safety of the air transportation industry, which becomes ever more important as the air traffic increases. The problem consists in enforcing a minimum distance threshold between flying aircraft, which naturally results in a bilevel formulation with a lower-level subproblem for each pair of aircraft. We propose two single-level reformulations, present a cut generation algorithm which directly solves the bilevel formulation and discuss comparative computational results.

Domaines: Informatique [cs]; Recherche opérationnelle [math.OC]

Martina Cerulli, Claudia d'Ambrosio, Leo Liberti. Flying safely by bilevel programming. Advances in Optimization and Decision Science for Society, Services and Enterprises, 2020, ⟨10.1007/978-3-030-34960-8_18⟩. ⟨hal-02869682⟩