5 présentations

• 
  15h30 - 15h55

  Conception et analyse d'un système d'optimisation des trajectoires aériennes pour les avions

  • Wissem Maazoun, prés., Polytechnique Montréal
  • Antoine Saucier, Polytechnique Montréal
  • Francois Soumis, GERAD et Polytechnique
  • Steven Dufour, Polytechnique Montréal
  • Charles Pestieau, Polytechnique Montréal

  Nous présentons une méthode d’optimisation des trajectoires aériennes qui minimise tous les coûts associés au vol d’un avion. Nous minimisons le coût total du vol (coût du carburant, coût du temps, coût de survol des territoires) en tenant compte du type d’avion, des conditions météorologiques et des coûts de retard.

• 
  15h55 - 16h20

  Interactions between Operations and Planning Decisions in ATC

  • Thibault Lehouillier, prés., GERAD
  • Francois Soumis, GERAD et Polytechnique
  • Cyril Allignol, Université de Toulouse

  In a context of growing trafic, it is necessary to determine which decisions can save money for companies. We present a study of the cost evolution of operations decisions when changing planning policies. Results will enlighten subjects for optimization and therefore indicate priorities for our future research.

• 
  16h20 - 16h45

  EOSID Optimization with Genetic Algorithm

  • Bastien Talgorn, prés., Polytechnique Montréal

  Aircraft takeoff weight is calculated on the SID trajectory considering the case of an engine failure. In case of mountainous landscape, an alternative trajectory can be designed to reduce obstacle constraints. This trajectory, called EOSID (Engine Out SID), is optimized with a genetic algorithm to maximize the takeoff weight.

• 
  16h45 - 17h10

  Stochastic Programming with Recourse for Aircraft Separation under Uncertainty

  • Jeremy Omer, prés., GERAD

  The optimization of air traffic control aims at increasing the airspace capacity and flexibility of use. Meteorological forecast and trajectory predictions being inexact, uncertainty is an important issue. A stochastic program with recourse is thus developed to explicitly include errors on wind predictions and speed measures when separating conflicting aircraft.

• 
  17h10 - 17h35

  Frameword for the Aircraft Conflict Resolution Problem

  • Cyril Allignol, prés., Université de Toulouse

  We propose a new framework for the conflict resolution problem, separating the model from the solver, to be able to: enhance the model with as many refinements as necessary, and compare different resolution methods on the same data. We compare an Evolutionary Algorithm and Constraint Programming on this optimization problem.