Skip to Main content Skip to Navigation

Optimisation du trafic aérien dans de grands aéroports

Abstract : The air traffic growth induces congestion and flight delays both at the airports and in the sur-rounding airspaces. In fact, the airports are limited in terms of capacity and represent the major bot-tlenecks in the air traffic management system. Efficient planning and control are critical to enhancethe airport operation efficiency and to reduce flight delays. In prior research, several sub-problemsassociated with airport operations have already been discussed separately, such as runway schedul-ing, taxiway scheduling, terminal airspace management, etc. However, these operations are closelyrelated and can affect each other. This motivates the development of an integrated optimization ap-proach for managing air traffic at airport and in the surrounding airspace. In this thesis, we suggest atwo-level optimization approach which works on both the macroscopic and the microscopic levels.Following the prediction horizon of different problems, we consider first a long term horizon with anabstract network of airport and TMA. Then, we consider a shorter horizon with a detailed networkof airport components.In the first part of the thesis, we focus on the integrated optimization of airport operation prob-lem and terminal airspace management problem at a macroscopic level. The airside is modeled asan abstract network: terminal, taxi network, and runway are seen as specific resources with a definedmaximum capacity, and the TMA is modeled by a predefined route network structure. This level ofabstraction aims at identifying the airport congestion situations. We develop an optimization modelto minimize flight delays, resolve airspace conflicts, and mitigate airport congestions by controllingspeed, arrival and departure times, and assigned runway, while keeping various operational con-straints. An adapted simulated annealing (SA) metaheuristic combined with a time decompositionapproach is proposed to solve the corresponding problem. Computational experiments performed oncase studies of Paris Charles De-Gaulle airport show potential improvements on airport congestionmitigation and flight delay reduction.The second part of the thesis deals with the airport runway and taxiway scheduling problem ata microscopic level. In this part, we represent the airport (gate, taxiway, runway) with a detailedsurface node-link network, and we consider individual aircraft trajectories based on this graph. Weaim at resolving the ground conflicts among aircraft, assigning the pushback times, the taxi speedsand the positions (runway threshold or holding point) and the holding times. The optimizationmodel is designed to reduce runway queue length and minimize flight delays as well as taxi times with respect to safety concerns in surface traffic operations. A comparison with regard to baseline scenarios of the microscopic optimization benefits is presented for two major airports: Paris Charles De-Gaulle (CDG) airport and Charlotte Douglas International airport (CLT). Important gain in taxitime savings and runway queue length reduction are achieved, particularly at CLT since it is moreprone to congestion.The last part of the thesis focuses on a sub-problem of the microscopic level. It consists insequencing departures flights incorporating arrival crossings. In many hub airports with parallelrunways, arrivals have to cross departure runway to reach the taxiway. A better sequence of depar-tures taking into account arrival crossings could achieve less flight delay. Constraints for minimumrunway separations, flight time window restrictions, and holding queue capacity at runway thresholdare explicitly considered. We present two Integer Linear Programming (ILP) models and a SA algorithm. Case studies are conducted for the Southern pair of runways at CDG to reduce flight delays and obtain optimal sequence. Moreover, the solution quality and the computation time of different approaches are compared. The three proposed methods could significantly improve the solutions based on the simple first-come-first-served rule, and the SA is more suitable for implementation
Keywords : air traffic
Document type :
Complete list of metadatas
Contributor : Laurence Porte <>
Submitted on : Tuesday, November 17, 2020 - 3:20:58 PM
Last modification on : Tuesday, November 24, 2020 - 3:09:37 AM


Files produced by the author(s)


  • HAL Id : tel-03010033, version 1



Ji Ma. Optimisation du trafic aérien dans de grands aéroports. Optimisation et contrôle [math.OC]. Université Toulouse 3 - Paul Sabatier, 2019. Français. ⟨NNT : 2019TOU30052⟩. ⟨tel-03010033⟩



Record views


Files downloads