David Gianazza

Ingénieur des Ponts, des Eaux et des Forêts

Associate professor at the MAIAA lab. of the french university of civil aviation (ENAC Toulouse, France)

Member of the APO team (Parallel Algorithms and Optimization) of the Toulouse Institute of Research in Computer Science (IRIT) since 2008

Work experience

Since 2012	Associate professor at the MAIAA of the french university of civil aviation (ENAC Toulouse, France).
2010 - 2011	Head of the Planning, Optimization, and Modeling team, at the R&D department of DSNA/DTI (Direction des Services de la Navigation Aérienne, Direction de la Technique et de l'Innovation), the technical service of the french Air Traffic Services provider.
2004 - 2010	Researcher in R&D services of DSNA (Direction des Services de la Navigation Aérienne), the french Air Traffic Services provider.
2000 - 2004	Research engineer and Ph.D. student at the Global Optimization Lab (ENAC/CENA Toulouse, France).
1996 - 1999	Deputy chief of the "Computer-assisted air traffic control tools" division, at the Technical Service (STNA) of the french Air Traffic Services provider (DSNA, France).
1991 - 1994	Ops engineer at the Brest Air Traffic Control Center, France.

For more details, see the pdf version of my CV.

Habilitation à Diriger des Recherches (post-doctoral degree similar to tenure), Institut National Polytechnique de Toulouse (INPT), 2004
Ph.D. in Computer Science, Institut National Polytechnique de Toulouse (INPT), 2004
DEA (M.Sc.) in Computer Science, INPT, 1996.
Engineer degree, ("Ingénieur de l'Aviation Civile"), Ecole Nationale de l'Aviation Civile (ENAC, the french university of civil aviation), 1996
Engineer degree ("Ingénieur des Études et de l'Exploitation de l'Aviation Civile"), ENAC, 1989.

My research interests focus on optimization and Machine Learning methods applied to Air Traffic Management problems.

Here are a few examples of the topics I've been working on:

ground-based trajectory prediction of climbing aircraft, using Machine Learning techniques or a point-mass model of the aircraft with tuned parameters (mass, thrust),
automatic resolution of conflicts between aircraft trajectories, using particle swarms or differential evolution,
wind approximation, with a least squares method applied on the observed aircraft trajectories,
Prediction of the air traffic controller workload, with a neural network tuned on historical data (past sector openings, recorded traffic),
Airspace configuration forecast, using tree search methods combined with a neural network,
3D-separation of aircraft trajectories, using a genetic algorithm hybridized with an A* algorithm.

For more details, see my research web page.

My publications are here. For memory, the publications of the former POM team members are there.

Reading, piano, badminton.

Last Update: