FABOP: Functional Airspace Blocks Optimized Process
La page internet de FABOP est aussi disponible en
français.
Summary
- Acronym: FABOP
- Project name: Functional airspace blocks optimized process
- Start date: 2004
- End date: -
Description
The main goal of the FABOP project is to enforce air safety. FABOP is a theoretical tool, but FABOP could be useful for concerned administrations who would draw a new European airspace.
The airspace is divided in elementary adjacent geographical sectors called controls sectors. Sectors are group together in airspace blocks. This airspace blocks are area where air traffic controllers are accredit to control airplanes. To enhance air traffic safety, we are studying about a new European airspace cut in blocks.
When air traffic controllers control one or more sectors, their workload depend on to three factors :
- Traffic supervision
- Conflicts resolution or potentials conflicts resolution
- Coordination between sectors
In the last case, controllers are stress when they have to manage significant traffic exchange between his sector and a sector in a block they do not know. Because controllers do not know as well traffic in other blocks than in theirs, they are more stress. Similarly, a good management of coordinations between blocks can reduce conflicts number and potentials conflicts number.
So, the FABOP project consist to cut the European airspace in blocks. And this project his particularly turn on cut in the upper space. Now, but this is only for the beginning, we want to cut blocks which respects this characteristics :
-
Airplane number which go through blocks must be minimized, and airplane numbers which are go through sectors in the same block must be maximized.
-
The sum of airplane which enter, go out and fly in a block must be on the same amount for each blocks.
To resolve this airspace block optimization problem, we use different metaheuristics :
-
A reference method : simulated annealing.
-
A new method who began to be know : ants colony.
-
A new method we have develop : fusion/fission
To limit the dimension of our study, we apply our cut on flights which go throw Europeans countries of the "country core area". The article "A theoretical approach to defining the European Core Area" define what is this European core area. Countries are : Spain, France, United Kingdom, Italy, Switzerland, Luxembourg, Belgium, Netherlands, Germany, Austria and Denmark.
Examples
In this section, we present you two optimization methods. The animations below (refresh the page if necessary) represent the algorithm process of our two optimization method. The big blue lines represent countries boundaries and small dark lines represent sectors boundaries. A set of sectors with the same color represent a block. Sometimes colors are closed, so be careful.
Each animations represent a cut of the airspace at a specific altitude. Of course we make the cut in three dimension. But we can not represent the cut in three dimension. The animation is a succession of steps of the optimization process find at the flight level 320. Flight level 320 is at the altitude 30 000 foots, nearly 10 000 meters.
A solution must be visually acceptable. But you can see an aptness of each solution find, in an other way than visually, with a diagram of the energy of each step on the up right corner. Be careful, the scale of the diagrams change between the two animations.
-
The first animation represent the process of a simulated annealing. The process begin with a randomly initialization. Each initial airspace block contain randomly sectors among all sectors. The animation is the result of 10 minutes of simulation. Best result are obtain if we wait more.
-
The second animation represent the process of fusion/fission. The process begin with a randomly initialization too. But the animation is the result of only 30 seconds of compute.
As you can see, the fusion/fission method we develop is much powerful for time computation and for results, than the simulated annealing.
We insist in the fact that cuts we present here are just illustratives examples. We do not have finish our work. For the moment, they can not be taken as a viable model for a possible airspace cut. If you want more information about the airspace cut, please contact Charles-Edmond Bichot.
Useful links
Main publications
- A tool to design Functional Airspace Blocks
Charles-Edmond Bichot, Nicolas Durand
7th USA/Europe Air Traffic Management Research and Developpment Seminar (2007/07/2)
- A new Method, the Fusion Fission, for the relaxed k-way graph partitioning problem, and comparisons with some Multilevel algorithms.
Charles-Edmond Bichot
Journal of Mathematical Modelling and Algorithms (JMMA), 6(3): 319-344, published online the 16th March 2007 (2007/03/16)
- Airspace block organization with metaheurisitics and partitioning packages
Charles-Edmond Bichot, Nicolas Durand
In proceedings of the 2nd International Conference on Research in Air Transportation (ICRAT) (2006/06/24)
- Metaheuristics versus spectral and multilevel methods applied on an Air Traffic Control problem
Charles-Edmond Bichot
In proceedings of the 12th IFAC Symposium on Information Control Problems in Manufacturing. (2006/05/18)
- A metaheuristic based on fusion and fission for partitioning problems
Charles-Edmond Bichot
In proceedings of the 20th IEEE International Parallel and Distributed Processing Symposium (IPDPS). (2006/04/25)
- Optimisation par fusion et fission. Application au problème du découpage aérien européen.
Charles-Edmond Bichot, Jean-Marc Alliot, Nicolas Durand, Pascal Brisset
Journal Européen des Systèmes Automatisés, September-October 2004, vol 38, pp 1141-1173 (2005/10/05)
- A theoretical approach to defining the European core area
Charles-Edmond Bichot, Jean-Marc Alliot
(2005/06/28)
- Optimisation par colonies de fourmis appliqué au découpage de l'espace aérien européen en zones de qualification.
Charles-Edmond Bichot, Jean-Marc Alliot
RIVF2005 (2005/02/22)
All the publications of the log are available
here.