Skip to Main content Skip to Navigation

Algorithmes génétiques hybrides en optimisation combinatoire

Abstract : In this thesis, we are interested in solving combinatorial problems with genetic algorithms. These algorithms are interested but they often need a lot of computations. Thus, we are focused on hybrid algorithms. Hybrids algorithms are algorithms built using several different methods. In our work, we build such algorithms by merging genetic algorithms and heuristics. There is two ways to do it. One is called the direct representation scheme and the other the indirect representation scheme. This two ways are studied by using them on three different problems: Multiprocessor tasks graph scheduling, VLSI cells placements, optimisation of cellular networks. For each of these problems, hybrids algorithms have shown their efficiency. For the probklem of optimisation of cellular networks, a new modelization have been made. This new approach allows us to do in one step the choice of emitters and the frequency's allocations.
Document type :
Complete list of metadatas
Contributor : Pascal Rebreyend <>
Submitted on : Thursday, November 10, 2005 - 2:34:18 PM
Last modification on : Thursday, November 21, 2019 - 2:15:24 AM
Long-term archiving on: : Friday, April 2, 2010 - 11:06:30 PM


  • HAL Id : tel-00010950, version 1



Pascal Rebreyend. Algorithmes génétiques hybrides en optimisation combinatoire. Autre [cs.OH]. Ecole normale supérieure de lyon - ENS LYON, 1999. Français. ⟨tel-00010950⟩



Record views


Files downloads