1 PhD position on Combining metaheuristics and exact methods for multi-objective optimization in France | Inria

The goal of the DOLPHIN (Discrete multiobjective Optimization for Large-scale Problems with Hybrid dIstributed techNiques) project team is the modeling and resolution of large multi-criteria combinatorial problems using parallel and distributed hybrid techniques. We are interested in algorithms using Pareto approaches which generate the whole Pareto set of a given multi-objective problem (MOP). For this purpose, the research actions can be resumed as follows.

The DOLPHIN team has a large expertise in designing multi-objective metaheuristics. For some important multi-objective problems, such as arc routing, flow-shop scheduling and unit commitments problems, exact methods can tackle efficiently small instances of those problems. Until now only few exact methods have been proposed to solve multi-objective problems. They are based either on branch-and-bound approaches or dynamic programming. However, those methods are limited to two objectives and are, most of the time, not able to be used on a complete large scale problem. Therefore, search subspaces have to be defined in order to be able to use exact methods. The hybridization allows to use the exploration capacity of metaheuristics, as well as the intensification ability of exact methods, which are able to find optimal solutions in a restricted search space.

The main goal of this thesis is to develop some original cooperation schemes to solve large scale multi-objective problems using some tight coupling of metaheuristics and exact methods. We will focus on well known problems such as arc routing, flow-shop scheduling and/or unit commitment problems.

• E-G. Talbi, « Metaheuristics : from design to implementation”, Wiley, 2009.
• E-G. Talbi, “Combining metaheuristics with mathematical programming, constraint programming and machine learning”, 4OR, 2013.