GPU-Based approaches for multiobjective local search algorithms. A case study: the flowshop scheduling problem
INRIA Dolphin Project / Opac LIFL CNRS, Villeneuve d’Ascq Cedex, France
Evolutionary Computation in Combinatorial Optimization, Lecture Notes in Computer Science, 2011, Volume 6622/2011, 155-166, 2011
@article{van2011gpu,
title={GPU-Based Approaches for Multiobjective Local Search Algorithms. A Case Study: The Flowshop Scheduling Problem},
author={Van Luong, T. and Melab, N. and Talbi, E.G.},
journal={Evolutionary Computation in Combinatorial Optimization},
pages={155–166},
year={2011},
publisher={Springer}
}
Multiobjective local search algorithms are efficient methods to solve complex problems in science and industry. Even if these heuristics allow to significantly reduce the computational time of the solution search space exploration, this latter cost remains exorbitant when very large problem instances are to be solved. As a result, the use of graphics processing units (GPU) has been recently revealed as an efficient way to accelerate the search process. This paper presents a new methodology to design and implement efficiently GPU-based multiobjective local search algorithms. The experimental results show that the approach is promising especially for large problem instances.
September 7, 2011 by hgpu
