Experimental feedback on bioobjective permutation scheduling problems solved with a population heuristic

Population heuristics present native abilities for solving optimization problems with multiple objectives. Using evolution and adaptation mechanisms, a population of individuals evolves in order to describe a good approximation of the efficient solutions. The resolution of the 1 | | (EC,- ,Tmaar) permutation scheduling problems, for which an exact algorithm is available, is investigated using a population heuristic based on genetic algorithms. The aim here is not to put in competition a heuristic method with an exact method but to give an experimental feedback on the resolution abilities of biobjective permutation scheduling problems with a population heuristic. The paper reports the aspects analyzed in this study : first, the pertinence of using genetic information into a population algorithm and second, a detailed multicriteria analysis of efficient solutions for this class of scheduling problems.