A hybrid evolutionary approach for set packing problem

Abstract

In this paper, we present a hybrid approach comprising an evolutionary algorithm with guided mutation (EA/G) and a local search to solve the set packing problem (SPP). EA/G is a recently proposed evolutionary algorithm that can be considered as a cross between genetic algorithms (GAs) and estimation of distribution algorithms (EDAs) and that tries to overcome the shortcomings of both. Guided mutation in EA/G generates offsprings through a probability model based on a combination of global statistical information and location information of the solutions found so far. We have compared our approach with the state-of-the-art approaches. Computational results show the effectiveness of our approach.