A Meta-Optimization Approach For Covering Problems In Facility Location

In this paper, we solve the Set Covering Problem with a meta-optimization approach. One of the most popular models among facility location models is the Set Covering Problem. The meta-level metaheuristic operates on solutions representing the parameters of other metaheuristic. This approach is applied to an Artificial Bee Colony metaheuristic that solves the non-unicost set covering. The Artificial Bee Colony algorithm is a recent swarm metaheuristic technique based on the intelligent foraging behavior of honey bees. This metaheuristic owns a parameter set with a great influence on the effectiveness of the search. These parameters are fine-tuned by a Genetic Algorithm, which trains the Artificial Bee Colony metaheuristic by using a portfolio of set covering problems. The experimental results show the effectiveness of our approach which produces very near optimal scores when solving set covering instances from the OR-Library.