Survey on distributed approaches to swarm intelligence for graph search problems

Abstract

In this paper a survey on existing distributed approaches to swarm intelligence approaches for graph search problems is presented. In particular we reviewed papers on Ant Colony Optimization(ACO) and Bee Colony Optimization (BCO). The comparison criteria that have been used are computational efficiency and speedup. The conclusion of this study is that the coarse grained master-slave model is the most studied. However, we found a large amount of papers sustaining that the island approach offers better solution quality. The added flexibility of the communication strategy between the islands makes this model preferred by the most recent papers.