Abstract: An ant colony optimization model for service selection was proposed on the basis of the summation analysis of the existing ant colony algorithms. The phoneme and heuristic information related with service selection were defined， and the six ant colony algorithms were applied to solve it. These algorithms were experimentally analyzed for different aspects and compared with the recently proposed service selection algorithm. The results showed that the performances of these algorithms were different and the ACS algorithm had better performance than the other compared algorithms.

Key words: ant colony algorithm, service selection, QoS, swarm intelligence, information phoneme