This paper addresses the challenge of service composition in multi-cloud environments, where services are composed of components distributed across multiple edge clouds to meet diverse user requirements. The authors propose an improved ant colony optimization (ACO) algorithm to optimize the quality of service (QoS) parameters such as latency and response time while ensuring service stability. The proposed algorithm introduces a multi-pheromone mechanism to prioritize specific QoS attributes and a mutation operation inspired by genetic algorithms to avoid local optima and enhance convergence speed. The simulation results demonstrate that the proposed algorithm outperforms traditional ACO algorithms and other existing methods in terms of fitness, stability, and QoS parameters. The multi-pheromone mechanism effectively optimizes critical QoS parameters like latency, while the mutation operation improves the algorithm's performance and robustness.This paper addresses the challenge of service composition in multi-cloud environments, where services are composed of components distributed across multiple edge clouds to meet diverse user requirements. The authors propose an improved ant colony optimization (ACO) algorithm to optimize the quality of service (QoS) parameters such as latency and response time while ensuring service stability. The proposed algorithm introduces a multi-pheromone mechanism to prioritize specific QoS attributes and a mutation operation inspired by genetic algorithms to avoid local optima and enhance convergence speed. The simulation results demonstrate that the proposed algorithm outperforms traditional ACO algorithms and other existing methods in terms of fitness, stability, and QoS parameters. The multi-pheromone mechanism effectively optimizes critical QoS parameters like latency, while the mutation operation improves the algorithm's performance and robustness.