A service composition and optimal selection method considering candidate service capabilities in cloud manufacturing

Service Composition and Optimal Selection (SCOS) has received increasing attention in cloud manufacturing, as it selects appropriate services from the candidate service set to complete the manufacturing task and satisfy the users. Existing research typically assumes that a subtask is performed by only one candidate service to facilitate modeling and solution. However, SCOS faces the challenge of successfully providing a service composition when there is insufficient capacity in the candidate service. To fill the gap, this paper proposes a service composition and an optimization selection model based on Multi-service Collaborative Manufacturing (SCOS-MSCM) to be able to select multiple candidate services for co-manufacturing to supplement the capacity. To address the SCOS-MSCM, an improved NSGA-II algorithm (I- NSGA-II) is presented, which contains several optimization strategies. Numerical experiments verify that I- NSGA-II outperforms the comparison algorithms. The case study shows that the service composition provided by SCOS-MSCM optimizes the minimum time by 37%, the minimum cost by 5.7%, and the maximum processing quality by 1.3% compared to the general SCOS.