Correlation-aware QoS modeling and manufacturing cloud service composition

Abstract

Recently, cloud manufacturing has attracted much attention from both academic and industry communities. Manufacturing cloud service composition and optimization is critical to the optimal resources allocation in cloud manufacturing. Since there are many manufacturing cloud services available with similar functions but different quality of service (QoS), and with potential quality correlations among them, such correlations must to be considered for manufacturing cloud service composition. In this paper, a correlation-aware manufacturing cloud service description model is presented to characterize the QoS dependence of an individual service on other related services. Based on such a model, a service correlation mapping model is proposed for getting correlation QoS values among services automatically. In addition, an effective approach for the correlation-aware optimal service selection is proposed based on a genetic algorithm. A case study indicates that services composition of higher quality can be obtained when such correlations are considered. And the effectiveness and efficiency of the proposed approach are demonstrated via simulation studies.