Article: Resource collaboration-aware service composition optimisation for cloud manufacturing Journal: International Journal of Data Mining and Bioinformatics (IJDMB) 2026 Vol.30 No.7 pp.84 - 107 Abstract: Within cloud manufacturing environments, service composition decisions are affected not only by service functionality and performance, but also by state changes and interactions among underlying manufacturing resources. However, implicit and dynamic resource collaboration relationships are difficult to characterise and incorporate into service composition, which may cause collaboration misalignment during scheme execution. To address this issue, a resource-collaboration-aware cloud manufacturing service composition optimisation method is proposed. Service associations induced by resource interactions are modelled as a service collaboration graph, and a graph attention network is used to learn collaboration embeddings that capture collaboration strength and structural heterogeneity. The embeddings are further mapped into a computable collaboration-degree metric and integrated with service quality and robustness in a multi-objective optimisation model. An improved GAT-NSGA-II algorithm is then designed to guide composition optimisation. Simulation results show that the proposed method improves the collaboration consistency and execution stability of service composition schemes. Inderscience Publishers - linking academia, business and industry through research

Title: Resource collaboration-aware service composition optimisation for cloud manufacturing

Authors: Jianjia He; Jiaqi Zhou; Xi Cheng; Zhuoyu Zhang

Addresses: Business School, University of Shanghai for Science and Technology, Shanghai, 200093, China; School of Intelligent Emergency Management, University of Shanghai for Science and Technology, Shanghai, 200093, China; Center for Super Networks Research (China), University of Shanghai for Science and Technology, Shanghai, 200093, China ' Business School, University of Shanghai for Science and Technology, Shanghai, 200093, China ' Business School, University of Shanghai for Science and Technology, Shanghai, 200093, China ' Business School, University of Shanghai for Science and Technology, Shanghai, 200093, China

Abstract: Within cloud manufacturing environments, service composition decisions are affected not only by service functionality and performance, but also by state changes and interactions among underlying manufacturing resources. However, implicit and dynamic resource collaboration relationships are difficult to characterise and incorporate into service composition, which may cause collaboration misalignment during scheme execution. To address this issue, a resource-collaboration-aware cloud manufacturing service composition optimisation method is proposed. Service associations induced by resource interactions are modelled as a service collaboration graph, and a graph attention network is used to learn collaboration embeddings that capture collaboration strength and structural heterogeneity. The embeddings are further mapped into a computable collaboration-degree metric and integrated with service quality and robustness in a multi-objective optimisation model. An improved GAT-NSGA-II algorithm is then designed to guide composition optimisation. Simulation results show that the proposed method improves the collaboration consistency and execution stability of service composition schemes.

Keywords: resource collaboration awareness; service composition optimisation; SCO; graph attention network; cloud manufacturing.

DOI: 10.1504/IJDMB.2026.154756

International Journal of Data Mining and Bioinformatics, 2026 Vol.30 No.7, pp.84 - 107

Received: 27 Jan 2026
Accepted: 06 May 2026
Published online: 13 Jul 2026 *

Title: Resource collaboration-aware service composition optimisation for cloud manufacturing

Keep up-to-date