Cloud service composition of collaborative manufacturing in main manufacturer-suppliers mode for aviation equipment

• Published in: Robotics and computer-integrated manufacturing Vol. 84; p. 102603
• Main Authors: Yin, Yongcheng, Yang, Bo, Wang, Shilong, Li, Shaobo, Fu, Guang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2023
• Subjects: Aviation equipment manufacturing; Carnivorous plant algorithm (CPA); Cloud platform; Main manufacturer-suppliers; Service composition; Carnivorous plant algorithm (CPA); Service composition; Main manufacturer-suppliers; Cloud platform; Aviation equipment manufacturing
• ISSN: 0736-5845; 1879-2537
• DOI: 10.1016/j.rcim.2023.102603

•The MSCM-SC model is firstly proposed for SC of aviation manufacturing.•The main characteristics of the M-S mode are fully considered in the MSCM-SC model.•ECPA is designed based on three improvement strategies for solving MSCM-SC problems.•Comparisons show that ECPA outperforms commonly used state-of-the-art algorithms.•The MSCM-SC model is validated by a case study of aviation manufacturing. Manufacturing of aviation equipment is characterized by long R&D cycles, multi-agent collaborations, and high technical difficulties. At present, the main manufacturer-suppliers (M-S) mode is mainly used for the supplier cooperative management of the aviation industry. To realize the efficient collaboration between the main manufacturer and the suppliers in terms of resources, businesses and processes, cloud platforms are gradually applied to the collaborative manufacturing of complex products. In the M-S mode, the manufacturing processes of aviation equipment are characterized by parallel design, multi-agent collaboration, hierarchical management of suppliers and early supplier involvement, which are seldom considered in the existing service composition (SC) methods in the cloud platform. Therefore, a SC model for aviation equipment collaborative manufacturing in the M-S mode (MSCM-SC) is proposed in this paper, in which the subtask correlation and multi-service collaboration are considered. Concurrently, a balance coefficient and a effect factor are put forward to modify the evaluation criteria of service resources, so the calculation method of the objective function can be established. To solve the above model accurately, a multi-layer code is designed and an enhanced carnivorous plant algorithm (ECPA) is proposed. In ECPA, a multi-population strategy and a Cauchy mutation perturbation strategy are developed to improve its exploration ability. Moreover, a variable step size attraction strategy is proposed to improve the exploitation ability of ECPA. Finally, comparative experiments are carried out to verify the effectiveness of ECPA and the MSCM-SC model in this paper. The superiority of ECPA is proved by comparing with several state-of-the-art algorithms by solving 22 classical benchmark test functions, while the effectiveness of the MSCM-SC model is verified by comparison with the traditional SC (T-SC) model on 9 aviation equipment collaborative manufacturing cases with different scales.