Cloud manufacturing is an emerging manufacturing paradigm, with the issue of service composition being one of its most significant challenges. The current researches take the view that all manufacturing services are of equal importance. However, in fact, manufacturing services that owns scarce resources are more important than ordinary one, especially in the production of complex products. Hence, it is significant to meet the requirements of manufacturing services that own scarce resources first. Meanwhile, due to the increasing complexity of products, a high degree of synergy is required in service composition. Our study explores a new framework of social relationships from the perspective of synergy. A two-layer social network model is first proposed for resource allocation that takes synergy and priority into account. The upper layer of this model is occupied by the manufacturing services with scare resources. The lower layer is occupied by ordinary manufacturing services. All services are connected by social relationships. For the master-slave characteristic of the model, the two-layer network is embedded into bi-level programming. Then, an improved genetic algorithm is designed to solve the problem. Finally, the effectiveness of the model and algorithm are verified using a structural part of the Modern Ark 60 as an example.

云制造是新兴的制造范例，服务组合问题是其最重大的挑战之一。当前的研究认为所有制造服务都是同等重要的。但是，实际上，拥有稀缺资源的制造服务比普通服务更为重要，尤其是在生产复杂产品时。因此，重要的是首先满足拥有稀缺资源的制造服务的要求。同时，由于产品复杂性的增加，在服务组合中需要高度的协同作用。我们的研究从协同作用的角度探索了一种新的社会关系框架。首先提出了两层社交网络模型用于资源分配，该模型考虑了协同作用和优先级。该模型的上层被资源匮乏的制造服务所占据。下层被普通的制造服务所占据。所有服务均通过社会关系连接。对于模型的主从特性，将两层网络嵌入到双层编程中。然后，设计了一种改进的遗传算法来解决该问题。最后，以Modern Ark 60的结构部分为例，验证了模型和算法的有效性。设计了一种改进的遗传算法来解决该问题。最后，以Modern Ark 60的结构部分为例，验证了模型和算法的有效性。设计了一种改进的遗传算法来解决该问题。最后，以Modern Ark 60的结构部分为例，验证了模型和算法的有效性。