Optimization Method Of Cloud Manufacturing Service Composition Based On Service Cluster And Improved Artificial Bee Colony Algorithm

In the cloud manufacturing environment,manufacturing resources or manufacturing business are encapsulated as cloud manufacturing services and published in the cloud platform for users to use on demand.By invoking and integrating cloud manufacturing services,users can rapidly expand their manufacturing capabilities.For complex manufacturing requirements,it is necessary to adopt cloud manufacturing service composition to respond.With the emergence of various cloud manufacturing platforms,the number of cloud manufacturing services is also increasing rapidly,which provides users with mo re choices.However,the increase of the number of cloud manufacturing services also makes it more difficult to construct the cloud manufacturing service composition.The service composition consists of multiple nodes,and the increase of the number of cloud manufacturing services will make the composition instances grow exponentially,and the difficulty of finding the optimal process increases dramatically.At present,swarm intelligence algorithms are often used to solve composition optimization.However,the optimization quality,efficiency and stability of the existing methods need to be improved.In addition,the existing cloud manufacturing service composition optimization models are mainly constructed based on the Quality of service(QoS).They lack of consideration of the Quality of collaboration(Qo C)between cloud manufacturing services.The rationality of these composition optimization needs to be further improved.To address the above problems,this paper proposes a cloud manufacturing service composition optimization method based on service cluster and improved artificial bee colony algorithm.This method constructs a service response framework oriented to service clusters,generates candidate response sets for service composition based on service clusters,and constructs a composition optimization model considering QoS and Qo C.An ABC algorithm integrating multi-search strategy island model was designed to improve the quality and stability of optimization solution.The work and contributions of this paper are as follows:(1)A service response framework oriented to service clusters is constructed.An algorithm for finding the candidate service response set for cloud manufacturing service composition oriented to service clusters is designed.The service search space is reduced and the efficiency of generating candidate service response set is improved under the proposed service response framework.(2)A cloud manufacturing service composition optimization model combining QoS and Qo C is established.The Qo C decision factors are mined from three perspectives of service cooperation intention,migration cost and historical cooperation intensity.The quantitative method for above Qo C decision factors is presented.For different process patterns,a service composition optimization model integrating QoS and Qo C is established.Compared with the single Qos-oriented composition optimization model,the proposed model is more suitable for cloud manufacturing service composition optimization scenarios.(3)An artificial bee colony algorithm with multi-search strategy island model is proposed.The candidate solution space is divided into multiple islands,and different search strategies are used to optimize the sub-populations independently.The migration mechanism is improved to enhance the effectiveness of information exchange between sub-populations.The diversity of solutions is improved by increasing the current solution vector and search method,and the searchable solution vector area is expanded.Various search mechanisms can also avoid the algorithm falling into local optimum and improve the quality of service composition selection.Experimental results show the effectiveness of the proposed method.