| In recent years, with the rapid development of cloud computing, Internet, Internet of things and other technologies, has led to the transformation of many traditional industries. In order to accelerate and promote the upgrading of the manufacturing industry, China made the 2025 plan, cloud manufacturing is an important means of development of the manufacturing industry. For manufacturing enterprise supplier selection, compared with the traditional manufacturing, cloud manufacturing has a larger range of choices, which can match the individual needs in real time, so that the distribution of manufacturing resources are highly shared. Meanwhile, huge supply service resources bring new challenges for the supplier selection in cloud manufacturing environment. Therefore, the study on the service matching method has become one of the hot spots currently.Currently service matching almost for all services, but cannot adapt to domain objects and lack of research on supplier service matching; at the same time, the huge cloud manufacturing resources lead to a lot of similar demand. It must need QoS (quality of service) which is obtained according to the user’s personality and preferences to further select. Traditional matching mechanism is lack of selection and ranking of fuzzy QoS information. In addition, the traditional method recommend optimal service only, it is difficult to Meet the needs of independent choice. In view of the disadvantages of the traditional matching method, this thesis takes the supplier in the cloud manufacturing environment as the research object, combined with the ontology and fuzzy QoS, proposes three stage matching method, the specific work is as follows:Firstly, this thesis analyzed the characteristics of cloud manufacturing environment, characteristics of the supply chain in this environment, and the workflow and matching processes on cloud platforms, Based on the traditional semantic matching method, this paper proposed the supplier service matching technology framework in cloud platforms.Then, according to the characteristics of information asymmetry in cloud manufacturing, we designed service description language which based on service ontology, defined the structure of supplier service ontology on the basis of ontology modeling meta language study. A description method of supplier service is designed, it mainly consists of two aspects: functional description and QoS description, which provide semantic support for function matching and QoS matching in supplier service matching method.Then this thesis designed the three stage service matching method. Functional matching is mainly aimed at the resource concept of functional information, we adopted the matching algorithm which is based on semantic similarity. In addition to the consideration of influencing factors:the upper concept set, depth and density on the concept of ontology, but also added the attribute influencing factors, it avoids the problem that the concept can’t match to the ontology model because of different expressions, which can result in the problem of wrong or partial matching result. In terms of fuzzy QoS matching, we used FCM algorithm which based on triangular fuzzy number, optimized the initial cluster center, and introduced user preference. In order to meet the needs of choosing supplier independently, the output of the matching method is a certain amount of ranking results set which based on comprehensive matching degree.Finally, the effectiveness of the matching method through simulation experiments has been verified. This method can meet the demand of the reality better. Also through the comparative analysis we verified that the optimized algorithm improved the efficiency of service matching. This study puts forward a new idea about how to solve the problem of service matching in cloud manufacturing environment. |
PDF Full Text Request