Research On Trust Mechanism In Task Scheduling Of Cloud Manufacturing

Manufacturing is an important industry for the economic pillar of China and even the entire world.Under a such background,the reform and optimazation of the manufacturing industry is imperative.Therefore,a grid-based manufacturing model was established.Under such a model,in order to facilitate manufacturing resources in the manufacturing industry,on-demand services are more convenient to handle and calculate.Based on cloud computing,one of the technical support cores,a new manufacturing model of cloud manufacturing was built up,making cloud manufacturing users more convenient and efficient when using network services.The key issue involved in cloud manufacturing is the scheduling problem.The core of cloud manufacturing scheduling is to assign reasonable resources to tasks according to different requirements of users(subjects).However,because cloud manufacturing platform is built in a complex,open,multi-node,and dynamically changing condition,there are many security issues in cloud manufacturing task scheduling,such as instability,lack of reputation,delayed access to interactive information,and difficulty in obtaining accurate information.What's more,obtaining reliable and complete information is hard too.Therefore,the research of trust mechanism in cloud manufacturing task scheduling is of very much necessity.To provide cloud manufacturing users a secure and trustworthy service platform,a trusted cloud platform in cloud manufacturing scheduling is firstly established in the paper,and a multi-agent-based trust management model is used to build a trusted environment.Firstly,a multi-agent-based trust management model framework was established.Comparing advantages and disadvantages of centralized management and distributed management mode,and analyzing the operating mechanism of the model sub-modules and the overall framework.After this,centralized distributed management approach is adopted.Combining the advantages of centralized and easy-to-efficient management with small distributed computing storage pressure,the platform features much advantages,and then trusted information is collected based on it.Through the research and analysis above and the construction of a trusted environment,a trust evaluation model based on interval-valued intuitionistic fuzzy sets and multi-attribute group decision making is presented to calculate the level of trust.Definiting the current version of interval intuitionistic fuzzy sets,studying interval intuitionistic fuzzy entropy,ordering interval-valued intuitionistic fuzzy numbers and distance formulas,determinating attribute weight coefficients,and determinating attribute weights and the ranking of grey correlation results.Through all these above,a complete trusted environment was constructed.Based on the characteristics of interval-valued intuitionistic fuzzy sets,combined with example analysis and experimental simulation,it's finally proved that this approch solves the problem that the evaluation results of existing trust assessment models are too subjective,the data type and scale are single,the control information is uncertain,the unsensitivity to malicious deception,and insufficient degree and coverage of entity information assessment.