Supply-Demand Mutual Selection Mechanism In Cloud Manufacturing Service Intelligent Allocation Mode

The new generation of industrial revolution led by advanced information technology is profoundly shaping global manufacturing patterns.As a new model of smart manufacturing for service-oriented architecture,Cloud Manufacturing is a typical representative of this process.The Cloud Manufacturing platform brings together a wide variety of manufacturing resources and manufacturing capabilities.Rational planning,organization and scheduling based on customer needs to achieve efficient and low-cost supply and demand configuration is a key factor for the operation performance of Cloud Manufacturing platform.This thesis studies the intelligent allocation mode of Cloud Manufacturing services and its supply and demand mutual selection operations,and uses fusion Soft Set theory to complete the design of relevant decision-making mechanisms.The main work includes:(1)In cloud manufacturing service supply and demand market system for existing studies,the role of platform operators is too powerful while the automation of supply and demand is limited.This thesis sorts out the framework of supply and demand operation of cloud manufacturing services,designs an intelligent configuration system for supply and demand based on the empowerment of participating participants,and constructs the top framework of the supply and demand mutual selection decision mechanism including the three core departments of service invitation evaluation,service composition optimization,and service migration decision.In addition,One-dimensional Hybrid Fuzzy Soft Set and its normalization is defined based on the complex and diverse information forms on the cloud manufacturing platform Strategies to achieve uniform pre-processing of decision data.(2)In order to realize the evaluation and selection of customers by suppliers of cloud manufacturing services,the service invitation evaluation mechanism is studied from a granular perspective.Evaluation indexes for service invitation are established,thus forming the basis of cognition and response to demands.N-Granularity Image,which is defined based on N-Soft Sets,and a following strategy of uniformity calibration and missing information filling are used to achieve effective granularity recognition.N-Soft Rough Fuzzy Linguistic Set is defined based on N-Soft Approximation Space to generate the alliance comprehensive evaluation and then risk attitude-aware group decision mechanism of service invitation evaluation for joint response is constructed.(3)A forward-looking service composition optimization mechanism is studied for realizing customer-oriented optimal configuration of cloud manufacturing services.Based on composition strategy considering service relevance,task accomplishment prospects of each service composition are expressed as the integration of service trust and response,and prospect prediction based strategy of service composition and selection is established.For decision-making method on the one hand,Level Soft Set performs filtering with service trust to achieve optimal scale control.On the other hand,Tolerance Level Soft Set strategy is implemented to make flexible and balanced decisions based on execution prospects,and volatility analysis improves the decision recognition performance.(4)In order to realize cloud manufacturing services migration with high efficiency and low consumption,which requires the participation of customers and interrupted services,the service migration decision mechanism under the concept of General Remanufacturing is studied.A dual evaluation index including execution and remanufacturing performance is put forward to form General Remanufacturing-Oriented service migration strategy.In this decision-making process,on the one hand,the demander filter response services based on execution prospects with Level Soft Set,so that service performance for migration is discriminated from forward-looking perspective.On the other hand,current supplier measure trusted remanufacturing performance with Information Entropy-based D-S reasoning method to achieve efficient resource reuse.