Research On The Theory And Method Of Cloud Manufacturing Service Evaluation

After several years of development,cloud manufacturing has developed into a new manufacturing model and intelligent manufacturing methods.It has been highly concerned by many research institutions both at home and abroad.Cloud manufacturing has become a new research hotspot in the field of manufacturing,and some important achievements have been achieved in theoretical research and practical application.As we can see from the entire cloud manufacturing service lifecycle activities,the cloud manufacturing platform brings together a large number of virtualized manufacturing service resources.Through the decomposition of cloud manufacturing service tasks,demand analysis,functional requirements and matching,the integrated manufacturing service resources are decomposed into a number of different independent service combinations,achieving “centralized use of decentralized resources,centralized resource decentralized services” and the purpose of high efficiency and low consumption.However,facing the gradual dynamics,complexity,diversity,and discreteness of cloud manufacturing services,higher requirements are placed on the service capabilities of cloud manufacturing services.Therefore,it is very important to conduct evaluation and research on cloud manufacturing services.This can not only enhance the competitiveness of different services,but also concentrate on superior resources,reduce production costs,and achieve complementary advantages among cloud manufacturing service resources.Firstly,on the basis of fuzzy theory,based on the Fuzzy Comprehensive Evaluation(FCE),the hierarchical comprehensive evaluation index system and the membership function are established by combining the qualitative and quantitative methods.Analytic hierarchy process(AHP)and Entropy Method are used to modify the weight of evaluation index,and the combined weight are obtained.Finally the comprehensive evaluation model with five levels of “excellence”,“good”,“medium”,“qualified” and “poor” are established.And through the verification of relevant examples,the established rating evaluation model can effectively evaluate and classify the cloud manufacturing services according to the different needs of service users.Secondly,using the Technique for Order Preference by Similarity to Ideal Solution(TOPSIS)and Grey Relating Analysis Theory(GRAT)the cloud manufacturing service provider preferred decision model is established.The model should be able to consider the actual distance of the sample in hyperspace,and to consider the relevance of the sample at the change of the shape,which makes the selection decision more reasonable and more believable.At the same time,based on the Improved Particle Swarm Optimization(IPSO)algorithm,the projection pursuit model is optimized,and the analysis of the sample index under high-dimensional non-normal and nonlinear data is realized.That model avoids the randomness of evaluation index weights from person to person,enhances objectivity,and has the characteristics of stable calculation results,less required parameters,and fast convergence.Through examples of simulation show that the model can scientifically,objectively and efficiently optimize the cloud manufacturing service provider,discover its constraints and deficiencies in time,and provide a reference for cloud manufacturing service provider's optimal decision-making.