Research On Collaboration Technology Of Multiple Parts Value Chain Based On Cloud Service Platform

With the rapid development of economic globalization and the continuous transformation of information technology,the pattern of automobile industry is undergoing significant changes.Specialized division of labor and socialized cooperation of enterprise groups are deepening,and the links within and between industrial chains are becoming increasingly close,which has promoted the rapid development of automobile manufacturing industry towards scale and ecology.Nowadays,the competitive advantage of automobile manufacturing enterprises is not limited to their own capabilities,but depends more on the core competitiveness level of each division of labor in the industrial chain,the close degree of cooperation between enterprises and the upstream and downstream of the industrial chain,and the multi-chain coordination ability among different industrial chains.As a cloud service platform supporting multiple parts value chain collaboration,it breaks the constraints of traditional supply chain collaboration that can not be cross-regional and cross-chain based on the integration,optimization and redistribution of multi-chain superior resources.It closely links the enterprises in multi-chain,so as to build a cooperative system of mutual trust,resource sharing,win-win cooperation,symbiosis and complementarity.This dissertation is supported by the national key research and development program,the national science and technology support program and the national high technology research and development program(863 Program),etc.This dissertation studies the related theories,methods and technologies of multiple parts value chain collaboration based on cloud service platform.The main aspects are as follows:(1)Based on the analysis of supply chain theory,value chain theory and cluster supply chain theory,this dissertation pointed out that the existing theories are difficult to support multiple parts value chain collaboration.Then a multiple parts value chain collaboration system is proposed,and a theoretical model of multiple parts value chain collaboration is established,which considers cost reduction and value increment in multiple parts value chain.On this basis,a collaborative optimization technology of multiple parts value chain is studied,and a collaborative optimization game strategy of multiple parts value chain is proposed.It constructs a positive sum game mapping relationship among enterprises in multiple parts value chain,and finally provides effective decision-making for multiple parts value chain collaboration.(2)Based on the analysis of the traditional supply chain and cluster supply chain collaboration modes,this dissertation points out that the existing collaboration modes are difficult to support multiple parts value chain collaboration.Then a multiple parts value chain collaboration mode is explored,and a collaborative architecture of multiple parts value chain based on cloud platform is constructed,which establishes a five-tier architecture including multi-chain data resource integration,multi-chain data processing and analysis,multi-chain business collaborative modeling,multi-chain workflow scheduling optimization and multi-chain business collaborative execution.It realizes the integration,processing,analysis and utilization of information resources in multiple parts value chain,and finally provides strong technical support for the efficient collaboration of multiple parts value chain.(3)Based on the research of high-dimensional data analysis and mining technology for multiple parts value chain collaboration,this dissertation proposes a self-organizing input/output mapping neural network model based on BP neural network,genetic algorithm and particle swarm optimization.This model can remove the noise information which seriously interferes with regression analysis,and filter out the redundant data with low impact,so as to realize denoising and dimensionality reduction.On this basis,a key feature recognition model is proposed,which can analyze the core features of data by calculating the weight of data features and the threshold of key feature discrimination,so as to improve the quality of multi-chain data analysis and utilization,and make it more suitable for multiple parts value chain collaboration.(4)Based on the research of uncertain programming optimization technology for multiple parts value chain collaboration,this dissertation proposes an uncertain programming optimization system based on stochastic simulation,fuzzy inference,neural network and heuristic algorithm.Aiming at the randomness in multi-chain uncertain environments,a distribution feature mapping neural network is proposed to fit the probability distribution characteristics of arbitrary one-dimensional random variables.On this basis,the multiple random fuzzy features in complex multi-chain environments are fitted by combining the adaptive neuro-fuzzy inference system.Finally,Monte Carlo simulation technology and expectation model are used for heuristic optimization to find the optimal decision-making scheme for the uncertain programming problem of multiple parts value chain.It provides effective decision-making for multi-chain collaboration in complex uncertain environment.(5)Aiming at the workflow scheduling optimization requirement of multiple parts value chain based on cloud platform,a cloud workflow scheduling model oriented to QoS and cost-awareness is established.Meanwhile a coding rule dedicated to scheduling scheme is proposed to analyze the scheduling strategy of the proposed model at different stages.A genetic algorithm based on two segment coding of tasks order division is proposed.With tenant leases and virtual machine instance loads as constraints,the population iterative evolution through genetic recombination and mutation processes of two segments is developed,and both cloud service charge saving and cloud resources cost saving are achieved.(6)Based on the above research results,this dissertation gives the concrete realization and validation of multiple parts value chain collaboration technology based on cloud platform,high-dimensional data analysis and mining technology,uncertain programming optimization technology,cloud workflow scheduling technology,etc.