【作者】 张以文；

【导师】 倪志伟；

【作者基本信息】 合肥工业大学 ， 管理科学与工程， 2013， 博士

【摘要】 随着全球经济一体化趋势加剧,用户对产品不断提出了多样化和个性化的需求,企业面对不断变化的市场,竞争日趋激烈,为求得生存与发展必须具有高度的柔性和快速反应能力。为此,现代企业向组织结构简单化、扁平化方向发展,将知识、技术、资金、原材料、市场和管理等资源联合起来的虚拟企业组织形态应运而生。虚拟企业的核心理念是共享企业资源,提高制造系统的敏捷性、柔性和市场应变能力,从而达到多方共赢的目的。近年来,众多网络新技术不断涌现,为虚拟企业合作提供了新的应用背景和培植环境。云计算的出现,为企业(特别是中小企业)提供了美好的网络化应用新模式；云制造将进一步促进我国企业信息化建设向网络化、智能化和服务化方向发展,以实现动态智能的制造资源和制造能力等的共享与分配；制造云服务可以为企业提供动态按需获取的制造资源或制造能力服务,为实现全生命周期的任务协同提供支持。虚拟企业的本质特征敏捷性也是云制造的目标,因此,云制造为实现虚拟企业敏捷性提供了新的手段,云制造服务平台是高度开放的,企业能够通过云端以云服务的形式实现制造资源或制造能力的自由接入,降低了企业的准入门槛,可实现松耦合、紧耦合等不同形式的企业协作方式,为虚拟企业资源高度共享和协作提供了新的应用集成平台。本文以云制造模式为背景,对动态虚拟企业合作若干关键问题进行了分析研究。论文的主要工作如下：1)在分析云计算和各种先进制造模式概念、提出的背景及研究发展现状基础上,系统性地总结了敏捷制造、网络化制造和云制造等模式的内涵与特征,并进行了分析与比较。接着本文详细分析了动态虚拟企业的特征和内涵,为全文接下来的研究工作奠定了应用背景和理论基础。2)在分析动态虚拟企业信息化服务平台建设需求和典型的虚拟企业合作系统模型基础上,系统总结了云制造模式对动态虚拟企业合作模式产生的重要影响,提出一种云制造架构下的动态虚拟企业合作系统模型。基于该模型,进而提出了一种面向云服务的动态虚拟企业组建过程模型。同时,为实现云制造服务平台下云服务智能匹配问题,设计了基于本体技术的云服务选择算法,对云服务进行了功能和非功能属性本体建模,从语义角度提高云服务选择的准确性。3)针对云制造模式动态性和不确定性特征,基于分层的思想建立了虚拟企业合作评价指标体系,并提出了一种支持混合指标的主客观指标权重确定算法,为合作伙伴的合理选择奠定了基础。为解决面向云制造服务平台大规模合作伙伴选择问题,提出一种改进的自适应遗传伙伴选择算法,返回若干组优化的合作伙伴组合。基于以上结果,进一步提出阶段式的伙伴再选择算法,利用盟主约束、理想点聚类算法以及基于人工神经网络的交互式遗传算法不断降低问题求解复杂性,充分考虑云制造服务平台合作成员的整体相容性和敏捷性等,从而获得最优的伙伴组合。4)在总结了虚拟企业知识管理的关键问题基础上,提出一种云制造服务平台虚拟企业知识管理体系架构。文中系统地分析了云制造服务平台虚拟企业知识资源及知识服务的典型特征,给出了云制造服务平台虚拟企业知识管理的关键技术解决方案。最后,在具体“863”项目中加以实施,结果表明了文中所提理论和方法的正确性和可实现性。更多还原

【Abstract】 With the growing trend of global economic integration, users propose continuously the diversified and individualized requirements for market products. Therefore, in order to survive and develop, enterprises must have a high degree of flexibility and rapid response capability under the environment of protean market and furious competition. As a result, organizational structure of modern enterprise has turned toward simplification and flatness and the virtual enterprise form emerges and develops rapidly through combining knowledge, technology, fund, raw material, market and management, etc. The core concept of virtual enterprise is to share enterprise resources, improve the agility, flexibility of manufacturing system and market strain ability, and hence achieve the goal of multiparty win-win.In recent years, many new network technologies are emerging, which provide the new background and breeding environment for virtual enterprise cooperation. Specifically, cloud computing provides a new network applications method for enterprises (especially for medium-small sized enterprises). Moreover, a new service-oriented networked manufacturing model called cloud manufacturing was put forward, to further promote the network, intelligent and service development of enterprise information construction, and realize the dynamic intelligent sharing and distribution of manufacturing resources and manufacturing capabilities. In addition, manufacturing cloud service can provide dynamic on-demand access to manufacturing resources or manufacturing capacity for enterprises, so as to provide support for the full life cycle of task collaboration. Agility, which is the essential characteristics of virtual enterprise, is also the goal of cloud manufacturing. Therefore, cloud manufacturing also provides a new means for the realization of agility. Since cloud manufacturing service platform is highly open, enterprise can realize the free access to manufacturing resources and manufacturing capabilities through cloud&client with the form of cloud service. As a result, cloud manufacturing service reduces the admission threshold for enterprises, and can realize various forms of business cooperation such as the loose coupling and the tight coupling, thereby providing a new application integration platform for resources sharing and business collaboration of virtual enterprise.Taking Cloud manufacturing model as a background, this dissertation conducts the analysis and research on several key issues of dynamic virtual enterprise cooperation. The main works are summarized as follows:1) On the basis of analyzing the concepts, backgrounds, and current development situation of cloud computing and existing advanced manufacturing models, this dissertation summarizes systematically the characteristics and connotation of agile manufacturing, networkization manufacturing and cloud manufacturing and gives the results of analysis and comparison. Moreover, this dissertation analyzes the connotation and characteristics of dynamic virtual enterprise in details, thereby establishing the foundation of application background and theory for the following research.2) After analyzing demands of information service platform for dynamic virtual enterprise and typical cooperation system models of virtual enterprise, this dissertation summarizes systematically the important impacts of cloud manufacturing mode on the cooperation mode of dynamic virtual enterprise. Afterwards, a new cooperation system model is proposed for dynamic virtual enterprise under cloud manufacturing architecture. With the proposed model, a cloud service oriented building process model is put forward for dynamic virtual enterprise. Moreover, to realize cloud service intelligent matching of cloud manufacturing service platform, a new cloud service selection algorithm is designed based on ontology technique, where the ontology model of functional and non-functional attributes of cloud service is created so that the accuracy of cloud service selection is improved from the semantic perspective.3) Considering the dynamic and uncertain characteristics of cloud manufacturing, cooperation evaluation index system of virtual enterprise is established based on hierarchical thinking, and an algorithm supporting hybrid index is proposed to determine subjective and objective index weight, thereby establishing the foundation for reasonable selection of cooperation partners. In order to solve the large-scale cooperation partner selection in cloud manufacturing service platform, an improved AGA partner selection algorithm is proposed and some optimized combinations of cooperation partners are obtained. Based on the above results, a stagewise partner selection strategy is further proposed, where the complexity of problem-solving is reduced through using the leader constraint, ideal point clustering algorithm and artificial neural network-based interactive genetic algorithm. This strategy considers fully the overall compatibility and agility of cooperation partners in cloud manufacturing service platform, so as to obtain the optimal partner combination.4) On the basis of summarizing the key problems of knowledge management in virtual enterprise, this dissertation proposes a kind of knowledge management system architecture in cloud manufacturing service platform. Afterwards, this dissertation analyzes systematically the typical characteristics of knowledge resource and knowledge service of virtual enterprise in cloud manufacturing service platform, and gives the key technique solution to knowledge management in cloud manufacturing service platform. Finally, the results are obtained from the application in national863plan project which prove the accuracy and feasibility for the proposed theories and methods.更多还原