节点文献

支持流程推荐的工作流协同建模工具的设计研究

【作者】 吴颜

【导师】 沈洁;

【作者基本信息】 扬州大学 , 计算机应用技术, 2007, 硕士

【摘要】 工作流技术是近年来计算机应用领域中迅速发展的一项新技术,作为工作流技术的实现,工作流管理系统已广泛应用与众多行业中。工作流系统成功的关键在于输入的工作流的质量。因此,工作流建模成为当今工作流研究的一个热点问题。在传统的工作流管理系统中建模工具一般只支持单用户工作,工作流模型由专业建模人员来建立。随着工作流管理系统应用领域和应用规模的不断扩大,其所管理的企业业务流程已不可能只局限在企业中的一个小部门内部,而是一些跨部门跨企业的业务流程,这些流程只靠单个专业建模人员已经很难完成其工作流模型的建立,它需要不同人员的分工合作,充分利用各自的专业知识共同来完成其工作流模型的建立。同时,工作流过程通常具有极高的复杂性,而大量的模型信息却不能够得到有效的利用,建模人员在建模过程中缺乏参考,产生“信息孤岛”。这就需要一种针对工作流模型的挖掘分析方法,使得工作流模型信息得到合理利用,降低建模难度。本文在深入研究了工作流技术、计算机支持的协同技术以及面向XML的数据挖掘技术的基础上,首先提出了一种综合考虑内容、结构以及工作流文档潜在信息的工作流模型推荐算法;其次设计了一种可扩展的协同框架,并实现了协同过程与具体应用的松散耦合;最后将工作流建模工具与该协同框架进行了集成,实现了建模过程的多人跨地域协同,同时在建模工具中应用本文提出的工作流模型推荐算法,对工作流模型进行相似性比较,为建模人员产生推荐。在具体的研究工作中,本文主要的创新点有以下几个方面:1.深入研究了XML的挖掘技术以及工作流模型的XML文档,从内容、结构以及工作流文档潜在信息三个方面对文档进行量化,提出了基于XML的工作流模型推荐算法。为工作流模型的挖掘提供了一种有效的方法。2.对计算机支持的协同工作技术进行研究,设计了一种可扩展的协同框架,实现了协同过程与具体应用的松散耦合,并充分考虑了系统的资源问题,采用多个协同服务机制,避免了系统资源紧缺的问题。3.将工作流建模工具集成到该协同框架上,实现了建模过程的多人跨地域协同操作,并在协同建模工具中应用基于XML的工作流模型推荐算法,在用户建模操作过程中,为用户动态挖掘网络中的相关工作流模型,推荐给用户参考。有效降低了建模难度,提高了建模效率。实验结果表明,本文设计的支持流程推荐的工作流协同建模工具可以支持多人跨地域协同建模,同时可以产生较准确的流程推荐,有效提高了建模效率,降低了建模难度。

【Abstract】 Workflow is a new technology that develops rapidly recent years in computer applications fields,and as realization of workflow technology, workflow management systems have widely applied in many trades.The quality of workflow input into workflow system is key to the success of the implementation of workflow system.Therefore,workflow modeling has grown to be one of the major research topics in the area of workflow management.Until recently , little attention has been paid to the collaborative modeling in most of conventional workflow management systems. Workflow models must be designed by professional modeling designers. With workflow management systems used in industry becoming increasingly complex , the business processes managed by them are not only within an organization , but also across organizations. These business processes can not be modeled by a single modeling designer , because this can not make the best of the knowledge of experts in different domain. So that , we have developed a workflow prototype system Wflow recently , and a collaborative modeling tool has been developed. In this paper , the main techniques of this tool implementation are discussed , and a workflow model is presented. Meanwhile,workflow always complex and it’s model information can’t be used efficiently,there is no reference for modeling designer during modeling.They need a method of workflow model information mining to use these information and debase difficulties in modeling.In this paper,we lucubrated workflow technology, Computer Supported Cooperative Work and XML mining.Based on these technologies,a workflow model recommendation arithmetic is be brought forward,which is combined content,structure and workflow document ’s latent information;and then we designed a cooperative frame; finally we integrated the frame and workflow modeling tool to support cooperative modeling for group, simultaneous,we use workflow model recommendation arithmetic to measure the similarity of workflow model document,given recommendation to modeling designer. Innovations during the research are exhibited below:1. Lucubrated workflow technology ,a workflow model recommendation arithmetic is be brought forward,which is combined content,structure and workflow document ’s latent information.2. Research on Computer Supported Cooperative Work, designed a cooperative frame,this frame made cooperative process and application incompact coupling .3. Integrated the frame and workflow modeling tool to support cooperative modeling for group, simultaneous,we use workflow model recommendation arithmetic to measure the similarity of workflow model document,given recommendation to modeling designer.Experimentation results indicated that the workflow modeling tool we brought forward can support cooperative modeling for group and given relative workflow recommendation,it is enhanced modeling efficiency and reduced modeling difficulty.

【关键词】 工作流模型XPDL相似度量XML文档CSCW
【Key words】 Workflow modelXPDLsimilarity measureXML documentCSCW
  • 【网络出版投稿人】 扬州大学
  • 【网络出版年期】2007年 06期
  • 【分类号】TP311.52
  • 【被引频次】4
  • 【下载频次】136
节点文献中: