【作者】 张劲松；

【导师】 钟毅芳；

【作者基本信息】 华中科技大学 ， 机械设计及理论， 2004， 博士

【摘要】 全球化的市场竞争环境迫使企业采取与之相应的产品开发模式,企业在更加专业化的同时加强了跨企业的分工与协作。本文用协同开发链描述和表达分布式环境下的产品开发网络,它以任务流为纽带,将产品开发各个阶段的开发成员紧密联系在一起。本文结合多个课题,重点研究了协同产品开发链的理论框架及其若干关键技术,包括产品开发链过程建模和管理、基于结构模型的产品配置、基于协同代理的产品数据及应用集成、复合产品数据权限管理等。在协同开发链理论框架研究方面,本文从内涵、概念模型以及协同控制等三个方面研究了协同开发链的基本原理,指出任务流是开发链的本质属性,分析了开发链的组成要素;产品开发链的组织模型与市场需求、开发目标以及产品功能结构存在密切联系,本文从成员选择、产品对象及需求分解两个方面研究了开发链的构建原理;为了支持协同开发链的实施,本文提出了协同开发链支撑系统层次结构,明确了各个层次的基本组成,研究了协同开发链的工作模式。为了管理和控制开发链的基本要素,协同开发链需要一些管理功能对其提供支撑,本文介绍了产品对象管理、协同开发过程管理、产品配置管理、数据安全管理、组织结构管理等功能。在协同开发链的过程建模与管理方面,通过分析协同开发链产品开发过程的特点,建立了基于事件驱动的开发链过程模型,研究了事件驱动过程链与工作流模型的映射规则;通过分析开发链任务之间的关系,研究了基于路由规则和任务约束的柔性工作流模型;为了确保工作流模型的完备性,提出了规则遗漏检测算法和不完善路由分支检测算法。协同开发链的分层递阶性使得其开发过程也具有相似的特点,本文提出了基于任务分解的任务流管理模型,通过任务逐级分解,降低了流程的复杂性以及开发过程的可操作性,采用这种任务分解和工作流相结合的方法,实现了对产品开发链任务流的有效管理。产品配置过程是协同开发链的重要一环。本文提出了基于结构模型的智能化产品配置过程,使配置知识与特定模型相关,降低了配置知识库的复杂度;通过对产品配置规则的深入分析,总结归纳了变量配置规则、配置约束规则以及版本有效性规则,介绍了配置知识库的构成及管理,并给出了配置方案的推理求解过程。配置方案的一致性和有效性通过配置知识库的正确性来保证,本文提出了基于测试案例的配置知识库诊断模型,通过测试案例集来诊断知识库中多余的约束规则以及应增加的规则。产品数据管理是协同开发链的重要支撑功能。本文重点研究了产品数据管理的两个问题:其一是产品数据及应用集成,其二是产品数据权限管理模型。本文提出了基于协同代理的产品数据和应用集成框架,协同代理负责本地工作区产品数据和设计资源的管理,并承担协同过程中的数据交换和协同控制,有效实现了应用工具之间的集成,并基<WP=5>于协同代理形成了协同开发环境。通过分析产品开发过程中权限管理相关对象及它们之间的关系,提出了基于任务和角色的复合产品数据安全模型,研究了产品开发流程中的动态权限控制,给出了产品数据存取权限的计算方法。在实践方面,本文基于IntePDM系统实现了协同产品开发链的相关支撑功能,包括开发链过程建模和过程管理、基于结构模型的产品配置、产品数据及应用集成机制、产品数据权限管理模型等,并将这些研究成果应用于东风柳汽公司、桂林橡胶机械厂等制造企业的产品开发系统集成项目中,取得了良好应用效果。更多还原

【Abstract】 Global market competition requires manufacturing enterprises to improve their product development. While enterprises become more and more specialized, cooperation or collaboration among enterprises is intensified simultaneously. Commerce processes in supply chain become agile collaborative product development chain (CPDC) based on task flow. CPDC don’t only describe organization structure of distributed development team, i.e. physical level of CPDC, but also represent development process chain based on task decomposition relations. Based on several projects, the whole dissertation focuses on CPDC, chiefly including some key technologies to support the implementation of CPDC effectively.Firstly this dissertation establishes general theory frame of CPDC and proposes a multi-view model that consists of organization, process and environment. System structure and running pattern of CPDC are discussed, and then five constituents of CPDC are introduced, which include development object, organization structure, development activity, product data, and collaborative control etc. In order to manage and control CPDC’s constituents some basic management functions are needed, such as product object management, collaboration development process management, product configuration management, data security management, organization management and so on. System framework of CPDC is proposed, which consists of five levels, i.e. application interfaces, functions or tools for product development, process control, data management, and database system. In process modeling and management, based on analysis of the characteristic of product development process(PDP) in CPDC, an event-driven PDP chain model is proposed and a formal representation for PDP chain is given. Using the mapping between PDP chain and workflow, a flexible workflow model enhanced with rules and constraints is presented, which improves the adaptability and flexibility of CPDC. In order to manage CPDC, task flow model based on task decomposition is proposed, which reduces the complexity of CPDC. Through task flow and workflow, development process in CPDC can be organized and managed effectively. Product configuration is one of key technologies in CPDC in the environment of mass customization. Intelligent product configuration approach based on product structure model(PKM) is proposed, in which configuration knowledge, i.e. configuration rule or constraint, is attached to a specific PKM for the sake of simplifying configuration knowledge base(CKB). The construction and management of CKB is introduced, and knowledge-based configuration solving process is given. In order to ensure the validity of knowledge base, <WP=7>case-based diagnosis model is discussed, through which redundant or missing configuration rule in CKB can be detected.In data management of CPDC, this dissertation discusses product data integration and data access control model in development process. Product data integration framework based on collaborative agent(CA) is presented, which overcomes the deficiency of conventional integration approach. CA can not only manage the product data in local work area, but also support the collaborative work between application tools. As for product data access control in CPDC, task-role-based access control model(TACM) is proposed after analyzing access control objects and relations among them, and dynamic right computation process in TACM is given. Finally, this dissertation introduces the CPDC’s implementation in some projects, which testified the advancement and practicability of research achievements in this dissertation.更多还原