【作者】 张黎；

【导师】 石为人；

【作者基本信息】 重庆大学 ， 控制理论与控制工程， 2003， 硕士

【摘要】 随着信息技术和通讯技术的迅速发展，动态联盟的出现，异地/异时的分布式群体决策己经成为新的趋势。分布式群体决策过程实际上是一种群体交互与协作的过程，这不是一般意义上的合作，而是快速、敏捷和准确的协作，需要有新的技术、新的方法支持决策成员的行为、协作方式和决策过程。Agent技术是分布式智能研究中的热门技术，有关Agent和多Agent系统的应用研究己成为国际上一个生机勃勃的研究领域，并在复杂工业过程智能控制、人机协同、新型企业组织、智能机器人、信息系统集成与协调等方面取得了成绩。本文以重庆市制造业信息化重大专项子项目“重庆青山工业有限责任公司CAD/CIMS一期工程(2001-03)”为选题背景，并获得重庆市应用基础研究项目“基于Agent的主动信息发现与服务理论”（No．6802）的资助，针对企业信息集成普遍存在的适应企业业务变迁、人机协同和集成异质信息资源等方面能力不足的问题，深入研究对群体决策提供广泛支持以及对各种不可预料的群体活动进行灵活支持的理论、方法和技术，引入多Agent理论和技术应用于企业群决策支持系统（GDSS—Group Decision Support Systems）。应用基础研究是本文的重点。以实现柔性的、开放的分布式智能GDSS为目的，与分布式人工智能相结合，以实现GDSS体系结构的柔性、智能性和开放性的具体实现方式以及GDSS的决策过程、决策模式、内部控制机制及算法为目标，着重开展了以下研究工作：1．综述了群决策支持系统技术发展历史、群决策理论和应用发展现状及其存在的问题，研究了GDSS和DSS的区别与联系。2．研究了各种人工智能的优缺点及其在GDSS中的功能，综述了Agent和多Agent技术发展历史、多gent技术的理论和应用发展现状，提出利用面向对象技术和Multi-Agent系统来构建GDSS的思想，就多Agent技术在GDSS中的研究方向和系统开发方法进行了探讨。3．分析了企业GDSS决策支持过程，并提出了一个基于分布式对象的GDSS集成结构，通过对Multi-Agents环境下各类Agent的内部结构、通讯、实现功能、过程算法以及实现自治、协作Agent的具体方法的研究，提出了以部门为分布式群体决策单元建立企业经营GDSS的思路。对GDSS的决策步骤、决策方法和决策过程进行了详细分析，设计了基于群Agent群、部门Agent群、Server Agent群的企业经营GDSS系统结构模型。最后，给出了基于任务的群体一致性及Agent协同谈判模型与算法4．从企业GDSS产生的社会、经济和技术背景入手，分析了GDSS研究的<WP=5>现状和面临的问题，从而提出了开展GDSS研究的必要性。以青山工业有限责任公司为例，建立了基于Multi-Agent的企业经营GDSS的组织体系结构和开发过程，在企业经营GDSS中对所研究的体系结构、方法、概念的运用进行了详细研究。最后总结了全文的工作，并对基于多Agent技术进行企业群决策支持研究进行了展望。更多还原

【Abstract】 With the rapid development of information technology and communication technology, and the appearance of dynamic alliance, the distributed group decision in different location/different time have been becoming a new trend. The distributed group decision actually is a process of group interaction and cooperation, not general cooperation, but rapid, agile and exact coordination that require some new technologies and new methods to support decision members’ action, coordination pattern and decision process.Agent has become an important technology in both artificial intelligence and mainstream computer science. Agent and Multi-agent systems (MAS) technology is a rapidly development area and research. Agent and MAS are being used in broad fields: complex industry control systems, electricity system management, enterprise information integration and new enterprise organization. This work is support by Chongqing Research Award of Applications Foundation: Agent-based Initiative Information Discovery and Service (No: 6802). This dissertation is devoted to the theoretical and practical study of enterprise group decision support systems (GDSS) based on MAS. The main contents are as follows:1. The historical development of GDSS is reviewed from theory and practice. Great emphasis is place on the lacks of GDSS, and similarities and differences between DSS and GDSS. 2. The methodologies of artificial intelligence and their functions in GDSS are studied, and the historical development of GDSS is reviewed from theory and practice as well. Then the thought of using object-oriented technology and MAS to construct GDSS is proposed. Great emphasis is place on research methods of GDSS based MAS.3. The integrated framework of flexible GDSS based distributed objects is proposed. Then this dissertation puts forward the thought of establishing enterprise management GDSS that takes departments as basic decision units by the study on the internal structure, communication, function, process algorithm, implementation pattern of various agent under multi-agent environment, and implementation methods of autonomous and cooperation agents. After analyzing the decision steps, decision methods and process in detail, this dissertation proposes the whole framework of enterprise management GDSS based group agent group, department agent group and<WP=7>Server agent group. And fuzzy models and algorithms of task-based GDSS and cooperative negotiation of agents.are presented as well.4. An application system of MAS-based GDSS is build in Qingshan Industry Co. Ltd (Qingshan). This dissertation studies the social, economic and technical background of the emergence of GDSS in Qingshan and proves the necessity of investigating GDSS, and then establishes an actual agent-based enterprise GDSS. Finally, the dissertation concludes with a series of vistas of future enterprise GDSS based on MAS.更多还原