【作者】 朱佳俊；

【导师】 郑建国；

【作者基本信息】 东华大学 ， 管理科学与工程， 2010， 博士

【摘要】 群决策理论研究始于20世纪70年代,而后经过多学科的交叉组合,逐步形成了系统的理论体系,使得群决策的研究成果被广泛应用于经济管理、系统科学等。由于群决策问题存在偏好、权重等内在的复杂性,引入可拓理论方法进行相关的不确定性分析,认识其内在机理,以辅助管理决策。本文以物元分析、群决策理论为基础,提高群决策的效率与效果为目的,对可拓群决策的理论作了全面系统的分析与研究,运用管理科学与信息技术等方法,对不确定条件下可拓群决策过程的优化进行了深入的探讨,并结合具体实例进行仿真研究。论文研究工作主要包括以下几个方面：首先,针对可拓群决策数据缺失的特点及数据处理的要求,通过分析关联函数的距、位值,提出了可拓对应性的含义、性质,阐述了其在可拓群决策中数据缺失处理的优越性,建立了基于对应性的可拓群决策区间分析的模型与步骤,不仅减少了由于主观估计而产生的决策偏差,而且可以提高数据预测与分析的效果,为不确定条件下方案的识别、判断和选择提供依据。其次,针对可拓群决策的有效性和系统性的问题,把可拓集与群决策优化相结合,从决策的可靠性与敏感性、关联性与可变性出发,研究了不确定条件下多维可拓群决策的权重优化与可拓关联、矛盾问题的描述与可拓变换、决策偏好下多目标可拓群决策优度与可拓区间,以实现决策对象在群体交互和个体偏好集结基础上的比较与选择,拓宽了可拓群决策的研究范围和应用领域。第三,针对用物元可拓集描述的群决策数据挖掘问题,把关联函数与决策函数相结合、可拓变换与等价关系相结合,研究了不确定条件下可拓群决策的分类与聚类、规则提取与数据分析,构建了不确定条件下的可拓数据挖掘模型,解决了多方案可拓分类识别的系统决策问题,以提高不确定条件下群决策分析的系统最后,通过对金融危机下国内玩具业的发展现状、趋势的描述,将可拓群决策定量与定性分析的方法引入到世纪友谊玩具有限公司的“小鲁班”系列玩具的改进与创新中。在市场调研、创意比赛及区域销售业绩等统计资料的基础上,运用可拓群决策方法,对不确定条件下“小鲁班”产品研发、改造、组合销售三个环节进行分类与决策,以辅助企业制定产品发展战略。更多还原

【Abstract】 With the research of group decision making theory beginning in the 1970’s, then through the overlap and combination of multi-disciplinary, a systematic theory system of group decision making has been gradually taken shape. Its research results are widely used in economic management, systems science, and etc.. Due to inherent complexity such as preferences, weights, and etc. in group decision making problems, the introduction of extension theoretical and method helps a related uncertain analysis and understanding of its inherent mechanism in order to aid management decisions.Based on matter-element analysis and group decision-making theories, aiming at the promotion on efficiency and effectiveness of extension group decision-making, this thesis makes a comprehensive and systematic analysis and research of theoretic system of extension group decision-making. It applies methods of management science, information technology, and etc. to profound investigation into optimization of extension group decision-making process under uncertainty, and combines with specific examples in simulation studies. It mainly revolves the following aspects:Firstly, on account of features of missing data and requirements of data processing in extension group decision-making, it brings forward the meaning, nature of extension correspondence and analyzes its superiority in data prediction in extension group decision-making by analyzing the distance, bit value of correlation function, model and steps of data interval analysis of extension group decision-making based on extension correspondence. Not only does this method reduce the decision-making bias by subjective estimation, but also improves the results of data forecasting and analysis, and provides a basis for recognition, judging and selection extension group decision-making under uncertainty.Secondly, in accordance with the effectiveness and systematization of extension group decision-making, on the combination of extension set with group decision optimization, it studies the weight optimization and extension association, contradictory problem description and extension transformation, the goodness and the extension space under decision-making preferences of the multi-dimensionality group decision-making under uncertainty from the perspectives of reliability and sensitivity analyses, relevance and variability of extension group decision-making. It achieves the goal of comparison and selection of object based on group interaction and individual preferences assembly, and broadens the scope of study and application of extension group decision-making.Thirdly, in terms of the problem of data mining in group decision-making with matter-element extension set, on the combination of correlation function with decision one and integrating extension transformation with equivalence relation, it studies classification and clustering, rules extraction and data analysis of extension group decision-making under uncertainty. It builds a data mining model of extension group decision-making as a result of solving systematic decision-making problems of multi-project extension classification and recognition, so as to improve the systematization of extension group decision-making analysis under uncertainty.Lastly, through the description of status quo and developing trend of domestic toy industry under the financial crisis, the quantitative and qualitative analysis methods of extension group decision-making are introduced into the improvement and innovation of "SLUBAN" toy series of Century Youyi Toys Co., Ltd.. According to the statistics of market research, creative competition and regional sales performance, the extension group decision-making methods are used to make classification and decision-making of product development, transformation and package sales of "SLUBAN" toys so as to aid the enterprise to work out its product development strategies.更多还原