模糊信息条件下多属性决策的几个重要问题研究

发布时间：2018-03-24 04:31

本文选题：模糊多属性决策　切入点：对偶犹豫模糊集　出处：《西安建筑科技大学》2017年硕士论文

【摘要】：随着时代发展,决策过程中蕴含的信息量越来越大,决策问题也越来越复杂,这就使得决策者在对事物做出判断的时候往往变得犹豫不决.决策结果仅仅依靠决策者经验和直觉得到的时代已经成为历史,采用科学的决策方法对各备选方案进行评价、权衡并选取最优方案是亟不可待的.模糊信息条件下多属性决策中的模糊理论在科技研究、社会生产、生活决策等领域都有广泛的应用,故模糊信息条件下多属性决策的出现填补了决策理论研究的空缺.但由于任何事物都是纷繁复杂的具有的模糊性并且人类认知结构也具有不确定性,决策者一般难以对决策信息给出精确的数值,最初的属性值形式为三角模糊数、区间数、直觉模糊数等,现今人们为了得到更精确的决策结果则用区间三角模糊数、对偶犹豫模糊数、区间值对偶犹豫模糊数等不同形式的模糊信息给出属性值.本文对不同属性下的模糊多属性决策问题做了如下研究:首先,研究了对偶犹豫模糊集的距离测度和相关系数,并将其应用于属性权重未给出的模糊信息条件下多属性决策中.基于对偶犹豫模糊集,给出对偶犹豫模糊集的Hamming距离测度公式和两个对偶犹豫模糊信息之间相关关系的相关系数,并给出定义和加权相关系数计算公式,使决策运算更快捷有效.其次,探讨区间三角模糊集的Hamming距离,结合TOPSIS方法对属性权重部分已知的问题进行决策.由区间三角模糊数的定义,给出任意两个区间三角模糊集的标准化的Hamming距离,将其与极大偏差法结合求部分已知的权重向量.最后结合经典TOPSIS方法计算得到最优方案.然后,研究了区间直觉模糊数的规范化方法和由区间直觉模糊加权几何平均算子对动态多属性决策问题的决策.建立属性值为区间直觉模糊数的规范化方法,且把区间直觉模糊加权几何平均算子推广到了多阶段的情形.最后,研究了区间值对偶犹豫模糊熵与相似性测度给出了其定义和公式,由此构造了熵权重模型;由距离与相似性测度的关系给出三种区间值对偶犹豫模糊集的距离公式.由以上给出一种区间值对偶犹豫模糊集的决策方法.最终给出区间值对偶犹豫模糊集多属性决策方法的计算步骤,并验证了该方法的应用性.
[Abstract]:With the development of the times, the amount of information contained in the decision-making process is increasing, and the decision-making problems are becoming more and more complex. As a result, the decision makers tend to hesitate when they judge things. The era when decisions are based solely on the experience and intuition of the decision makers has become a thing of the past, and scientific methods of decision making are used to evaluate the alternatives. It is very important to weigh and select the optimal scheme. Fuzzy theory in multi-attribute decision making under the condition of fuzzy information has been widely used in the fields of science and technology research, social production, life decision making and so on. Therefore, the emergence of multi-attribute decision making under the condition of fuzzy information fills the gap in the study of decision theory. However, because everything is complicated and fuzzy and human cognitive structure is uncertain, Decision makers generally find it difficult to give accurate values for decision information. The initial attribute values are triangular fuzzy numbers, interval numbers, intuitionistic fuzzy numbers, etc. Nowadays, people use interval triangular fuzzy numbers in order to obtain more accurate decision results. Attribute values are given for different forms of fuzzy information such as dual hesitation fuzzy number, interval value dual hesitation fuzzy number and so on. In this paper, the fuzzy multi-attribute decision making problem under different attributes is studied as follows: first, This paper studies the distance measure and correlation coefficient of dual hesitation fuzzy sets, and applies them to multi-attribute decision making under the condition of fuzzy information not given by attribute weights. The Hamming distance measure formula of dual hesitation fuzzy set and the correlation coefficient between two dual hesitation fuzzy information are given, and the definition and calculation formula of weighted correlation coefficient are given to make the decision operation more efficient. In this paper, the Hamming distance of interval triangular fuzzy sets is discussed, and the problem of known attribute weights is determined by TOPSIS method. Based on the definition of interval triangular fuzzy numbers, the standardized Hamming distance of any two interval triangular fuzzy sets is given. It is combined with the maximum deviation method to obtain the partial known weight vector. Finally, the optimal scheme is obtained by combining the classical TOPSIS method. In this paper, the normalization method of interval intuitionistic fuzzy numbers and the decision of interval intuitionistic fuzzy weighted geometric averaging operators for dynamic multi-attribute decision making problems are studied, and the normalization method of interval intuitionistic fuzzy numbers with attribute values as interval intuitionistic fuzzy numbers is established. The interval intuitionistic fuzzy weighted geometric averaging operator is extended to the multi-stage case. Finally, the definition and formula of interval valued dual hesitation fuzzy entropy and similarity measure are studied, and the entropy weight model is constructed. From the relation between distance and similarity measure, the distance formulas of three interval valued dual hesitation fuzzy sets are given. A decision method for interval valued dual dual hesitation fuzzy sets is given. Finally, the interval valued dual dual hesitation fuzzy sets are given. The calculation steps of the attribute decision method, The application of this method is verified.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O159;O225

【参考文献】