基于ITFN信息关联输入的改进群体MULTIMOORA决策方法  

An improved MULTIMOORA approach for group decision making based upon interdependent inputs of intuitionistic trapezoidal fuzzy numbers

作　　者：陈振颂 李延来[1,2] 

机构地区：[1]西南交通大学交通运输与物流学院,成都610031 [2]西南交通大学综合交通运输智能化国家地方联合工程实验室,成都610031

出　　处：《系统工程理论与实践》2014年第12期3138-3156,共19页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(71371156;70971017);西南交通大学优秀博士学位论文培育项目

摘　　要：提出基于直觉梯形模糊数(intuitionistic trapezoidal fuzzy number,ITFN)极小、极大期望值的序关系判别准则,并引入风险系数构建ITFN相对完善的带有决策者风险偏好的运算规则,在此基础上定义直觉梯形模糊Bonferroni(intuitionistic trapezoidal fuzzy Bonferroni,ITFB)平均算子,验证其相关性质.针对决策者之间、属性之间分别存在关联关系且权重均未知的多属性群决策问题,提出基于ITFN信息关联输入的改进群体MULTIMOORA决策方法.首先,构建直觉梯形模糊决策矩阵序列,予以标准化处理,并将其转化为极小期望决策矩阵序列;其次,综合利用基于熵权法和考虑决策者偏好关联的基于2-可加模糊测度与Choquet积分联合的主客观赋权法确定决策者权重及属性权重;最后,分别引入WITFB平均算子及ITFN的Hamming距离以改进传统MULTIMOORA决策方法,基于优势理论可对方案展开综合排序以确定最优方案.通过算例分析验证本文方法的可行性及有效性.A ranking method of intuitionistic trapezoidal fuzzy numbers （ITFNs） is proposed based on the notions of a minimum expectation and a maximum expectation. Considering risk preferences of de- cision makers, a novel concept of a risk coefficient is introduced to construct improved operational laws of ITFNs. Furthermore, an intuitionistic trapezoidal fuzzy Bonferroni （ITFB） mean operator is proposed based on the improved operational laws, and then the relative properties of the ITFB mean operator are investigated. With respect to a multi-attribute group decision making problem, in which decision-makers are interdependent, attributes are interdependent, and decision-makers＇ weights and attributes＇ weights are both unknown, an improved MULTIMOORA approach for group decision making of interdependent ITFNs inputs is proposed. In this approach, firstly, a set of intuitionistic trapezoidal fuzzy decision ma- trixes is constructed, and then a set of normalized minimum expectation decision matrixes is obtained by calculating the minimum expectation one corresponding to each intuitionistic trapezoidal fuzzy decision matrixes. Secondly, in order to determine decision-makers＇ weights and attributes＇ weights, an entropy weight approach for determining the attributes＇ weights associated with each decision-maker is integrated into an objective and subjective synthetic approach, which has considered interactions of decision-makers＇ preferences, for obtaining decision-makers＇ weights based on the combination of a 2-additive fuzzy measureand a Choquet integral. Finally, a weighted ITFB mean operator and a Hamming distance of ITFNs are respectively introduced to improve the traditional MULTIMOORA approach, the improved MULTI- MOORA approach is employed to obtain a ranking of alternatives corresponding to each one of three ordering approach, and then a dominance theory is utilized to summarize the three rankings into a single one. A practical case is used to illustrate the validity and feasibility of the proposed

关 键 词：直觉梯形模糊数 ITFB平均算子 MULTIMOORA 极小期望值 CHOQUET积分 多属性群决策 

分 类 号：C934[经济管理—管理学]