基于加权可能性均值的直觉梯形模糊数矩阵博弈求解方法  

Method based on weighted possibility mean for solving matrix games with payoffs of intuitionistic trapezoidal fuzzy numbers

作　　者：万树平[1] 张小路[1] 王凌[1] 王圣尧[1] 方晨[1] 

机构地区：[1]江西财经大学信息管理学院,南昌330013

出　　处：《控制与决策》2012年第8期1121-1126,1132,共7页Control and Decision

基　　金：国家自然科学基金项目(71061006,70861002);教育部人文社科项目(09YGC630107);江西省自然科学基金项目(20114BAB201012);江西省教育厅科技项目(GJJ12265);江西财经大学优秀青年学术人才支持计划项目;江西财经大学第6届学生科研课题项目

摘　　要：针对支付值为直觉梯形模糊数(ITFN)的矩阵博弈求解问题,提出了一种基于加权可能性均值的求解方法.定义了ITFN新的运算法则,并引入ITFN的下、上加权可能性均值和加权可能性均值的概念,根据加权可能性均值给出了ITFN新的排序方法;运用新的排序方法,将求解局中人最优策略问题转化为求解双目标线性规划问题.实例分析验证了所提出方法的可行性和有效性.For the problem of matrix games with payoffs of intuitionistic trapezoidal fuzzy numbers（ITFNs）, a solving method based on weighted possibility mean is proposed. The new operation laws for ITFNs are defined. The notions of lower and upper weighted possibility means for ITFNs are introduced as well as the weighted possibility mean. A new ranking approach for ITFNs is given according to the weighted possibility mean. According to the new ranking approach, the optimal strategies of two players can be obtained by solving the bi-objective linear programming model. The example analysis verifies the feasibility and effectiveness of the proposed method.

关 键 词：直觉梯形模糊数 矩阵博弈 可能性均值 双目标规划 直觉模糊集 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]