计算机仿真
計算機倣真
계산궤방진
COMPUTER SIMULATION
2010年
1期
316-318,333
,共4页
张细香%刘建勋%雷静%杨保安
張細香%劉建勛%雷靜%楊保安
장세향%류건훈%뢰정%양보안
加性一致性%满意一致性%模糊互补判断矩阵%阈值%仿真
加性一緻性%滿意一緻性%模糊互補判斷矩陣%閾值%倣真
가성일치성%만의일치성%모호호보판단구진%역치%방진
Additive eousistency%Satisfactory eousisteney%Fuzzy complementary judgment matrix%Threshold%Simulation
由于事物的不确定性、信息获取的限制、甚至专家知识的局限性,决策者在决策过程中很难用精确数来表示其偏好,而喜欢用模糊语言来表示.当决策者给出的模糊互补判断矩阵不满足加性一致性,加性一致性指标阈值的确定是一个值得研究的问题.一般来说,决策者给出的模糊互补矩阵具满意一致性,且加性一致性模糊互补矩阵一定是满意一致性模糊互补矩阵.MATLAB中计算机仿真方法构造了大量的满意一致模糊互补判断矩阵,通过计算其加性一致性指标值及其概率分布,获取模糊互补矩阵的加性一致性指标的逼近阈值(0.2).如果模糊互补判断矩阵的加性一致性指标值小于等于0.2,则矩阵是可接受的;否则需要修订.
由于事物的不確定性、信息穫取的限製、甚至專傢知識的跼限性,決策者在決策過程中很難用精確數來錶示其偏好,而喜歡用模糊語言來錶示.噹決策者給齣的模糊互補判斷矩陣不滿足加性一緻性,加性一緻性指標閾值的確定是一箇值得研究的問題.一般來說,決策者給齣的模糊互補矩陣具滿意一緻性,且加性一緻性模糊互補矩陣一定是滿意一緻性模糊互補矩陣.MATLAB中計算機倣真方法構造瞭大量的滿意一緻模糊互補判斷矩陣,通過計算其加性一緻性指標值及其概率分佈,穫取模糊互補矩陣的加性一緻性指標的逼近閾值(0.2).如果模糊互補判斷矩陣的加性一緻性指標值小于等于0.2,則矩陣是可接受的;否則需要脩訂.
유우사물적불학정성、신식획취적한제、심지전가지식적국한성,결책자재결책과정중흔난용정학수래표시기편호,이희환용모호어언래표시.당결책자급출적모호호보판단구진불만족가성일치성,가성일치성지표역치적학정시일개치득연구적문제.일반래설,결책자급출적모호호보구진구만의일치성,차가성일치성모호호보구진일정시만의일치성모호호보구진.MATLAB중계산궤방진방법구조료대량적만의일치모호호보판단구진,통과계산기가성일치성지표치급기개솔분포,획취모호호보구진적가성일치성지표적핍근역치(0.2).여과모호호보판단구진적가성일치성지표치소우등우0.2,칙구진시가접수적;부칙수요수정.
Due to the uncertainty, constraints of knowledge, decision makers cannot provide exact numbers to ex-press their preferences, and prefer to use fuzzy linguistic terms to express their preferences. If the given complemen-tary matrix is not additive consistent, how to choose the threshold to judge the given complementary matrix' s accepta-bility is worthy studying. Generally speaking, it is easy for a decision maker to give a fuzzy complementary matrix with satisfactory consistency. Meanwhile, a fuzzy complementary matrix with additive consistency must be satisfactory consistent. Thus, thousands of fuzzy complementary judgment matrices with satisfactory eousisteney were built in MATLAB and the additive consistency index of these matrices were computed. Through analyzing the distribution of the matrices' additive eousisteney indexes, the approximate threshold of additive consistency index was deduced. Therefore, if the additive consistency index of a given fuzzy complementary matrix is smaller than or equals 0.2, it is acceptable; otherwise, it needs to be revised.