摘要
粗糙集权重确定方法利用属性重要度来确定属性的客观权重,这种方法克服了权重确定过分依赖专家经验知识的不足,但也忽略了各属性指标本身的重要性问题。本文将粗糙集权重确定方法与专家主观赋权的层次分析法相结合,基于方差最大化的原理提出了一种主客观组合赋权方法。粗糙集结合主观层次分析法的组合权重很好地将两者的优点有效结合,同时方差最大化原理也对传统组合权重确定方法进行了改进,使得权重的确定更加合理,更能有助于决策者做出决策。最后,用一个算例应用证明了该种组合赋权决策方法的可行性。
Rough set method weights the attributes according to the significance degree, which overcome the shortage of common weighting methods that overly relied on the expert＇s transcendental knowledge. But the autologous improtance of the attributes may always be neglected. In this paper, Analytic Hierarchy Process（AHP） is used together with rough set weighting method to put forward a new combination weighting method based on maximizing deviation principle. This new combination weighting method may integrate the merits of both subjective and impersonal weightiing methods, maximizing deviation improves the traditional combination weighting method, and may lead to a more reasonal weighting result and easy decision. Finally, an applied example proves the feasibility of the new combination weighting method.
出处
《中国管理科学》
CSSCI
北大核心
2011年第6期141-148,共8页
Chinese Journal of Management Science
基金
国家自然科学基金资助项目(70872010)
国家社会科学基金重大项目(10zd&029)
关键词
粗糙集
层次分析法
组合赋权
方差最大化
rough set
analytic hierarchy process
combination weighting
maximizing deviations
