摘要
将粗集和证据理论相结合应用到多属性决策评价中,证据理论应用中会存在“焦元”爆炸问题,我们采用基于粗集特征属性约简的办法在保证分类质量不变的情况下,一方面削减了证据之间的相关性,尽可能采用少的属性对决策问题进行评价,防止发生“焦元”爆炸;另一方面能够从决策表蕴含的信息中得出各属性的权重,避免专家赋权的主观性,以配送中心选址问题为例证明了此方法是可行的、有效的。
In this paper, rough sets and evidence theory are applied to the research on the evaluation of multi-attribute decision making. Focal element increases rapidly while applying the evidence theory. A feature reduction algorithm based on rough set is used when classification result is almost invariable. On one hand, in order to reduce the dependence between evidences, the authors adopt a little attributes to evaluation decision-making problem, on the other hand, in order to prevent from focal element increasing rapidly, the authors draw the weight of every attribute from the decision form and prevent subjectivity points of the experts. Finally, taking the site selection of distribution center for an example proves that the method is feasible.
出处
《科学学与科学技术管理》
CSSCI
北大核心
2005年第9期39-42,共4页
Science of Science and Management of S.& T.
关键词
粗集
证据理论
多属性决策
配送中心选址
rough set
evidence theory
multi-attribute decision making
site selection of distribution center
