主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2014, Vol. 22 ›› Issue (5): 98-103.

• 论文 • 上一篇    下一篇

基于样本集的区间数灰靶分类决策模型及应用

梁燕华1, 郭鹏2, 朱煜明2   

  1. 1. 杭州电子科技大学管理学院, 浙江 杭州 310018;
    2. 西北工业大学管理学院, 陕西 西安 710072
  • 收稿日期:2011-09-20 修回日期:2012-10-10 出版日期:2014-05-20 发布日期:2014-05-14
  • 作者简介:梁燕华(1979-),女(汉族),河南南阳人,杭州电子科技大学管理学院,讲师,研究方向:多属性决策、项目评价.
  • 基金资助:

    国家自然基金资助项目(71373964);国家社会科学基金资助项目(10BJY024);教育部人文社会基金资助项目(13YJC630177);科研启动经费项目(KYS035613029)

A Sample-Set based Interval Gray Target Classification and Decision-Making Model and Its Applications

LIANG Yan-hua1, GUO Peng2, ZHU Yu-ming2   

  1. 1. Management School, Hangzhou Dianzi University, Hangzhou 310018, China;
    2. Management School, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:2011-09-20 Revised:2012-10-10 Online:2014-05-20 Published:2014-05-14

摘要: 基于灰靶思想的不确定背景分类决策问题分析,本文提出了区间数灰靶分类决策模型。该模型将灰靶决策拓展到决策信息为区间数的情况,提出了区间数的靶心距测度方法;根据靶心距提出了灰靶分类决策中的靶心分类临界值设置方法;以决策对象的靶心距与临界值之间的偏差总量最小为目标,建立了指标权重和分类临界值的确定模型;依据求解的权重与分类临界值对决策对象集进行分类评价。算例分析验证了该模型的有效性和可行性,可以很好地解决决策对象众多、分类数不确定等特性的多属性分类决策问题。

关键词: 灰靶, 区间数, 分类, 样本集, 多属性决策

Abstract: Based on the analysis of gray target decision making under uncertainty, an interval gray target classification and decision-making model is put forward. In this model,the gray target decision-making is extended to uncertain circumstance in which the interval numbers serve as the decision-making information, and a method for measuring the target-center distances of the interval numbers is proposed. According to the target-center distance, a method for determining the critical values of target-center classification in the gray target classification and decision-making is presented. Besides,a determined model involving the target weights and the critical values of the classification is constructed, aiming at the minimum value of the accumulate deviation between the target-center distances in the case and the critical values.At last, classification and ranking of the new objects are conducted according to the calculated weights and the critical values of the classification. Case analysis shows the proposed model is effective and feasible. As a result, this model can be applied to multi-attribute classification decision-making problems with abundant objects and uncertain number of classification.

Key words: gray target, interval numbers, classification, sample-set, multi-attribute decision making

中图分类号: