首页 | 本学科首页   官方微博 | 高级检索  
     

基于改进证据理论的群决策专家意见集结方法
引用本文:史超,程咏梅. 基于改进证据理论的群决策专家意见集结方法[J]. 空军工程大学学报(自然科学版), 2011, 0(5): 85-90
作者姓名:史超  程咏梅
作者单位:西北工业大学自动化学院,陕西西安710072
基金项目:国家自然科学基金资助项目(61075029);航空科学基金资助项目(20100853010)
摘    要:针对Demsper组合规则无法有效集结群决策中高冲突专家意见的局限,提出了利用冲突系数和Jousselme距离综合表示证据冲突程度进而通过证据冲突度计算专家客观权重的方法,采用经典的Demsper组合规则对经权重修正后的专家意见进行了集结;提出了一种新的可能度计算公式及基于可能度的信任区间的决策方法。数值实例表明:提出的方法可以有效集结高冲突专家意见,与几种改进方法相比收敛效果更好。基于可能度的信任区间法与最大信任函数法和最大似真函数法相比,决策结果更加科学。

关 键 词:群决策  证据冲突  可能度

The Expertise Aggregation Method Based on the Improved Evidence Theory in Group Decision Making
SHI Chao,CHENG Yong-mei. The Expertise Aggregation Method Based on the Improved Evidence Theory in Group Decision Making[J]. Journal of Air Force Engineering University(Natural Science Edition), 2011, 0(5): 85-90
Authors:SHI Chao  CHENG Yong-mei
Abstract:This paper, considering Demsper rule''s disadvantage in aggregating the expertise of high conflict, presents the evidence conflict degree based on conflict coefficient and Jousselme distance. The objective expert weight is computed by using the evidence conflict degree. Then Demsper rule is used to aggregate the modified expertise. Finally, a new probability degree formula is established and the belief interval method based on probability degree is used for decision-making. The numerical examples illustrate that the proposed method is more efficient in combining the conflict expertise than several improved DS methods and compared with the maximum belief function method and plausibility function method, the belief interval method proposed is more scientific.
Keywords:group decision-making   evidence conflict   probability degree
点击此处可从《空军工程大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《空军工程大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号