摘要
针对风险评估中群决策数据的处理可靠性问题,提出专家权重对于数据处理的重要性。从汽车碰撞安全研究的行业特色出发,依据专家社会因素指标,运用系统聚类和K-均值聚类分析方法,对专家进行专家权威权重聚类;依据专家提供意见的数据一致性和个体与群体一致性程度,运用基于信息熵的方法对专家进行专家意见权重聚类;将两种聚类结果进行复合聚类,得到各位专家的最终权重。结果表明,在综合专家权威权重和专家意见权重的基础上,经复合聚类得到的专家权重具有较好的均衡性和可靠性,为系统项目的研究奠定基础。
Aiming at data-processing reliability problems in risk-assessment group decision-making,the importance of expert-weight for data processing was presented.Starting from the industrial characteristics of automobile-crash-safety study,according to expert social factor indicators,by using two analysis methods of system clustering and K-means clustering,experts were clustered using expert-authority-weights.According to the data consistency of expert＇s opinions and the consistency level between an individual and a group,by using the method based on information entropy,experts were clustered with expert-opinion-weights.The two clustering results were compositely clustered to get the final weight of each expert.The results showed that the expert-weights obtained with the composite clustering have good proportionality and reliability based on compositing expert-authority-weights and expert-opinion-weights,the finally obtained expert-weights provides a foundation for the study of systems projects.
出处
《振动与冲击》
EI
CSCD
北大核心
2014年第7期73-78,85,共7页
Journal of Vibration and Shock
基金
长沙理工大学道路灾变防治及交通安全教育部工程研究中心开放基金资助项目(kfj080301)
国家自然科学基金(51175050)
关键词
汽车碰撞安全
风险评估
群决策
社会因素指标
复合聚类
专家权重
automobile-crash-safety
risk assessment
group decision-making
social factor indicators
composite clustering
expert-weight
