摘要
为解决传统层次分析法在权重分配过程中主观性过强的缺陷,改进层次分析法,提出基于信度系数的人机界面评价指标权重分配方法.根据人类思维具有模糊性和灰色性的特点,将灰色关联分析引入到层次分析法中,从整体相似性的角度构造专家判断的信度系数,解决了多专家评判信息的集结以及不确定信息的处理,实现了认知特性的定量描述;将其与指标权重合成,获得了符合认知特性且更加科学可靠的综合权重分配.通过算例分析验证了该方法的可行性.
In order to avoid the disadvantage of excessive subjectivity in the weitht assignment of traditional analytic hierarchy process （AHP） and improve the AHP, a method for weight assignment of evaluation indexes in human-machine interface based on reliability coefficient was proposed. Based on both vagueness and grey characteristics of human thought, the grey correlation analysis was introduced into the AHP. The reliability coefficient of expert judgements was established in terms of the integrated similarity. And thus, the information aggregation of multi-expert evaluation and disposal of uncertain information were resolved, and the quantitative description for the cognitive characteristics was realized. Through synthetizing the reliability coefficient with the index weight, the more scientific and reliable comprehensive weight assignment in accordance with cognitive characteristics was obtained. The feasibility of the proposed method is proved with the example analysis.
出处
《沈阳工业大学学报》
EI
CAS
北大核心
2013年第1期53-57,共5页
Journal of Shenyang University of Technology
基金
国家自然科学基金资助项目(10871223)
黑龙江省教育厅科学技术研究资助项目(12511391)
佳木斯大学科研资助项目(L2012-001)
关键词
信度系数
灰色关联分析
层次分析法
人机界面
信息集结
相似性
认知特性
reliability coefficient
grey correlation analysis
AHP
human-machine interface
information aggregation
similarity
cognitive characteristic
