为提高工业设计方案评价结果的可信度,建立了评价群体共识度模型,引入围绕中心点的划分聚类算法处理设计方案评价中的“少数意见”,通过形成意见簇促进评价群体对设计方案的认知沟通,以加速意见收敛.研究了围绕中心点的划分聚类原理,提出基于围绕中心点的划分聚类的工业设计方案评价共识度达成流程,以户外检测车的设计方案评价为例,验证了该方法能够在共识度不一致时,通过围绕中心点的划分聚类识别“少数意见”,促进评价群体间的交流沟通,实现共识度的达成.最后,通过与K-均值算法的聚类效果比较,显示了该方法的优越性.
To improve the reliability of results in industrial design evaluation,a consensus model was established and Partitioning Around Medoids (PAM) clustering algorithm was applied to deal with minority opinions.With PAM,opinions were clustered into several groups and minority opinions were extracted for discussion and communication,which would accelerate convergence of opinions in design evaluation.By researching principles of PAM clustering algorithms,a process to reach consensus in evaluation of industrial design schemes was proposed and outdoor detection equipment was taken as an example to verify the validity of this method.Results reflected that when consensus did not be reached,PAM clustering would help for recognizing minority opinions to promote exchange of ideas and communication,resulting in consensus reaching.The method also showed the superiority compared with K-means clustering.