摘要
为减少模块划分过程中专家打分等量化方法主观性太强对结果的影响,提供一条从用户需求出发的设计路径,将博弈论和模糊聚类算法结合,设计一套客观的、贴近客户实际需求的模块划分方法和方案评价体系;采用合作博弈中的核与Sharpley Value等理念从产品数据中提炼出设计所需的参数,建立关联矩阵;然后采用改进模糊聚类算法生成模块划分的动态聚类树;接着采用合作博弈中的核仁解等理念对各种模块划分方案进行评价,得到最优方案。最后以消费类机电产品中的洗衣机为例,验证方法的可行性和有效性。
In order to reduce the impact of the subjectivity of quantitative methods such as expert scoring on the results,and provides a design path from the user's needs,this paper combines the game theory and fuzzy c-means algorithm,and designs a set of objective module division method and scheme evaluation system close to the actual needs of customers;The concept of Corel and Sharpley Value in cooperative game is used to extract the design parameters from the product data,and the incidence matrix is established;then the improved fuzzy clustering algorithm is used to generate the dynamic clustering tree of module divison;then the concept of Nucleolus solution in cooperative game is used to evaluate various module divison schemes,and the optimal scheme is obtained.Finally,taking the washing machine of consumer electromechanical products as an example,the feasibility and effectiveness of the method are verified.
作者
丁翔宇
刘丹
DING Xiang-yu;LIU Dan(Key Laboratory of Advanced Manufacturing Technology of the Ministry of Education,Guizhou University,Guiyang 550025,China)
出处
《组合机床与自动化加工技术》
北大核心
2021年第12期160-164,共5页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金资助项目(72061006)
黔科合平台-JXCX[2021]001。
