基于K-means^(++)的制造业服务流程模块划分  

Manufacturing Service Process Module Division Based on K-means^(++)

作　　者：张雪 耿秀丽[1] 金运婷 ZHANG Xue;GENG Xiuli;JIN Yunting(usiness School,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学管理学院,上海200093

出　　处：《机械设计与研究》2021年第3期154-160,共7页Machine Design And Research

基　　金：国家自然科学基金资助项目(71301104);高等学校博士学科点专项科研基金资助课题(2013312012002);上海市高原学科资助项目(GYXK1201)。

摘　　要：针对日益复杂产品延伸服务的客户个性化设计问题,提出了基于K-means^(++)的制造业服务流程模块划分方法。首先,分析各服务活动间的相关性得出综合相关度矩阵,通过K-means^(++)算法获得不同的模块化方案。其次,分别以数值和概率语言的形式给出定量和定性评价指标下的各模块化方案评估结果,并引入FlowSort多属性决策方法进行方案优选。最后,应用案例表明所提方法可考虑产品延伸服务间的关联关系得出合理的流程模块化方案。以上研究有助于揭示制造业产品延伸服务流程中的关联关系及整体特性,并通过科学的流程模块化方案实现简化复杂流程、快速响应客户需求的效果。In order to solve the problem of customer personalized design of increasingly complex product extension service,this paper proposes a service process module partition method based on K-means^(++).Firstly,the correlation between each service activity is analyzed to obtain the comprehensive correlation matrix,and different modularization schemes are obtained through K-means^(++)algorithm.Secondly,the evaluation results of each modular scheme under quantitative and qualitative evaluation indexes are given in the form of numerical and probability language respectively,and the FlowSort multi-attribute decision making method is introduced to optimize the scheme.Finally,the application case shows that the proposed method can consider the correlation between product extension services to get a reasonable process modularization scheme.The above research is helpful to reveal the correlation relationship and the overall characteristics of the manufacturing product extension service process,and realize the effect of simplifying the complex process and responding quickly to the customer demand through the scientific process modularization scheme.

关 键 词：服务流程模块化 K-means^(++) 概率语言术语集 FlowSort 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]