摘要
为了对客户资源进行评价、分类和有效管理,建立了面向钢铁企业直接客户的客户价值评价指标体系。针对自组织映射神经网络聚类算法在权值初始化方面的不足,提出了一种基于K-means改进的自组织映射的聚类算法。运用层次分析法和基于K-means改进的自组织映射聚类算法,设计了一套客户分类方法。通过对可能出现的客户类型进行分析,提出了相应的客户关系发展建议。最后通过应用实例,验证了该评价指标体系和分类方法的可行性和有效性。
In order to evaluate, classify and manage customer resources effectively, an evaluation index system for the steel enterprise's direct customer was established. In order to overcome the shortcomings of weight initialization of self-organizing map (SOM), a new classification method of K-SOM combining K-means algorithm with SOM was proposed. Then a customer classification method based on analytic hierarchy process (AHP) and K-SOM was provided. The characteristics of each type were analyzed and a set of suggestions on customer management was presented for different types of customer. Finally, a case study was provided to demonstrate the practicality and efficiency of the evaluation index system and methods.
出处
《计算机集成制造系统》
EI
CSCD
北大核心
2009年第8期1650-1655,共6页
Computer Integrated Manufacturing Systems
基金
国家自然科学基金资助项目(70572098)~~