摘要
分析了各种因素对工序质量损失的影响,提出了工序质量损失率概念,建立了基于决策树、K-均值聚类和基尼指数相结合的工序质量损失原因挖掘决策树模型,将模型应用于工序质量损失数据集市训练集样本,挖掘质量损失原因。利用获取的知识规则对测试集进行诊断验证,得到的结果具有很高的真实性和有效性。为降低质量损失、持续改进生产质量提供了决策支持。
In order to analyze the causes of quality loss in processing, a new concept of quality loss ratio in processing is presented and a decision-tree model of knowledge mining is established based on the solution which integrates decision-tree, K-average value clustering and Gini-index. The developed model is applied to knowledge mining analysis of the causes of quality loss in processing based on the sample of quality loss in processing. The developed knowledge acquisition is validated by tests, and the results are proved to be valid.
出处
《机械科学与技术》
CSCD
北大核心
2009年第10期1353-1358,共6页
Mechanical Science and Technology for Aerospace Engineering
基金
国家863高技术研究发展计划项目(2007AA04Z187)
国家自然科学基金项目(50705076
50705077)资助
关键词
工序质量损失率
决策树
基尼指数
K-聚类
quality loss ratio in processing
decision tree
gini-index
K-means clustering
