基于决策树C4.5算法的制造过程质量管理

Manufacturing Process Quality Management Based on Decision Tree C4.5 Algorithm

针对大多数制造业生产过程中对产品质量只注重问题处理,忽略挖掘问题潜在原因,造成质量分析滞后、质量信息利用不充分的问题,引入事先预警的理念,利用决策树C4. 5算法,对大量生产加工数据进行挖掘,建立制造质量分析模型,运用该模型挖掘质量数据中潜在的质量影响因素,找到影响生产的关键因素。文章以滚齿加工过程累积的质量数据来训练检测模型,试验表明,所建模型对数据分类准确率高,满足企业需求,为企业持续改进质量提供决策支持,降低了生产的不合格率。

For the product quality management of most manufacturing industries only focus on problem solving,ignoring to find potential reasons,which results in the problem of the lag of quality analysis and inadequate utilization of quality information.This paper introduced the concept of early warning in advance,using decision tree C4.5algorithm to mine the mass data of production processing,establishing a production process quality analysis model,using this model to mine the quality data for potential affecting factors,find the key factors affecting production.The experiment shows that the model has high accuracy in classification of data,and provides decision support for continuous improvement of the quality of enterprises,and reduces the rate of disqualification of production.