基于改进随机森林的半自动化产线制程能力评估研究  

Research on Process Capability Evaluation of Semi-Automated Production Lines Based on Improved Random Forest

作　　者：张放 ZHANG Fang(Hitachi Elevator Motor(Guangzhou)Co.,Ltd.,Guangzhou,Guangdong 510670,China)

机构地区：[1]日立电梯电机(广州)有限公司,广东广州510670

出　　处：《自动化应用》2024年第23期25-27,30,共4页Automation Application

摘　　要：半自动化产线制程能力关系产线生产质量,评估该项能力是半自动化产生企业生产前的重要环节,同时也是线程技术调整和优化的依据,为此,提出基于改进随机森林的半自动化产线制程能力评估的研究。选择产线质量因子、制程性能指数作为评估指标,通过加权抽取样本代替投票方式抽样实现对随机森林的改进,利用改进随机森林构建评估模型,通过多个决策树综合决策分类,实现对半自动化产线制程能力定量和定性评估。通过实例应用的验证,所设计的半自动化产线制程能力评估方法显示出合理性,并展现出良好的应用前景。The process capability of semi-automated production lines is related to the production quality of the production line.The evaluation of this capability is an important link before the production of semi-automated production enterprises,and it is also the basis for adjusting and optimizing thread technology.Therefore,this paper proposes a research on process capability evaluation of semi-automated production lines based on improved random forest.Selecting the production line quality factor and process performance index as evaluation indicators,improving the random forest through weighted sampling instead of voting sampling,constructing an evaluation model using the improved random forest,and achieving quantitative and qualitative evaluation of the semi-automated production line process capability through multiple decision trees for comprehensive decision classification.Through the verification of practical applications,the semi-automated production line process capability evaluation method designed in this study demonstrates its rationality and demonstrates good application prospects.

关 键 词：改进随机森林 半自动化产线 制程能力 投票方式 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]