基于数据驱动的潮流玩具产品质量管理与控制研究  

Research on Data-driven Quality Management and Control of Trendy Toy Products

作　　者：吕远智 LYU Yuanzhi(Beijing Paopao Mart Cultural and Creative Co.,Ltd.,Beijing 100123,China)

机构地区：[1]北京泡泡玛特文化创意有限公司,北京100123

出　　处：《移动信息》2024年第7期299-301,共3页MOBILE INFORMATION

摘　　要：针对潮流玩具产品质量管控困难、产品质量低的问题,文中提出了利用数据驱动对其进行智能管理及优化控制的方法。研究利用卷积神经网络提取质量数据中的深层次特征,然后利用长短时记忆网络对产品质量进行预测,根据预测结果将产品分为不合格产品与合格产品。同时,采用关联规则挖掘的方法建立规则库,根据规则库与不合格产品进行质量诊断,并根据诊断结果对产品生产参数进行优化、调整。实验结果表明,使用该质量管控方法后,不同种类的潮流玩具产品的生产不合格率下降了3.5%以上。由此可知,该方法能在使用过程中优化提升产品质量,进一步提高产品的生产效率。Aiming at the problems of difficult quality control and low product quality of trendy toy products,this paper proposes a method of intelligent management and optimal control using data-driven.This paper studies the use of convolutional neural networks to extract deep-level features in quality data,and then uses long and short-term memory networks to predict product quality,and classifies products into substandard products and qualified products according to the prediction results.At the same time,the method of association rule mining is used to establish a rule base,and the quality diagnosis is carried out according to the rule base and substandard products,and the production parameters of products are optimized and adjusted according to the diagnosis results.The experimental results show that after using the quality control method,the production failure rate of different types of trendy toy products decreases by more than 3.5%.It can be seen that this method can optimize and improve product quality during use and further improve product production efficiency.

关 键 词：潮流玩具 质量管控 质量预测 关联规则 长短时记忆网络 

分 类 号：TP321[自动化与计算机技术—计算机系统结构]