电子行业全生命周期产品质量智能感知、监测、计算和可解释预测数据
收藏资源简介：
电子行业全生命周期产品质量智能感知、监测、计算和可解释预测数据主要面向产品质量全生命周期多源异构数据集成与计算技术研究，主要记录了基于工业互联网的全生命周期质量数据的感知体系架构和多源异构数据的实时高精度感知技术，实现供应数据可视化、生产数据可视化过程中涉及到的设备位置数据、厂商位置数据、生产产线检测数据、订单数据、新品过程数据和风险管理数据；电子产品研发、测试、生产、售后、质量智能监测模型、动态追踪、根因分析过程中涉及的产品阶段性研发数据，原材料及生产数据，产品层级数据，供应商及产线数据；实现供应业务过程挖掘、生产业务过程挖掘和多源工业大数据融合过程中涉及到的电子产品多源异构、全生命周期结构化标准数据：产品数据、物料数据、生产设备基础数据、生产设备备件数据、制造过程数据；基于面向设计、制造、售后的数据清洗与选择方法和质量数据字典的统一数据标准，实现对全价值链质量关联数据的实时采集与处理的过程中涉及到的多种感知传感器工艺数据和通过端边物联网传输后清洗的质量关联结果数据；基于人工智能技术的结构化、半结构化和非结构化数据的高效融合方法，通过工业互联网产品质量多模态数据采集，构建融合多端的数字化质量信息计算平台，实现售后数据聚类、质量趋势分析和市场趋势可视化过程中涉及到的半结构化生产设备日志数据；基于多粒度情感分析的自动化标记方法，实现售后数据情感分析过程中涉及到的情感分析语义定义数据和情感分析过程数据。其中，电子产品多源异构、跨全生命周期结构化标准数据来源为供应链生产厂家产线中的相关IoT设备以及联想对应的数据中台，半结构化生产设备日志数据来源为主流互联网的产品信息数据、售后数据等，情感分析语义定义和过程数据来源为中关村在线（https://www.zol.com.cn/)等网站。合计数据量约为661MB。
This dataset focuses on intelligent perception, monitoring, computing, and interpretable prediction data for product quality across the entire life cycle of the electronics industry, targeting research on multi-source heterogeneous data integration and computing technologies for full-life-cycle product quality. It mainly records the perception system architecture of full-life-cycle quality data based on the industrial internet and real-time high-precision perception technologies for multi-source heterogeneous data, involving equipment location data, manufacturer location data, production line inspection data, order data, new product process data, and risk management data involved in supply data visualization and production data visualization; It covers product phased R&D data, raw material and production data, product hierarchy data, and supplier and production line data involved in the processes of R&D, testing, production, after-sales service, intelligent quality monitoring models, dynamic tracking, and root cause analysis of electronic products; It includes structured standard data covering multi-source heterogeneous and full-life-cycle electronic products involved in supply business process mining, production business process mining, and multi-source industrial big data integration: product data, material data, basic production equipment data, production equipment spare parts data, and manufacturing process data; Based on unified data standards including data cleaning and selection methods for design, manufacturing, and after-sales scenarios as well as quality data dictionaries, it involves various process data from perception sensors and cleaned quality-related result data transmitted via edge-end IoT, which are involved in the real-time collection and processing of quality-related data across the entire value chain; Using efficient fusion methods for structured, semi-structured, and unstructured data based on artificial intelligence technologies, it constructs a multi-terminal integrated digital quality information computing platform through multimodal data collection of product quality via the industrial internet, involving semi-structured production equipment log data involved in after-sales data clustering, quality trend analysis, and market trend visualization; Using automated annotation methods based on multi-granularity sentiment analysis, it covers sentiment analysis semantic definition data and sentiment analysis process data involved in the sentiment analysis of after-sales data. Specifically, the structured standard data of multi-source heterogeneous and cross-full-life-cycle electronic products is sourced from relevant IoT equipment in the production lines of supply chain manufacturers and Lenovo's corresponding data platform; the semi-structured production equipment log data is sourced from mainstream internet product information data, after-sales data, etc.; the sentiment analysis semantic definition and process data are sourced from websites such as ZOL (https://www.zol.com.cn/). The total data volume of this dataset is approximately 661 MB.
