Yimei Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yimei Xu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Hairong Hao
School of Information Engineering, Eurasia University, Xi' an, China
1 shared publication - Hongfei Hu
School of Information Engineering, Xi'an Eurasia University, Xi'an, Shaanxi, China
1 shared publication - Jiarui Fan
A School of Information Engineering, Xi'an Eurasia University, China
1 shared publication - Ming Chen
School of Information Engineering, Xi'an Eurasia University, Xi'an, Shaanxi, China
1 shared publication - Pengju He
School of Information Engineering, Xi'an Eurasia University, Xi'an, Shaanxi, China
1 shared publication
Latest Publications
- Open Access
- 24 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 235-240
Design and Implementation of a HarmonyOS-Based Student Mental Health Management System
- 24 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 235-240
Open AccessDesign and Implementation of a HarmonyOS-Based Student Mental Health Management System
- Open Access
- 26 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 170-175
Application and Optimization of Deep Learning-Powered Intelligent Andon System in Lean Manufacturing
The current manufacturing industry urgently needs to improve quality and efficiency while achieving lean and intelligent upgrades. Traditional Andon systems often fall short due to their slow response times and strong reliance on human judgment. To address this, this study proposes a framework model for an intelligent Andon system based on deep learning and the fusion of heterogeneous data sources. This model utilizes industrial cameras and sensors to collect data related to quality, equipment, and environmental conditions, and employs convolutional neural networks for proactive sensing and fault prediction. It guides quality control and system optimization in the manufacturing process. The research shows that compared to traditional manual intervention models, the new system can reduce downtime response time by more than 60%, decrease spare parts inventory by 30%, and increase product yield by approximately 8 percentage points. This achievement not only enriches the theoretical foundation of Andon systems but also provides a technological path for the upgrading of the manufacturing industry, offering both theoretical and practical value. Future efforts should focus on accumulating industrial-grade data and algorithms to continuously strengthen system effectiveness and expand application scenarios.
- 26 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 170-175
Open AccessApplication and Optimization of Deep Learning-Powered Intelligent Andon System in Lean Manufacturing
The current manufacturing industry urgently needs to improve quality and efficiency while achieving lean and intelligent upgrades. Traditional Andon systems often fall short due to their slow response times and strong reliance on human judgment. To address this, this study proposes a framework model for an intelligent Andon system based on deep learning and the fusion of heterogeneous data sources. This model utilizes industrial cameras and sensors to collect data related to quality, equipment, and environmental conditions, and employs convolutional neural networks for proactive sensing and fault prediction. It guides quality control and system optimization in the manufacturing process. The research shows that compared to traditional manual intervention models, the new system can reduce downtime response time by more than 60%, decrease spare parts inventory by 30%, and increase product yield by approximately 8 percentage points. This achievement not only enriches the theoretical foundation of Andon systems but also provides a technological path for the upgrading of the manufacturing industry, offering both theoretical and practical value. Future efforts should focus on accumulating industrial-grade data and algorithms to continuously strengthen system effectiveness and expand application scenarios.