AI-Powered Smart Manufacturing Real-Time Monitoring System for Solder Paste Inspection

Authors

  • Syaeful Anas Aklani Universiti Teknikal Melaka Malaysia, Malaysia
  • Halizah Binti Basiron Universiti Teknikal Melaka Malaysia, Malaysia
  • Sek Yong Wee Universiti Teknikal Melaka Malaysia, Malaysia

DOI:

https://doi.org/10.37253/conescintech.v6i1.12763

Keywords:

Solder Paste Inspection, Artificial Intelligence, Anomaly Detection, Smart Manufacturing, Predictive Maintenance

Abstract

Solder paste inspection (SPI) constitutes a critical quality control process in surface mount technology (SMT) assembly lines, yet conventional systems operate reactively without predictive capabilities. This research develops an artificial intelligence-powered real-time monitoring system for SPI machines that addresses fundamental limitations in existing approaches. The proposed system architecture integrates four layers: data acquisition from multiple SPI machines, SQL Server database storage, machine learning analytics, and Flask-based web visualization. The Isolation Forest algorithm achieves 94.2% accuracy for anomaly detection (AUC-ROC 0.963), while Random Forest classification attains 93.1% prediction accuracy at 75% batch completion. Deployed across four KOHYOUNG SPI machines over 30 days, the system analyzed 1,247 production batches comprising 532,819 individual pad measurements. Results demonstrate 52.1% reduction in unplanned downtime, 32.5% decrease in rework costs, and 82.2% improvement in mean time to detect anomalies. The system generates estimated annual savings of USD 50,800 with return on investment period of 4.2 months. This research contributes an open-source reference implementation demonstrating practical application of machine learning for smart manufacturing quality control.

Downloads

Download data is not yet available.

References

Prasad, R. (2018). Surface Mount Technology: Principles and Practice. Springer, 3rd Edition.

Lee, S., & Kim, J. (2020). Analysis of solder paste printing defects in SMT assembly. Journal of Electronics Manufacturing, 28(3), 195-208.

Wang, L., Chen, X., & Zhang, Y. (2019). 3D solder paste inspection using structured light. IEEE Transactions on Components, Packaging and Manufacturing Technology, 9(5), 892-901.

Zhang, H., Li, J., & Wang, F. (2022). Big data analytics for SMT process optimization. Computers in Industry, 134, 103547.

Yamada, T., & Tanaka, M. (2020). Real-time defect detection in solder paste printing. IEEE Transactions on Industrial Informatics, 16(8), 5234-5243.

Park, J., Kim, S., & Lee, H. (2021). Predictive maintenance for SMT assembly equipment. Reliability Engineering & System Safety, 205, 107245.

Kim, D., & Lee, K. (2019). Centralized monitoring system for multiple SPI machines. Journal of Intelligent Manufacturing, 30(4), 1789-1802.

Brown, R., & Wilson, T. (2021). Alert fatigue in manufacturing monitoring systems. Human Factors in Manufacturing, 31(3), 245-258.

Smith, P., & Johnson, M. (2018). Evolution of solder paste inspection technology. Soldering & Surface Mount Technology, 30(2), 89-98.

Anderson, D., Williams, R., & Thompson, K. (2019). Advances in 3D solder paste measurement. Measurement Science and Technology, 30(6), 064005.

Martinez, R., & Garcia, L. (2020). Phase-shift profilometry for solder paste inspection. Optics and Lasers in Engineering, 128, 106028.

Kagermann, H., Wahlster, W., & Helbig, J. (2013). Recommendations for implementing Industrie 4.0. Final report.

Li, Y., Huang, Z., & Chen, X. (2018). Convolutional neural networks for industrial defect detection. Neurocomputing, 275, 1960-1970.

Zhang, Q., & Chen, J. (2019). Deep residual network for solder joint inspection. IEEE Access, 7, 156234-156245.

Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM Computing Surveys, 41(3), 1-58.

Liu, F. T., Ting, K. M., & Zhou, Z. H. (2008). Isolation forest. 2008 Eighth IEEE International Conference on Data Mining, 413-422.

Liu, D., Chen, Y., & Wang, L. (2018). One-class SVM for manufacturing anomaly detection. Control Engineering Practice, 72, 36-45.

Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32.

Patel, H., & Shah, M. (2019). Random forest for CNC tool wear prediction. Procedia Manufacturing, 34, 362-369.

Kumar, A., Singh, P., & Gupta, R. (2020). Ensemble methods for stencil life prediction. Journal of Intelligent Manufacturing, 31(5), 1245-1258.

Taylor, M., Jones, P., & Davies, R. (2021). Integration challenges in smart manufacturing systems. Journal of Manufacturing Technology Management, 32(6), 1234-1251.

Chen, Y., & Wang, X. (2022). Real-time analytics for manufacturing data streams. IEEE Transactions on Industrial Electronics, 69(3), 2876-2885.

Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). Why should I trust you?. Proceedings of the 22nd ACM SIGKDD, 1135-1144.

Williams, A., & Brown, S. (2020). Scalability challenges in industrial AI deployment. AI in Manufacturing, 45-62.

Chen, M., & Liu, W. (2021). Data-driven quality control in electronics assembly. International Journal of Production Research, 59(12), 3678-3695.

Wong, C., Chan, K., & Leung, H. (2021). Multi-frequency pattern projection for fine-pitch inspection. Precision Engineering, 68, 156-167.

Jardine, A. K. S., Lin, D., & Banjevic, D. (2006). A review on machinery diagnostics and prognostics. Mechanical Systems and Signal Processing, 20(7), 1483-1510.

Zuehlke, D. (2010). SmartFactory—Towards a factory-of-things. Annual Reviews in Control, 34(1), 129-138.

Silva, J., Oliveira, L., & Santos, R. (2019). Web-based monitoring for Industry 4.0. International Journal of Computer Integrated Manufacturing, 32(8), 767-781.

Grinberg, M. (2018). Flask Web Development: Developing Web Applications with Python. O'Reilly Media.

Published

2026-09-22