From Sensor Data to Maintenance Actions: An Industrial PHM Application for UltrasonicWelding Assembly Machines
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
Ultrasonic welding machines are widely used in high precision manufacturing processes, where progressive component degradation can lead to quality losses, increased reject rates, and unplanned downtime. Although modern machines generate large volumes of high-frequency process data, such as welding time, amplitude, and pressure, deploying effective predictive maintenance solutions remains challenging due to strong process variability, the lack of explicit failure labels, and the absence of historical run-to-failure datasets. This paper presents an industrial Prognostics and Health Management (PHM) framework developed for a multi-station ultrasonic welding machine used in pharmaceutical assembly, with the objective of enabling early detection of performance degradation and supporting predictive maintenance decisions. The proposed approach focuses on the construction of an interpretable, data-driven healthindicator at component level, derived from sensor data and explicitly designed to be understandable by machine experts. Domain knowledge provided by the machine manufacturer is integrated to interpret the health indicator evolution and to translate detected degradation patterns into concrete maintenance recommendations. A healthy reference behavior is established using data from machines operating under stable conditions, enabling relative deviation analysis and trend-based monitoring across heterogeneous stations. The framework was deployed in an industrial pilot and demonstrated the ability to identify abnormal behaviors associated with component wear and process sensitivity, including cases where conventional maintenance actions showed limited effectiveness. The results indicate that the proposed indicators can reveal degradation patterns earlier than traditional reject-rate monitoring, thereby supporting maintenance prioritization at component level. This work illustrates how interpretable, data-driven PHM methodologies, co-designed with machine experts, can be successfully integrated into real manufacturing environments, bridging the gap between raw process data and actionable maintenance insights.
How to Cite
##plugins.themes.bootstrap3.article.details##
PHM, Industrial Machines, Manufacturing, Exepert Knowledge
Calabrese, F., Regattieri, A., Botti, L., & Galizia, F. G. (2019). Prognostic health management of production systems: New proposed approach and experimental evidences. Procedia Manufacturing, 39, 260–269. doi: 10.1016/j.promfg.2020.01.333
Faes, K., et al. (2025). The influence of welding parameters on the performance of ultrasonic copper welds. Journal of Manufacturing and Materials Processing, 9(2), 55. doi: 10.3390/jmmp9020055
Fernandes, M., Corchado, J. M., & Marreiros, G. (2022). Machine learning techniques applied to mechanical fault diagnosis and fault prognosis in the context of real industrial manufacturing use-cases: A systematic literature review. Applied Intelligence. doi: 10.1007/s10489-022-03344-3
Kim, S., Lee, S., Kim, M., Yoon, H., & Youn, B. D. (2025). Interpretable sensor importance-based multisensor integration for condition monitoring of rotating machinery. In Proceedings of the Asia Pacific Conference of the PHM Society (Vol. 5). doi: 10.36001/phmap.2025.v5i1.4624
Lei, Y., Yang, B., Jiang, X., Jia, F., Li, N., & Nandi, A. K. (2020). Applications of machine learning to machine fault diagnosis: A review and roadmap. Mechanical Systems and Signal Processing, 138, 106587. doi: 10.1016/j.ymssp.2019.106587
Li, Z., & Gryllias, K. (2024). Remaining useful life prognostics of rolling element bearings based on state estimation techniques. In PHM Society European Conference 2024.
Lukens, S., & Markham, M. (2018). Data-driven application of PHM to asset strategies. In Annual Conference of the Prognostics and Health Management Society. Philadelphia, PA, USA.
Neyret Group. (2023). Continuous motion assembly machines. Retrieved from https://neyretgroup.com/en/solutions/continuousmotion-assembly-machines/. Accessed: 2026-01-20.
Oviedo, A. I., et al. (2025). Unsupervised real-time anomaly detection in industrial time series using clustering and autoencoder-based models. Technologies, 13(11), 534. doi: 10.3390/technologies13110534
Polverino, L., Abbate, R., Manco, P., Perfetto, D., Caputo, F., Macchiaroli, R., & Caterino, M. (2023). Machine learning for prognostics and health management of industrial mechanical systems and equipment: A systematic literature review. International Journal of Engineering Business Management, 15. doi: 10.1177/18479790231186848
Silva, J. J., da Silva, R. G. C., Morelli, C. L., Lopez, E. A. T., & Santos, T. F. A. (2025). Trends and future projections in ultrasonic welding research for hybrid materials. Polymers, 17(8), 1124. doi: 10.3390/polym17081124
Tian, B., Lu, K.-C., Eslaminia, A., Wang, Y., Shao, C., & Nahrstedt, K. (2023). WeldMon: A cost-effective ultrasonic welding machine condition monitoring system. In 2023 IEEE 14th Annual Ubiquitous Computing, Electronics and Mobile Communication Conference (UEMCON).
Zio, E. (2022). Prognostics and health management (PHM): Where are we and where do we (need to) go in theory and practice. Reliability Engineering & System Safety, 218, 108119. doi: 10.1016/j.ress.2021.108119
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.