A Context-Aware Edge-Cloud Multi‑Agent PHM Framework for Multi‑Tool CNC Turning Machines Using a Single Vibration Sensor
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Abstract
CNC turning machines operate under highly variable and context-dependent conditions, where multiple cutting tools and machine subsystems share a common structural vibration path. This creates a practical monitoring challenge: a single sensor captures a superposed vibration response, while tool wear and machine degradation evolve differently across tools, operations, and cutting conditions. This paper presents a context-aware edge-cloud multi-agent Prognostics and Health Management (PHM) framework for multi-tool CNC turning machines using a single spindle-mounted vibration sensor. The framework assigns a logical PHM agent to each tool, while a context router uses controller-side process information, such as tool identity, spindle speed, feed rate, and depth of cut, to transform the shared vibration stream into tool-specific feature streams. Each agent performs in situ baseline calibration, health indicator (HI) estimation, monitoring mode selection, online degradation tracking, and, where applicable, Remaining Useful Life (RUL) estimation. To improve deployment robustness, the library includes adaptive threshold estimation, online threshold refinement, and explicit handling of anomaly-dominant and degradation-dominant operating regimes. The framework is implemented as an edge-cloud architecture in which the edge performs feature extraction, context-aware routing, HI computation, and alert generation, while the cloud ingests telemetry and curated feature streams for visualization, storage, and lifecycle management. The study includes an industrial CNC case study with three tool types and an auxiliary evaluation on the UC Berkeley milling dataset. Results show that the framework can isolate tool-specific degradation behavior using a shared sensor, provide early warnings for rapidly degrading tools, and support low-latency edge deployment.
How to Cite
##plugins.themes.bootstrap3.article.details##
Prognostics and Health Management, CNC turning, tool wear monitoring, edge AI, multi-agent monitoring, context aware feature fusion, health indicator, remaining useful life
Agogino, A., & Goebel, K. (2007). Milling data set. NASA Ames Prognostics Data Repository. Moffett Field, CA.
Calabrese, F., Regattieri, A., Bortolini, M., Gamberi, M., & Pilati, F. (2021). Predictive maintenance: A novel framework for a data-driven, semi-supervised, and partially online prognostic health management application in industries. Applied Sciences, 11(8). doi: 10.3390/app11083380
Costa, V. L. L., Eberhardt, B., Chen, J., & Roßmann, J. (2023). Towards predictive maintenance: An edge-based vibration monitoring system in Industry 4.0. In 2023 International Conference on Sustainable Computing and Data Communication Systems (ICSCDS) (pp. 1430–1437).
Dan, L., & Mathew, J. (1990). Tool wear and failure monitoring techniques for turning: A review. International Journal of Machine Tools and Manufacture, 30(4), 579–598. doi: https://doi.org/10.1016/0890-6955(90)90009-8
Gaikwad, P., Mundhada, V., Nagre, H., & Patil, H. (2024). Bearing fault detection using vibration analysis. International Journal for Research in Applied Science and Engineering Technology, 12, 4644–4653. doi: 10.22214/ijraset.2024.62569
Hassan, I. U., Panduru, K., & Walsh, J. (2024). An in-depth study of vibration sensors for condition monitoring. Sensors, 24(3). doi: 10.3390/s24030740
Huang, Z., Zhu, J., Lei, J., Li, X., & Tian, F. (2021). Tool wear monitoring with vibration signals based on short-time Fourier transform and deep convolutional neural network in milling. Mathematical Problems in Engineering, 2021(1), 9976939. doi: https://doi.org/10.1155/2021/9976939
Krishnamurthy, R., S, S., Sreedhar, R., & Chandrashekar, A. (2025). IntelliMaint: An intelligent component-agnostic framework for health indicator generation and prognostics for electromechanical systems. Annual Conference of the PHM Society, 17. doi: 10.36001/phmconf.2025.v17i1.4380
Li, X. (2002). A brief review: Acoustic emission method for tool wear monitoring during turning. International Journal of Machine Tools and Manufacture, 42(2), 157–165. doi: https://doi.org/10.1016/S0890-6955(01)00108-0
Li, Y., Meng, X., Zhang, Z., & Song, G. (2020). A machining state-based approach to tool remaining useful life adaptive prediction. Sensors, 20(23). doi: 10.3390/s20236975
Munaro, R., Attanasio, A., & Del Prete, A. (2023). Tool wear monitoring with artificial intelligence methods: A review. Journal of Manufacturing and Materials Processing, 7(4). doi: 10.3390/jmmp7040129
Resende, C., Folgado, D., Oliveira, J., Franco, B., Moreira, W., Oliveira-Jr, A., ... Carvalho, R. (2021). TIP4.0: Industrial Internet of Things platform for predictive maintenance. Sensors, 21(14). doi: 10.3390/s21144676
Salvador Palau, A., & Dhada, M. (2019). Multi-agent system architectures for collaborative prognostics. Journal of Intelligent Manufacturing, 30. doi: 10.1007/s10845-019-01478-9
Wang, K., Wang, A., Wu, L., & Xie, G. (2024). Machine tool wear prediction technology based on multi-sensor information fusion. Sensors, 24. Retrieved from https://api.semanticscholar.org/CorpusID:269311267
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.