Enterprise Machine Learning Model Lifecycle Management: A Seven-Phase Framework for Production-Grade MLOps
Keywords:
MLOps, Machine Learning Lifecycle, Feature Store, Model Monitoring, Concept Drift, Continuous Training, Model Governance, Enterprise AI, AWS SageMaker, Azure Machine Learning.Abstract
Enterprise machine learning adoption has been happening at a pace that, consistently, far outstrips operational infrastructure to support it. Industry surveys between 2020 and 2024 show that between 78 and 87 percent of enterprise ML models never make it to production, a number that has been remarkably steady in the face of the rapid upskilling of algorithms and cloud economics. The production gap isn't caused by the mathematics of ML itself but by the ML model lifecycle, where ML models progress from data acquisition to training, validation, testing, deployment, monitoring, and finally, retirement. In the absence of an ML lifecycle management framework, ML models quickly decay in production. This can result in the considerable erosion of trust in investment in artificial intelligence․ We propose a seven-phase ML lifecycle management framework based on cloud-based enterprise platform architectural patterns of Amazon Web Services and the Microsoft Azure cloud platform. It evaluates the platform capabilities‚ governance controls‚ and operational automations needed for each stage to operate at scale reliably․ That comprises controlled feature stores, experiment tracking infrastructure, progressive deployment patterns and drift detection technology. It discusses MLOps lifecycle maturity governance and legislation in the context of the European Union Artificial Intelligence Act and United States federal AI policy. The framework provides a practical architecture reference point to enterprise engineers and technology leaders to close the divide and support sustainable production value from machine learning investments․
Downloads
References
D. Kreuzberger, N. Kühl, and S. Hirschl, “Machine learning operations (MLOps): overview, definition, and architecture,” IEEE Access, vol. 11, pp. 31866–31879, 2023. [Online]. Available: https://ieeexplore.ieee.org/document/10081336/
S. Amershi et al., “Software engineering for machine learning: a case study,” in Proc. IEEE/ACM 41st Int. Conf. Software Engineering: Software Engineering in Practice (ICSE-SEIP), 2019, pp. 291–300. [Online]. Available: https://ieeexplore.ieee.org/document/8804457/
D. Baylor et al., “TFX: a TensorFlow-based production-scale machine learning platform,” in Proc. 23rd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 2017, pp. 1387–1395. [Online]. Available: https://dl.acm.org/doi/10.1145/3097983.3098021
D. Sculley et al., “Hidden technical debt in machine learning systems,” in Proc. 29th Int. Conf. Neural Information Processing Systems (NeurIPS), 2015, pp. 2503–2511. [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2015/hash/86df7dcfd896fcaf2674f757a2463eba-Abstract.html
M. Mitchell et al., “Model cards for model reporting,” in Proc. ACM Conf. Fairness, Accountability, and Transparency (FAccT), 2019, pp. 220–229. [Online]. Available: https://dl.acm.org/doi/10.1145/3287560.3287596
A. Liu et al., “Concept drift detection delay index,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 5, pp. 4585–4597, 2023. [Online]. Available: https://ieeexplore.ieee.org/document/9729470/
P. Li et al., “High-dimensional multilabel data stream classification with concept drifting detection,” IEEE Trans. Knowl. Data Eng., vol. 35, no. 8, pp. 8085–8099, 2023. [Online]. Available: https://doi.org/10.1109/TKDE.2022.3200068
Y. Lu et al., “Learning under concept drift: a review,” IEEE Trans. Knowl. Data Eng., vol. 31, no. 12, pp. 2346–2363, 2019. [Online]. Available: https://doi.org/10.1109/TKDE.2018.2876857
P. Ruf, M. Madan, C. Reich, and D. Ould-Abdeslam, “Demystifying MLOps and presenting a recipe for the selection of open-source tools,” Applied Sciences, vol. 11, no. 19, p. 8861, 2021. [Online]. Available: https://www.mdpi.com/2076-3417/11/19/8861
M. Zaharia et al., “Accelerating the machine learning lifecycle with MLflow,” IEEE Data Eng. Bull., vol. 41, no. 4, pp. 39–45, 2018. [Online]. Available: http://sites.computer.org/debull/A18dec/p39.pdf
M. Zaharia, A. Ghodsi, R. Xin, and M. Armbrust, “Lakehouse: a new generation of open platforms that unify data warehousing and advanced analytics,” in Proc. 11th Conf. Innovative Data Systems Research (CIDR), 2021. [Online]. Available: http://cidrdb.org/cidr2021/papers/cidr2021_paper17.pdf
European Parliament and Council of the European Union, “Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act),” Official Journal of the European Union, 2024. [Online]. Available: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32024R1689
A. Paleyes, R.-G. Urma, and N. D. Lawrence, “Challenges in deploying machine learning: a survey of case studies,” ACM Comput. Surv., vol. 55, no. 6, pp. 1–29, 2023. [Online]. Available: https://dl.acm.org/doi/10.1145/3533378
E. Breck et al., “Data validation for machine learning,” in Proc. Machine Learning and Systems (MLSys), 2019. [Online]. Available: https://proceedings.mlsys.org/paper_files/paper/2019/hash/928f1160e52192e3e0017fb63ab65391-Abstract.html
K. Vanhaesebrouck et al., “On continuous integration/continuous delivery for automated deployment of machine learning models using MLOps,” in Proc. IEEE Int. Conf. Emerging Technologies and Factory Automation (ETFA), 2021. [Online]. Available: https://ieeexplore.ieee.org/document/9723793/
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
