Intelligent Framework Based on Machine Learning and Model-Driven Architecture for Optimization of Industrial IoT Systems

Abstract :

By increasing proliferation of IoT devices, the large volume of unlabeled data, and the need for fast processing and optimal resource management, new challenges have emerged in this field. In the meantime, the importance of semi-supervised methods for prediction and also for extracting key domain features to improve accuracy has become more important. In this research, a new approach to improving the efficiency of industrial IoT systems is presented, in which machine learning is enhanced by using Model-Driven Architecture. In this framework, domain-oriented conceptual models are used as the basis for extracting key features, structuring data, and reducing noise. This integration improves the machine learning process, especially in conditions of lack of labeled data. In the first step, PIM and PSM models are created to separate the system logic from the implementation details and adapt to technological changes. Then, the processed data and extracted features are fed into the semi-supervised learning algorithms Self-training, Label Propagation, and Semi-Supervised SVM to also use unlabeled data in the training process. Evaluation of this approach in an IoT-based industrial simulation environment shows that using MDA as the ML reinforcement layer increases the prediction accuracy by an average of 95%, and the robustness of the model against noisy data is also significantly improved. The results indicate that the combination of MDA and semi-supervised machine learning can improve the efficiency of IoT-based industries.

References:


1. Moin, A., et al., A model-driven approach to machine learning and software modeling for the IoT: Generating full source code for smart Internet of Things (IoT) services and cyber-physical systems (CPS). Software and Systems Modeling, 2022. 21(3): p. 987-1014.
2. Dral, P.O., et al., MLatom 3: a platform for machine learning-enhanced computational chemistry simulations and workflows. Journal of Chemical Theory and Computation, 2024. 20(3): p. 1193-1213.
3. M Casares, P.A., et al., GradDFT. A software library for machine learning enhanced density functional theory. The Journal of Chemical Physics, 2024. 160(6).
4. Bishop, C.M., Model-based machine learning. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2013. 371(1984): p. 20120222.
5. Eyke, N.S., B.A. Koscher, and K.F. Jensen, Toward machine learning-enhanced high-throughput experimentation. Trends in Chemistry, 2021. 3(2): p. 120-132.
6. Snoeck, M. and Y. Wautelet, Agile MERODE: a model-driven software engineering method for user-centric and value-based development. Software and Systems Modeling, 2022. 21(4): p. 1469-1494.
7. Akhbarifar, S., et al., RETRACTED ARTICLE: A secure remote health monitoring model for early disease diagnosis in cloud-based IoT environment. Personal and Ubiquitous Computing, 2023. 27(3): p. 697-713.
8. Boumahdi, F., et al., MDA4SOA+ d: A new model driven architecture to supporting decision making in SOA. Journal of King Saud University-Computer and Information Sciences, 2023. 35(5): p. 101544.
9. Wang, Y., et al., Machine learning-enhanced flexible mechanical sensing. Nano-Micro Letters, 2023. 15(1): p. 55.
10. Liu, J., et al., Internet of things challenges and future scope for enhanced living environments, in Advances in Computers. 2024, Elsevier. p. 201-246.
11. Park, Y.-S., B.U. Park, and H.-J. Jeon, Advances in machine learning-enhanced nanozymes. Frontiers in Chemistry, 2024. 12: p. 1483986.
12. Kwon, H., Z.A. Ali, and B.M. Wong, Harnessing semi-supervised machine learning to automatically predict bioactivities of per-and polyfluoroalkyl substances (PFASs). Environmental Science & Technology Letters, 2022. 10(11): p. 1017-1022.
13. Albayati, M.G., et al., Semi-supervised machine learning for fault detection and diagnosis of a rooftop unit. Big Data Mining and Analytics, 2023. 6(2): p. 170-184.
14. Ahmad, E., Model-based system engineering of the Internet of Things: A bibliometric literature analysis. IEEE Access, 2023. 11: p. 50642-50658.
15. Alulema, D., et al., SI4IoT: A methodology based on models and services for the integration of IoT systems. Future Generation Computer Systems, 2023. 143: p. 132-151.
16. Marques, G., Internet of Things: Architectures for Enhanced Living Environments. 2024: Elsevier.
17. Javed, A.R., et al., Future smart cities: Requirements, emerging technologies, applications, challenges, and future aspects. Cities, 2022. 129: p. 103794.
18. Mardani Korani, Z., et al., Model-driven engineering techniques and tools for machine learning-enabled IoT applications: A scoping review. Sensors, 2023. 23(3): p. 1458.
19. Moin, A. Data analytics and machine learning methods, techniques and tool for model-driven engineering of smart IoT services. in 2021 IEEE/ACM 43rd International Conference on Software Engineering: Companion Proceedings (ICSE-Companion). 2021. IEEE.
20. Moin, A., et al., Enabling automated machine learning for model-driven AI engineering. arXiv preprint arXiv:2203.02927, 2022.
21. Berrouyne, I., et al., A model-driven methodology to accelerate software engineering in the internet of things. IEEE Internet of Things Journal, 2022. 9(20): p. 19757-19772.
22. Wang, H. and D.-Y. Yeung, A survey on Bayesian deep learning. ACM computing surveys (csur), 2020. 53(5): p. 1-37.
23. Moin, A., et al. Supporting AI engineering on the IoT edge through model-driven TinyML. in 2022 IEEE 46th Annual Computers, Software, and Applications Conference (COMPSAC). 2022. IEEE.
24. Morin, B., N. Harrand, and F. Fleurey, Model-based software engineering to tame the IoT jungle. IEEE software, 2017. 34(1): p. 30-36.
25. Paszke, A., et al., Automatic differentiation in pytorch. 2017.
26. Atzori, L., A. Iera, and G. Morabito, The internet of things: A survey. Computer networks, 2010. 54(15): p. 2787-2805.
27. Bennett, K.P. and E. Parrado-Hernández, The interplay of optimization and machine learning research. The Journal of Machine Learning Research, 2006. 7: p. 1265-1281.
28. Nižetić, S., et al., Internet of Things (IoT): Opportunities, issues and challenges towards a smart and sustainable future. Journal of cleaner production, 2020. 274: p. 122877.
29. Aouedi, O., et al., A survey on intelligent Internet of Things: Applications, security, privacy, and future directions. IEEE Communications Surveys & Tutorials, 2024.
30. Moghaddasi, K. and M. Masdari, Blockchain-driven optimization of IoT in mobile edge computing environment with deep reinforcement learning and multi-criteria decision-making techniques. Cluster Computing, 2024. 27(4): p. 4385-4413.
31. Moin, A., S. Rössler, and S. Günnemann, ThingML+ augmenting model-driven software engineering for the Internet of Things with machine learning. arXiv preprint arXiv:2009.10633, 2020.
32. Rademacher, F., et al., Model-Driven Engineering of Microservice Architectures—The LEMMA Approach, in Ernst Denert Award for Software Engineering 2022: Practice Meets Foundations. 2024, Springer Nature Switzerland Cham. p. 105-147.
33. Verbruggen, C. and M. Snoeck, Practitioners’ experiences with model-driven engineering: a meta-review. Software and Systems Modeling, 2023. 22(1): p. 111-129.
34. GitHub - komeilmoghaddasi/IoT_Task_Offloading_Datase: This synthetic dataset represents a scenario of 10,000 interactions between different types of IoT devices and edge servers. if you want to use this dataset please cite article: https://doi.org/10.1007/s10586-023-04195-4