The growing adoption of Federated Learning (FL) is reshaping the way machine learning models are trained across distributed, privacy-sensitive datasets. However, the scalable and efficient orchestration of data engineering pipelines in decentralized cloud environments remains a significant challenge. This paper presents a comprehensive architectural framework for scalable data engineering tailored for FL in heterogeneous and resource-constrained environments. By integrating modern distributed computing paradigms, such as Kubernetes-based orchestration, edge-aware data preprocessing, and secure federated communication, we propose a modular architecture that addresses data heterogeneity, scalability, and compliance. A case study in a healthcare IoT scenario validates the performance and flexibility of the proposed system. Our work serves as a blueprint for deploying robust FL systems in real-world decentralized cloud ecosystems.
Laura Conti, Andrew Collins· International Journal of Dat...· 0 citations
Industry 4.0 has evolved traditional manufacturing into a highly connected, data-driven, intelligent production environment. Digital twin (DT) and predictive control have been considered as two of the enabling technologies in industrial automation that can greatly enhance the manufacturing efficiency, operational cost reduction, and product quality improvement. Digital Twin — In the context of manufacturing, a digital twin is a virtual model of a manufacturing system that continuously receives live operational data using IoT devices, cloud computing and artificial intelligence (AI). By integrating predictive control strategies, Digital Twins predict system behaviour (particularly Model Predictive Control (MPC)), and provide timely optimised actions before any faults/ inefficiencies happen. This is a survey on the state of art works related to Digital Twin-based Predictive Control for intelligent manufacturing systems, including its architecture, enabling technologies and practical applications. This discusses the role of Digital Twins in real-time monitoring, predictive maintenance, process optimization, Quality assurance and energy-efficient manufacturing. Moreover, the novel integration of machine learning algorithms enables Digital Twins to leverage big data in manufacturing (various plant operational informatics) as well as dynamically update control strategies based on changing operational environments. While researchers have made strides towards realizing Digital Twins, challenges still remain regarding abundance of computational requirements, inherent cybersecurity security risks, interoperability across applications and systems, pricey to implement solutions and absence of standardized Digital Twin frameworks. These research gaps are highlighted in this paper, and future directions towards autonomous self-optimizing manufacturing systems are discussed. The conjoining between Digital Twin technology and predictive control proposed in this article offers a paradigm of smart manufacturing that facilitates increases in production flexibility, reductions in downtime, increases in resource utilization and sustainable development. These results show that Digital Twin-based predictive control can be an important technological basis for next-generation intelligent manufacturing environments.
Marco Bianchi, Laura Conti· International Journal of Int...· 0 citations