Abstract

Federated Learning (FL) combined with the Industrial Internet of Things (IIoT) enhances decision-making in industrial settings by leveraging decentralized machine learning (ML) to ensure data privacy, optimize edge computing, and facilitate adaptive model training. However, implementing FL and IIoT presents challenges due to distributed architectures, including communication, data transfer, and file management across wide area networks. This paper addresses these challenges by introducing FL and Clustered FL (CFL) models using Pegasus workflows. Evaluated with real data from airport baggage conveyor systems, it offers practical insights into FL's application in IIoT environments, contributing to advancements in intelligent industrial decision-making.

Metadata

publication

2024 20th International Conference on Distributed Computing in Smart Systems …, 2024

year

2024

publication date

2024/4/29

authors

Hamza Safri, George Papadimitriou, Frédéric Desprez, Ewa Deelman

link

https://ieeexplore.ieee.org/abstract/document/10621546/

resource_link

https://georgepapadimitriou.com/publication/safri-dcoss-iot-2024/safri-dcoss-iot-2024.pdf

conference

2024 20th International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT)

pages

259-263

publisher

IEEE