Loading...
Paper Type
Complete
Abstract
Industry 4.0 (I4.0) is a pivotal change to business models and processes. Artificial Intelligence (AI), especially Deep Learning (DL) on the edge, has made great progress towards building the Internet of Things (IoT) devices that provide real-time inferences using limited computational resources. Many industries are eager to adopt DL for IoT to leverage their competitive advantage. However, companies also have apprehensions regarding cost, reliability, security, networking, and trust around related technologies. These concerns call for an approach that standardizes DL to IoT. This paper introduces the concept of Federated Deep Learning (FDL) to enable I4.0 companies to adopt DL for IoT end devices. The conceptual model aims to provide a secure strategy to federate deep learning models on the edge and end nodes. To elucidate the application of the model, a framework is presented to explain how FDL may be applied to an I4.0 automobile manufacturing facility. The framework presents capabilities to transition current manufacturing facilities to future “smart factories”.
Recommended Citation
Elnagar, Samaa and Thomas, Manoj A., "Federated Deep Learning: A Conceptual Model and Applied Framework for Industry 4.0" (2020). AMCIS 2020 Proceedings. 24.
https://aisel.aisnet.org/amcis2020/ai_semantic_for_intelligent_info_systems/ai_semantic_for_intelligent_info_systems/24
Federated Deep Learning: A Conceptual Model and Applied Framework for Industry 4.0
Industry 4.0 (I4.0) is a pivotal change to business models and processes. Artificial Intelligence (AI), especially Deep Learning (DL) on the edge, has made great progress towards building the Internet of Things (IoT) devices that provide real-time inferences using limited computational resources. Many industries are eager to adopt DL for IoT to leverage their competitive advantage. However, companies also have apprehensions regarding cost, reliability, security, networking, and trust around related technologies. These concerns call for an approach that standardizes DL to IoT. This paper introduces the concept of Federated Deep Learning (FDL) to enable I4.0 companies to adopt DL for IoT end devices. The conceptual model aims to provide a secure strategy to federate deep learning models on the edge and end nodes. To elucidate the application of the model, a framework is presented to explain how FDL may be applied to an I4.0 automobile manufacturing facility. The framework presents capabilities to transition current manufacturing facilities to future “smart factories”.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.