The role of maintenance in the industry has been shown to improve companies productivity and profitability. Industry 4.0 revolutionised this field by exploiting emergent cloud technologies and IoT to enable predictive maintenance. Significant benefits can be obtained by taking advantage of historical data and Industrial IoTstreams, combined with high and distributed computing power. Many approaches have been proposed for predictive maintenance solutions in the industry. Typically, the processing and storage of enormous amounts of data can be effectively performed cloud-side (e.g., training complex predictive models), minimising infrastructure costs and maintenance. On the other hand, raw data collected on the shop floor can be successfully processed locally at the edge, without necessarily being transferred to the cloud. In this way, peripheral computational resources are exploited, and network loads are reduced. This work aims to investigate these approaches and integrate the advantages of each solution into a novel flexible ecosystem. As a result, a new unified solution, named SERENA Cloud Platform. The result addresses many challenges of the current state-of-the-art architectures for predictive maintenance, from hybrid cloud-to-edge solutions to intermodal collaboration, heterogeneous data management, services orchestration, and security.

A hybrid cloud-to-edge predictive maintenance platform / Marguglio, Angelo; Veneziano, Giuseppe; Greco, Pietro; Jung, Sven; Siegburg, Robert; Schmitt, Robert H.; Monaco, Simone; Apiletti, Daniele; Nikolakis, Nikolaos; Cerquitelli, Tania; Macii, Enrico (INFORMATION FUSION AND DATA SCIENCE). - In: Predictive Maintenance in Smart Factories / Marguglio A., Veneziano G., Greco P., Jung S., Siegburg R., Schmitt R. H., Monaco S., Apiletti D., Cerquitelli T., Nikolakis N., Macii E.. - ELETTRONICO. - [s.l] : Springer, 2021. - ISBN 978-981-16-2939-6. - pp. 19-37 [10.1007/978-981-16-2940-2_2]

A hybrid cloud-to-edge predictive maintenance platform

Monaco, Simone;Apiletti, Daniele;Cerquitelli, Tania;Macii, Enrico
2021

Abstract

The role of maintenance in the industry has been shown to improve companies productivity and profitability. Industry 4.0 revolutionised this field by exploiting emergent cloud technologies and IoT to enable predictive maintenance. Significant benefits can be obtained by taking advantage of historical data and Industrial IoTstreams, combined with high and distributed computing power. Many approaches have been proposed for predictive maintenance solutions in the industry. Typically, the processing and storage of enormous amounts of data can be effectively performed cloud-side (e.g., training complex predictive models), minimising infrastructure costs and maintenance. On the other hand, raw data collected on the shop floor can be successfully processed locally at the edge, without necessarily being transferred to the cloud. In this way, peripheral computational resources are exploited, and network loads are reduced. This work aims to investigate these approaches and integrate the advantages of each solution into a novel flexible ecosystem. As a result, a new unified solution, named SERENA Cloud Platform. The result addresses many challenges of the current state-of-the-art architectures for predictive maintenance, from hybrid cloud-to-edge solutions to intermodal collaboration, heterogeneous data management, services orchestration, and security.
2021
978-981-16-2939-6
978-981-16-2940-2
Predictive Maintenance in Smart Factories
File in questo prodotto:
File Dimensione Formato  
Serena_Book_Chapter2.pdf

non disponibili

Descrizione: Capitolo numero 2
Tipologia: 1. Preprint / submitted version [pre- review]
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 727.87 kB
Formato Adobe PDF
727.87 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2916180