Abstract
We are still in the midst of Industry 4.0 (I4.0), with more manufacturing lines being labeled as smart thanks to the integration of advanced ICT in Cyber–Physical Systems (CPS). While I4.0 aims to provision cognitive CPS systems, the nascent Industry 5.0 (I5.0) era goes a step beyond, aiming to build cross-border, sustainable, and circular value chains benefiting society as a whole. An enabler of this vision is the integration of data and AI in the industrial decision-making process, which does not exhibit yet a coordination between the Operation and Information Technology domains (OT/IT). This work proposes an architectural approach and an accompanying software prototype addressing the OT/IT convergence problem. The approach is based on a two-layered middleware solution, where each layer aims to better serve the specific differentiated requirements of the OT and IT layers. The proposal is validated in a real testbed, employing actual machine data, showing the capacity of the components to gracefully scale and serve increasing data volumes.
Highlights
The emergent fifth industry revolution, known as Industry 5.0 (I5.0), aims to establish value chains spanning heterogeneous industrial domains, enhancing re-use, increasing production flexibility, and exhibiting resiliency in times of disruption [1]
The contribution of this work is three-fold: (i) we present a practical solution for the Operation and Information Technology domains (OT/IT) convergence problem (ii), we present the implementation details of a two-layered middleware approach best fitting the needs of the OT/IT layers, and (iii) we validate the proposal in a real testbed
Edge computing plays a crucial role in enabling the design and implementation of novel distributed control functions with parts that are hosted on the edge nodes located in the production plant premises and close to the controlled sensors/actuators, primarily to increase reliability and decrease latency [13]. This edge-enhanced cloud architecture provides several benefits compared to a pure data center-based approach: application latency is reduced because of vicinity to end-points; inter-domain traffic is diminished because, for example, Multi-access Edge Computing (MEC) nodes stay in the telco operator network; sensitive information/processing can be maintained at industrial edge gateways in the premises of end-points, while global status visibility can be employed, e.g., when needed for global machine learning optimization, by interacting with pure data center-based cloud resources [14]
Summary
The emergent fifth industry revolution, known as Industry 5.0 (I5.0), aims to establish value chains spanning heterogeneous industrial domains, enhancing re-use, increasing production flexibility, and exhibiting resiliency in times of disruption [1]. The current manufacturing landscape comprises heterogeneous machines and production facilities capable of autonomous message exchange, generating data at an ever-increasing speed, and all data could provide useful information and could be used proactively for optimized control and business-related purposes [3] This capability could bring fundamental improvements to the industrial processes in manufacturing, engineering, supply chain, and life cycle management [4]. A Gateway component, deployed on the edge computing fabric, defined to handle the coordination of layers at the OT/IT boundary, is capable of conveying OT data towards the IT layer with reliability and security To this end, we rely on Apache Kafka, a Message-oriented Middleware (MoM) equipped with a rich ecosystem of plugins, capable of providing differentiated QoS to OT flows.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
