Interoperability and scalability are key requirements for the implementation of Industry 4.0 (I4.0) application scenarios targeting flexible and resilient production systems. In place of the rigid structures and heterogeneous communication solutions used in Operational Technology (OT) today, loosely coupled systems using standardized interfaces and information models are needed in order to decrease the high manual effort for (re-)configuration.
With this regard, the PublishSubscribe (PubSub) communication paradigm is well suited for the implementation of loosely coupled systems, that can e.g. be implemented using the MQTT application protocol.
However, MQTT is data agnostic and does not address interoperability on the information layer, i.e. the definition of topic trees and payloads are up to the user. Specifications like Sparkplug B (Sparkplug) and the standard OPC Unified Architecture (OPC UA) use MQTT as a transport protocol and add such missing features.
In this article, Sparkplug and OPC UA are compared according to the requirements of the I4.0 scenarios Plug & Work and Condition Monitoring.
