Why this matters now
We all know data is king. The entire purpose of the digital transformation initiatives of the last decade or two were laser focused on getting critical data out of the field and into the hands of personnel and systems that could make use of it.
However, in those years of collecting and storing data, operational technology (OT) teams learned a valuable lesson: data is extremely challenging to use without context. As processes become more complex and individual elements of the process work in tandem with others—especially as automation gets more complex with advanced process control and state-based control—analysts and software need to understand the operating conditions surrounding data, not just the value, for it to be truly meaningful.
Leveraging their combined decades of experience in the automation industry, Emerson’s Keith McNab and Daniel Smith explore this challenge in a recent article in Industrial Ethernet magazine. The key to maintaining context, they explain, is powerful open protocols like OPC UA and MQTT.
Takeaway: Data only becomes valuable for analytics and AI when it is delivered with full operational context.
TL;DR
- Data without context limits analytics and AI effectiveness.
- Traditional proprietary protocols deliver only basic values.
- OPC UA securely delivers contextualized industrial data.
- MQTT enables scalable, lightweight movement of that data.
- Open protocols are foundational for AI-ready OT architectures.
Why use open protocols?
Many traditional, proprietary protocols only provide the most basic information—typically a basic value and a timestamp. But as organizations pursue higher-level analytics to understand how systems interact, those basic values limit their ability to understand system behavior.
This challenge has been exacerbated by the rise of AI. AI software tools consume massive amounts of multivariate data in a very short time to identify patterns and connections that are often invisible to human analysts. To do so, that multivariate data must be available.
“OPC UA, MQTT, and AI are all technologies working together on industrial networks to bring more value to the data created during operations, greatly benefiting users. OPC UA elevates and secures data; instead of simply reporting a basic value with a timestamp and engineering units, OPC UA securely delivers contextualized information to provide more value and greater understanding.”
MQTT then moves that contextualized information at scale via a lightweight framework, helping teams build an enterprise operations platform.
Takeaway: Open protocols unlock the contextual, multivariate data AI requires to perform effectively.
Speaking the same language
Industrial environments rarely operate a single technology stack. Equipment from many vendors must intercommunicate for increasingly autonomous operations to succeed. OPC UA provides a shared language that enables seamless interoperability across vendors and devices.
“This increased flexibility gives users a better view into the process as a whole because they can collect data from many sources and bring it together to make better decisions via a wide array of tools, including AI models. In addition, the consistency offered by a shared language makes tools like AI agents far easier to create.”
“We’ve had predictive analytics and maintenance applications running in the cloud for some time now, but to get them configured, teams would have to tap an application engineer who would spend hours trying to map the data from the field to the data that resides or is consumed in the cloud. All that mapping was manual and took a lot of effort. Now, since we’ve standardized and normalized that information at the edge, nearly all of that manual effort is eliminated. It’s almost entirely automatic.”
As organizations increasingly turn to industrial AI, open protocols like OPC UA and MQTT reduce integration friction and accelerate ROI.
Takeaway: Standardized, open protocols dramatically reduce complexity and accelerate AI deployment.
