OEMs are winning specification battles by implementing unified namespace structures and data contextualization to help manufacturers standardize their unstructured data, enabling AI-driven digital transformation. While 59% of manufacturers use smart technology, only 43% effectively use their data, making OEMs' role in creating standardized data models and ontologies critical for scaling operations.
- 59% of manufacturers actively run smart manufacturing technologies, but only 43% of collected data gets used effectively
- Unified namespace structure (UNS) solves the "data spaghetti" problem by creating standardized asset data and defined equipment relationships across plants
- OEMs designing for UNS from the start are winning specifications competitors don't even know they're competing for
- Real-world results: Bühler's Grinding Optimizer reduced power consumption by 38% in pilot tests with targets of 8-10% energy savings plant-wide
- Data contextualization requires connecting measurements with machines, processes, products, and operational events—not just collecting raw sensor data
Legacy and greenfield plants are creating mountains of data. A recent Rockwell Automation survey shows 59% of manufacturers are actively running smart manufacturing technologies, while only 43% of the data collected gets used. Interoperability among equipment, controllers, and middleware, and sharing data throughout the plant, present many challenges for end users.
Original Equipment Manufacturers (OEMs) have implemented tools that have elevated food and beverage plants to different levels of digital transformation over the last 10 years. Now, agentic AI is driving a more comprehensive form of digital transformation through standardized asset data and defined equipment relationships.
The next chapter: Unified namespace structure
The industry is struggling to define and roll out standardized data initiatives and is looking for trusted sources. OEMs, system integrators, and external data specialists are emerging as trusted sources and showing how to scale digital transformations.
The first wave of Industry 4.0 included advanced sensors, adoption of open industrial protocols, and sharing data to middleware. Now, the data needs meaning.
“The unified namespace structure (UNS) conversation the industry's been having is really the middle chapter of a three-part story,” says David Ariens, Co-founder, IT/OT Insider Academy. IT/OT Insider Academy offers educational content on contextual data architecture for food and packaging operations. “Before UNS, plants were living with data spaghetti, think mixed protocols, point-to-point links between PLC, SCADA, and ERP, and no shared naming convention,” adds Ariens.
OEMs understand the moment and the serious challenges for food processors contending with unstructured data. “A vibration or temperature value alone does not explain the potential impact on equipment condition, production, quality, or energy consumption,” says Thiago Prada, Project Mgr., Automation and Digital Sales, Bühler Group. “Creating value requires connecting measurements with the machine, process conditions, product, and operational events.”
“Every company needs a common data model that spans assets, processes, systems, and departments throughout the organizational structure,” says Patrick Fetterman, Research Analyst, LNS Research.
Acquisitions and mergers have led to multiple vendors in plants with legacy packaging, processing, and controls, leading to disparate data. “Different asset names, data structures, and KPI definitions make it difficult to create a consistent view of performance across machines, lines, and facilities,” adds Prada.
“A lot of companies still lean on system integrators for every point-to-point connection, and this leads to no internal standard for the next project, and every integration is a bespoke job instead of a repeatable one,” according to Ariens.
So OEMs are stepping up, as they did in the first phase of Industry 4.0, and pushing processors toward action. Recently, Bühler Group combined time and vibration measurements, historical operator adjustments, process expertise, and AI to develop the Grinding Optimizer Mill System.
A recent use case with the mill system showed reduced power consumption in a single roller pair by as much as 38%, while maintaining quality within the target range. “This was a specific result for one adjustment, not a plant-wide expectation,” Prada adds. “More broadly, the company targets an 8% to 10% reduction in roller-mill energy consumption through AI-supported grinding optimization, with potential improvements in yield and product quality.”
Blentech, a supplier of industrial blenders, cookers, and kettles, has worked with customers for over a decade to move machine data across the enterprise. Blentech’s machinery uses API and MQTT communication systems and runs on Inductive Automation’s Ignition platform.
“I'm seeing awareness when we're getting customers to engage with data context,” says Daniel Voit, CEO and CRO, Blentech Corp. “There is a tendency for people to want to jump to AI as the solution and skip the creation of a robust data layer.”
According to Ariens, the next step is a standardized ontology layer on top of process data. “The UNS as most people know it proposes a solution with a centralized broker and an ISA-95-style namespace,” adds Ariens. “Progress, but still a flat tree since it tells you what data exists, not how it relates.”
Many reports point to the food industry’s struggles with standardized data projects within companies. The struggle is real, and OEMs reflect it.
“If a customer comes to us and says this is our tag structure, then we have a phase that's built in,” says Voit. “Now, the vast majority of customers are not actually asking us, but we are doing it and telling them they need it.”
Large food manufacturers have stopped large, cumbersome data standardization projects because they didn't connect to specific business cases. However, food and beverage producers are turning to external resources to address these challenges. Ruiz Foods turned to Infor and its extensive integrator ecosystem for its R&D data project. Ruiz Foods’ Michael Warter, SVP and CIO, shared its successful lab testing data standardization initiative at the 2024 IFT FIRST conference.
The business case was to standardize unstructured R&D and lab data in cloud-based Enterprise Resource Planning (ERP) and Product Lifecycle Management (PLM) systems. "Data was in disparate systems, and integration was vital in getting the systems to work together and move away from spreadsheets," said Warter during the panel discussion. "The first step provided regulatory, compliance, and track-and-trace benefits."
OEMs and data contextualization
OEMs’ interoperability efforts with machine designs through open networking protocols have been successful, and UNS appears to be the next hill to climb. “OEMs are increasingly being asked by end users to design machines that fit into a UNS-based data infrastructure, where information flows into a single, organized, real-time layer rather than siloed systems talking past each other,” said Nikki Gonzales, Dir. of Business Development, Weintek USA, in a 2026 OEM article.
Successful OEMs see the writing on the wall when it comes to winning customers. “OEMs who understand it and design for it from the start are winning specifications that others are not even aware they are competing for,” says Voit.
In addition to OEMs, data contextualization companies can provide answers. HiveMQ, HighByte, Litmus, and Haystack Data Solutions are partnering with automation and control companies like Rockwell Automation, Siemens, and more to help solve data modeling and contextualization.
“In the future, food producers will present a specification and say this is the tag structure we need,” adds Voit. “That's the reason we've built API and MQTT communication systems so that we can include the contextualization from our system in the data.”
Contextualizing data is essential for manufacturers and processors to scale and move to agentic AI. “Where the field is heading now is ontologies and graphs on top of that: explicit relationships between equipment, process, and product, plus proper historized storage instead of just current state,” adds Ariens of IT/OT Insider Academy.
With so much unstructured data in plants, this presents a sturdy challenge for food producers. “It's also forcing companies to realize ‘industrial data’ isn't just PLC tags; it's the CMMS entries, wiring diagrams, and RCAs that never made it into the historian,” says Ariens.
“The trend is clear: processing equipment is becoming not only a production asset, but also a reliable source of contextualized information,” adds Prada.
Digital transformation and success look very different across companies. As many have seen, digitizing processes was the first step; now OEMs and other vendors are helping companies standardize data models and, more importantly, add ontologies to scale and move toward a “complete digital transformation.”
