A New Data Foundation for Industrial AI in Manufacturing
Industrial AI in manufacturing refers to the use of machine learning, analytics and intelligent agents on factory data from machines, systems and business applications to improve quality, productivity and maintenance decisions at scale. Siemens’ partnership with HighByte is framed around this definition: transforming industrial data into an asset that can reliably feed AI models and applications across operations. Through the deal, HighByte Intelligence Hub becomes part of Siemens’ Industrial Edge ecosystem, alongside Intelligence Center X. This combination targets one persistent barrier to industrial AI manufacturing projects: fragmented, inconsistent data scattered across operational technology and information technology systems. By placing HighByte’s industrial DataOps engine directly on the edge, Siemens wants plants to build a shared, standards-based manufacturing data pipeline instead of bespoke integrations for each AI use case.
Closing the OT/IT Data Gap with Edge Computing Solutions
At the heart of the collaboration is OT IT data integration delivered through edge computing solutions. HighByte Intelligence Hub now runs natively on Siemens Industrial Edge, which manages application deployment and configuration close to production assets. Through the Industrial Edge Connectivity Suite, plants can pull data from PLCs, SCADA systems and industrial protocols, while HighByte extends that reach to IT systems such as MES platforms and enterprise applications. This creates a shared data layer that can be reused across multiple industrial AI manufacturing projects instead of rebuilt each time. According to Siemens Digital Industries, the combined approach “bridges the gap between shop floor operations and IT systems” by making data from diverse sources accessible, understandable and actionable across the enterprise.
From Raw Signals to a Reusable Manufacturing Data Pipeline
HighByte’s main contribution is industrial DataOps: modeling, orchestration and governance to build a reliable manufacturing data pipeline. Running on Industrial Edge, the Intelligence Hub can apply transformation rules to raw machine signals and IT data, aligning tags, units, equipment models and business context before the data reaches AI engines. This contextualization turns unstructured operational data into structured, reusable data sets for Intelligence Center X and other IT services. The same tools support pipelining, routing prepared data to AI models, analytics platforms or data lakes, while also handling bidirectional flows so commands from IT systems can move back through Industrial Edge to PLCs to adjust machine setpoints. The result is a standardized data layer that reduces one-off integration work and supports consistent industrial AI manufacturing deployments across lines and sites.
Scaling Industrial AI Manufacturing Without Deep OT Expertise
For many manufacturers, a shortage of OT experts has slowed industrial AI projects. Siemens and HighByte position their combined platform as a way to scale AI without requiring deep OT skill sets for every initiative. With contextualized, standardized data delivered by Industrial Edge and HighByte, developers can focus on model design and application logic instead of low-level protocols or plant-specific quirks. HighByte CEO Tony Paine stated that the integration with Industrial Edge provides “a direct path to contextualized and standardized data” needed to build data products and AI at scale with Intelligence Center X. The approach fits a broader industry shift toward reusable data products and industrial agents that can be deployed across assets, supported by a consistent OT IT data integration layer.
Early Results: Predictive Maintenance in Glass Production
An early deployment at flat glass producer Vivix Vidros Planos shows how the Siemens–HighByte stack can be used in practice. The company runs Siemens Industrial Edge, HighByte Intelligence Hub and Intelligence Center X across about 30 industrial applications in its production environment. One highlighted use case is predictive maintenance on a glass furnace, where contextualized industrial data flows from edge devices into AI models that monitor and predict furnace degradation to support maintenance planning and equipment utilization. Aristóteles Terceiro Neto from Vivix Vidros Planos said this combination lets developers build industrial AI models, agents and applications “at speed and scale, with very little support from [their] limited pool of OT experts.” For Siemens, these results help validate its strategy to deliver end-to-end edge computing solutions—from data capture to AI-powered decision making.
