While rapidly advancing AI technologies have enabled corporations to build useful AI applications from common enterprise IT data, the uniquely difficult nature of production data has left companies struggling to achieve rapid, repeatable and scalable results from their manufacturing operational data.
Today we’re excited to announce the integration of Sight Machine, Microsoft Fabric’s Real-Time Intelligence (RTI) and NVIDIA Omniverse libraries into a full AI-powered solution for connecting, structuring, analyzing and visualizing production data, and integrating the resulting insights with data from across the enterprise.
This combination leapfrogs the piecemeal assemblies of individual services that require years of integration and data engineering, with an elegant and powerful solution that manufacturers can have up and running in weeks.
Manufacturers can now consolidate all their industrial data into the Fabric environment, unlocking AI-driven value at enterprise scale, while leveraging Sight Machine’s industrial AI capabilities and an embedded 3D digital twin built with NVIDIA Omniverse libraries.
This powerful three-way integration delivers a unified ecosystem where:
Together, these technologies enable organizations to seamlessly connect, stream, store, contextualize, model, consume and agentify all kinds of production data. Data flows from the factory floor through Azure IoT Operations into a Microsoft Fabric Eventstream and Eventhouse and Sight Machine Factory Transform, where it undergoes progressive refinement through the streaming medallion architecture (bronze / silver / gold) – from raw operational data to business-ready manufacturing insights.
Manufacturers can now optimize production operations together with enterprise functions like sales, logistics and finance, and manage production data under the same governance standards applied to other enterprise data in Fabric. They can answer questions like, “Which plant has the capacity to fill this large new customer order next week – and how much will this rush order add to our production costs?”
Every layer of the technology stack in this tightly integrated ecosystem scales effortlessly – unlocking exponential speed, performance, and impact across the data-to-value journey.
At its plant in Mexico, Ronal manufactures light alloy wheels for passenger cars and commercial vehicles, a process that is energy-intensive and precise. The plant runs eight melting furnaces and 24 casting machines simultaneously. The production challenges include minimizing energy consumption while optimizing process parameters for the casting and furnace machines to ensure maximum quality for its wheels.
At Ronal, Sight Machine's Furnace Agent suggests recommended values for casting and furnace process parameters in real time. These agentic recommendations are made available within a 3D digital twin built with NVIDIA Omniverse libraries.
As Ronal evolves its long-term IT and OT architecture, Sight Machine’s integration with Fabric Real-Time Intelligence gives Ronal confidence that the entire technology, data platform, and AI agent stack scale seamlessly.
With its ability to integrate with Fabric Real-Time Intelligence, Sight Machine provides flexibility to its clients who are adopting Microsoft Fabric. Fabric Real-Time Intelligence customers who want to deploy AI solutions at scale can now leverage Sight Machine's proven manufacturing data processing and analysis capabilities alongside Microsoft's Azure and Fabric tools.
The same suite of manufacturing optimization applications and AI agents available to Sight Machine clients for streaming data are now available to Sight Machine clients adopting Fabric Real-Time Intelligence.
Data engineers, data scientists and software developers across the enterprise who are building custom applications and workflows can leverage Sight Machine's manufacturing intelligence, powerful visual interface capabilities built with Omniverse libraries, and their existing Fabric data assets within a single development framework.
Our innovative architecture empowers customers to access manufacturing data directly within their native Microsoft Fabric environment. Previously, customers required provisioning of individual data copies on demand – a process that created latency and operational overhead.
Customers can now consume raw production data (bronze layer) in real time, while also gaining low-latency access to structured, AI-ready models (silver layer) and curated data including KPIs (gold layer) from Sight Machine through native Fabric connections. This allows teams to seamlessly integrate contextualized manufacturing insights into business applications.
NVIDIA Omniverse libraries enable the development of a flexible interoperable 3D visualization layer that can access data from Fabric Real-Time Intelligence as well as from Sight Machine. These immersive environments enable operations teams to access the same, unified 3D visualization of the production line and collaborate more effectively. Agentic AI insights from Sight Machine overlaid onto the 3D models allow users to track entire factory operations to determine upstream and downstream impacts on real time events, and make decisions using what-if analysis and simulation.