The latest release of Cognite Data Fusion® brings the concept of Core Data Models across the full spectrum of industrial data and streamlines how users can interact with their data, providing a future-proof data platform for deploying AI solutions at scale.
With this release, Cognite continues to deliver on its North Star vision of helping energy and industrial companies contextualize their complex industrial data to scale advanced use cases such as turnaround planning, digital operator rounds, and production optimization —ultimately letting them deploy AI Agents for autonomous operations.
"With the launch of Cognite Data Models, a new, out-of-the-box data modeling functionality, it has never been easier for users to reflect their industrial reality at scale. Our customers in energy and manufacturing can now easily onboard and get real-time access to large amounts of multi-modal data for their decision-making," said Knut Vidvei, Head of Product Management at Cognite.
The power of data models to reflect your industrial reality
Cognite offers users a Core Data Model that provides a solid framework to let industrial data speak human. Users can richly express their data and industrial reality by extending it with Industry Data Models. Users are also encouraged to extend these models to create their own custom Data Models, providing incredible flexibility and structure at the same time.
These models serve as building blocks so users can solve their challenges without data duplication, offering guaranteed functionality without the need for additional configurations. Cognite’s Core Data Model crucially reduces deployment time for new customers by 50% to help them quickly achieve value and scale AI solutions across their enterprise.
One place for all your industrial data
Users can access all their industrial data via Cognite Data Fusion® in real time and at scale. All data is seamlessly contextualized, and the Data Model is automatically populated with the right data. Users can easily interact with their data, whether that is through the Industrial Search experience to find critical data, using no-code analytics in Cognite Charts for root cause analysis, or exploring and analyzing data for turnaround planning in Industrial Canvas.
The Core Data Model turbocharges access to industrial data by enabling users to onboard data directly from source systems into Cognite Data Fusion using extractors such as Cognite File, OPC UA, PI, and SAP OData extractors. Cognite Data Fusion now also offers hosted extractors for MQTT and Kafka, REST, and Azure Event Hub to help users significantly improve the flow of data from source systems to Cognite Data Fusion.
Automate your data workflows for next-level operations
Cognite is introducing the ability to natively define schedule-based triggers within data workflows to help users automate what actions should be taken when certain triggers happen. This simplifies the automation of data workflows and reduces complexity, allowing users to spend more time on their most critical operations.
Combined with write-back to SAP source systems, we now let users manage maintenance operations with a “single source of truth.” Cognite Data Fusion® now supports write-back to SAP when using Cognite out-of-box applications. This is possible by implementing a bi-directional data flow that enables seamless synchronization of SAP notifications data between SAP S4HANA and Cognite Data Fusion®. This allows frontline maintenance operators to have full trust in their data.
Getting even more value from your industrial data using AI
We are using AI to enhance everything users are already doing in Cognite Data Fusion®, whether they are finding the right data, doing real-time ROI calculations for turnaround planning, or creating advanced cause map diagrams and reliability work.
Real-time cost impact analysis for turnaround scoping: Users can directly see the cost impact of their scoping decisions to improve turnaround ROI. The cost calculations are automatically updated and give a real-time view of the effect of different work orders and resource scenarios under different scenarios. We offer the same for your maintenance flows to enable teams to make informed decisions and assess the financial impact of various scoping options.
AI-improved retrieval for multiple documents (beta): It's time-consuming for industrial users to identify relevant documents and search for specific data within each document. To make this easier, we are introducing the ability to search using natural language, which will search up to 100 documents simultaneously. This will allow users to use insights across multiple documents and get relevant information quickly and effectively.
Auto-aligned cause map diagrams (beta): This feature lets users easily create cause maps directly in Industrial Canvas, where all necessary data is readily available as evidence to support or reject a hypothesis. It includes auto-alignment when creating cause maps and built-in status indicators to streamline the process even further.
Rule-based coloring for visual condition monitoring (beta): Empower reliability engineers to better monitor the health of their plants and critical systems. No more checking each piece of equipment individually. Industrial Canvas is introducing support for color-coding assets and pipes on diagrams. This feature will categorize equipment based on several vital rules, such as corrosion rate, operating temperature, and more, providing a clear and immediate visual representation of conditions.
An incredible ecosystem and next-level simulators
This release also expands the openness of Cognite Data Fusion®, ensuring customers can work with any number of partners to quickly connect off-the-shelf applications or develop new services to meet real-world demands. Customers can now create simulation workflows using Cognite APIs, enabling users to scale simulations to thousands of models.
Combine simulators with plant data to generate new, valuable insights. We are introducing a vendor-agnostic service that supports integration with simulators from multiple vendors. This includes out-of-the-box connectors to industry-standard simulator tools and a Simulator API that allows for programmatic interaction and orchestration of simulator resources. This integration enables real-time asset performance optimization, predictive maintenance, and what-if scenario analysis, driving significant operational improvements.
A future-proof data foundation for AI agents
The enhancements to Cognite Data Fusion's core functionality support Cognite Atlas AI, an industrial agent workbench that extends Cognite’s Industrial DataOps platform to help industrial organizations use generative AI to carry out more complex operations with greater accuracy.
We also include support for Claude and Gemini on projects deployed on Amazon Web Services and Google Cloud Platform, respectively, ensuring no data is moved out of the customer tenant to use Generative AI features. This underpins Cognite’s commitment to helping customers establish a trustworthy data foundation to safely deploy AI agents at scale.
For more information or questions, please get in touch here or dive deeper into the September 2024 Product Release with the below materials that include selected deep dives: