Check the
documentation
Ask the
Community
Take a look
at
Academy
Cognite
Status
Page
Contact
Cognite Support
Posting on behalf of Koch Ag and Energy Solutions
Summary
We are requesting functionality to enable support for defining custom GraphQL schemas and resolvers on top of CDF. This capability would allow our team to build tailored, efficient APIs for operational applications that surface actionable insights from contextualized manufacturing data.
Context
We are using CDF to centralize and contextualize key plant data — including assets, sensor readings, events, shift logs, inspection documents, and relationships to business systems.
Our application team is developing internal tools for specific use case enablement.
These tools require fast and intuitive access to data, but CDF’s default GraphQL schema is too low-level for these user-facing applications.
Current Challenges
- Our frontend developers must manually chain multiple queries to get asset status, performance trends, and associated work orders.
- Common business logic (like calculating asset availability, filtering alarms, or joining time series to asset metadata) must be repeated in each app.
- We cannot easily present business-specific views like “ProductionLineHealth” or “ShiftSummary” without building and maintaining our own middleware.
Requested Capabilities
We are asking for the ability to:
- Define custom GraphQL object types that aggregate and shape CDF data into operational concepts (e.g.,
ProductionLine,MachineStatus,DowntimeSummary) - Implement custom query resolvers for calculated fields (e.g., OEE, MTBF, energy efficiency)
- Integrate external system data (e.g., SAP PM, shift logs, scheduling systems) into a unified query layer
- Apply business rules, KPIs, and filtering logic server-side for reuse across applications
- Secure the schema and field-level access based on user roles (operator vs planner, plant vs corporate)
Example Use Case
Custom Type: ProductionLineOverview
graphql
type ProductionLineOverview {
lineId: ID!
name: String
currentThroughput: Float
availability: Float
topFiveAlarms: [Alarm]
openWorkOrders: [WorkOrder]energy
UseToday: Float
} This schema would simplify dozens of low-level CDF queries into a single, reusable API call for our production dashboard.
Expected Value
| Benefit | Impact |
|---|---|
| Streamlined application development | APIs are aligned with plant workflows, not just data structure |
| Lower integration effort | Data from multiple sources can be presented in one logical schema |
| Faster time to insight | Field users access actionable metrics in fewer steps |
| Consistent KPI logic | Calculations like OEE or uptime are standardized |
| Broader adoption of CDF | More teams (maintenance, operations, digital) can build on it directly |
Potential Implementation Paths
- Allow deployment of custom GraphQL services inside CDF projects (e.g., based on Apollo Server)
- Provide schema extension hooks that pull and reshape data from CDF's GraphQL or REST APIs
- Offer first-party support for calculated fields or composite objects in GraphQL
- Enable hybrid access to external data systems within the same query context
Closing
This capability would significantly enhance how our teams leverage CDF in the field, drive app development efficiency, and improve visibility across operations.
We would be happy to participate in design validation or pilot this capability if/when it becomes available.
