A critical architecture decision is where machine logic executes: headless (cloud/edge) or app-based (terminal).
Trade-off matrix
The table below explains different logic execution methods.
| Method | Headless | Cost | Ideal use cases | Limitations |
|---|---|---|---|---|
| Machine triggers | Yes | Included in Tulip machine license | OEE tracking, state management, simple part counting | Limited actions (state changes, activity fields only), no expression editor, cannot call connector functions |
| Automations | Yes | Per-task pricing | Complex workflows, table writes, connector functions, external system integration | Cost scales with usage, no completion records |
| App triggers | No | Included | Human-machine workflows, rich UI interactions, complex logic with expressions | Requires app running on terminal, not suitable for 24/7 monitoring |
| DataOps platform (e.g. Litmus, High Byte) | Yes | Contact DataOps vendor | Advanced data normalization, time-series database, multi-protocol aggregation | Additional platform to manage, higher total cost, integration complexity |
| Edge device + Node RED | Yes | One-time hardware cost | Highly customized logic, customer-owned infrastructure | Less reliable (no failover/multi-threading), harder to scale, maintenance burden |
Decision criteria
Machine triggers
Use machine triggers when:
- You need basic OEE tracking (state, part count, defects)
- Logic is simple conditional checks (e.g., "if attribute > 10, set state to Running")
- Cost optimization is critical (included with Tulip)
Automations
Use Automations when:
- You need to write machine data to Tulip tables
- Integration with external systems via connector functions is required
- Headless operation is mandatory and budget supports per-task pricing
Example: "When Machine 5 outputs defect code, create record in Quality Issue table AND send Slack notification"
App triggers
Use app triggers when:
- Human operators are actively involved in the workflow
- You need the expression editor for complex calculations
- Workflow includes UI interactions (buttons, forms, scans)
Example: "Operator presses 'Complete Run', app calculates OEE from machine data and displays result on screen"
DataOps platform
Use DataOps platform when:
- You need sub-second data resolution (<1 Hz)
- You need to aggregate data from non-Tulip-supported protocols
- Time-series database is required for advanced analytics
- Customer has existing investment in the platform
Edge device
Use an Edge device + Node-RED when:
- You are running a proof-of-concept or pilot project
- Customer has strong embedded systems expertise
- Regulatory requirements prevent cloud-based logic execution
- Budget constraints prohibit subscription software
Further reading
- Machine monitoring app-based architecture patterns
- Create your first machine trigger
- Automations
- Edge IO
- Node-RED
Did you find what you were looking for?
You can also head to community.tulip.co to post your question or see if others have solved a similar topic!