Pharma Factory Journey
Batch 112A just failed compliance. No one can explain why.
The Challenge
Pharmaceutical manufacturing demands:
- Complete traceability - Every batch, every reading, every decision
- Audit-ready data - Inspectors need proof, not promises
- Rapid response - Compliance failures cost millions
- Cross-system visibility - Data scattered across historians, SCADA, MES
Traditional approaches fail because:
- Data lives in silos
- Query logic isn't tracked
- AI answers can't be audited
- Nobody owns the full picture
The Solution: Governed AI
OpenIndustrial provides the Control Plane that connects your data and governs your queries:
| Capability | How It Helps |
|---|---|
| Azi-built queries | Natural language to KQL, with your approval |
| Full audit trail | Every query, who ran it, when, what parameters |
| MCP-native AI | Connect Claude or Copilot to governed data |
| Role-based access | QA sees QA data, Ops sees Ops data |
This isn't AI with guardrails. This is a fundamentally different architecture—where AI proposes and you decide.
What You'll Build
By the end of this journey, you'll have:
| Component | Description |
|---|---|
| Connections | Temperature, pressure, batch data from your systems |
| Surfaces | Cleanroom monitoring, QA gating, batch tracking |
| Warm queries | Compliance checks, threshold monitoring, trend analysis |
| API endpoints | REST APIs for LIMS, MES, or custom dashboards |
| MCP integration | Any AI tool querying governed data |
Two Tracks
Track A: Hybrid Integration
"Govern Without Replacing"
You have existing systems—DeltaV, HighByte, SCADA, historians. This track adds governance on top:
- Connect to existing data sources via IoT Hub
- Create surfaces that organize your data
- Build compliance queries with Azi
- Get API endpoints without changing infrastructure
Track B: Full System
"Author It All"
Starting fresh or want complete control. Build everything with OpenIndustrial:
- Design your connection architecture from scratch
- Create a complete surface hierarchy
- Build a comprehensive warm query library
- Establish team permissions and workflows
Prerequisites
| Requirement | Details |
|---|---|
| OpenIndustrial account | Free tier works for this journey |
| Azure subscription | For IoT Hub connections (Hybrid track) |
| Data sources | At least one: simulator, historian, or sensors |
| Time | ~2 hours for Hybrid, ~4 hours for Full |
Don't have data sources? Configure a simulator to follow along. The patterns are the same.
Category Metrics
At the end of this journey, you'll measure:
| Metric | Target |
|---|---|
| Time to First Governed Query | < 30 minutes |
| Compliance queries built | 5+ |
| Audit coverage | 100% of queries logged |
| AI integration | At least 1 MCP-native AI connected |
Why This Matters for Compliance
| Before Control Plane | After Control Plane |
|---|---|
| "The AI said it was compliant" | "Here's the query, who approved it, and the audit trail" |
| Black-box recommendations | Transparent, reviewable logic |
| Vendor-locked insights | Portable, governed data |
| Scattered evidence | Single source of truth |
Choose Your Track
Hybrid Integration
Add governance to existing systems. Connect historians, SCADA, MES—without replacing them.
Full System
Build from scratch. Complete control over every connection, surface, and query.
What's Next
| Track | Link |
|---|---|
| Start with Hybrid | Hybrid Track → |
| Start with Full | Full Track → |
| See common patterns | Patterns → |