VÖRNTEC
|
BGV Product Demo
01
/ 03
The Problem
~£30M
estimated annual loss per operator
due to undetected gas leaks in upstream and midstream operations.
What We Do
We find gas leaks
using AI
on the data operators already have.
Every operator runs SCADA. We built an ML pipeline that reads that telemetry and detects anomalies before they become leaks — no new hardware.
No new sensors
No hardware
Existing SCADA data
🔧
Current solutions
Focus on expensive hardware — OGI cameras, drones, satellites, new sensor networks.
💬
Validated
Spoke with 20+ companies — the pain point is real and current solutions miss early signals.
VÖRNTEC
|
How it works
02
/ 03
How It Works
📡
SCADA + ERP
Pressure, temperature, flow, valve states
→
🧠
3-Layer ML Engine
Designed for scalability
Statistical baseline
Unsupervised ML
Supervised classifier
→
💬
Microsoft Teams
Conversational AI agent alerts + context
→
📋
Work Order in ERP
Auto-created in SAP PM, IBM Maximo, or any CMMS
Live Demo →
14-step interactive walkthrough on real Petrobras well data — from SCADA ingestion through fault classification to work order creation.
VÖRNTEC
|
Where we are
03
/ 03
Where We Are
Done
✓
ML pipeline
— validated on Petrobras 3W (32.8M obs, 1,119 wells)
✓
Integrations
— Teams agent, SAP PM work order flow
✓
Interactive demo
— full detection → alert → WO journey
In Progress
→
1st pilot pipeline
— conversations with UK, Norway & Argentina operators
→
Outreach
— cold + warm intros across upstream & midstream
Next
○
First no-cost pilot
on real operator SCADA data
○
Prove transfer
from public dataset → private production data
Feedback Ask
01
How can we improve our sales funnel from cold outreach to data handover?
02
How can we sharpen our story to land better with technical decision-makers?