Head of Logistics
Fleet and flow
Fleet utilisation, patrolling, milk-run delays — and where to cut trucks without starving lines.
Why an operational digital twin
A 3D scan shows your plant. An operational twin shows what is happening in it.
Scan, simulation, operational and predictive twins answer different questions. Only the operational one runs the shift — and it builds the spatial dataset every later AI layer needs.
Four layers people confuse
Each layer is useful. The mistake is expecting one to do the job of another.
| Layer | What it answers | What it cannot do |
|---|---|---|
| 3D scan / CAD / BIM | Where are the walls and racks? | Show the forklift circling the buffer now |
| Simulation twin | What if we change the dock layout? | Update itself every shift |
| Operational twin twinzo | What is happening on the shopfloor right now, and what should we do? | Replace ERP / MES / WMS as systems of record |
| Predictive twin Unlocking | What will starve Line 3 in 20 minutes? | Needs the live spatiotemporal history the operational twin builds |
The layout file is not the plant
When Line 3 is waiting, a static layout file cannot say why. Last shutdown’s scan still shows walls and racks — not the milk-run that is already late. Internal moves still start on the radio, so the search happens on foot while the model does not update. Digital twin on the CapEx slide usually means a 3D shell →
Simulation answers a capital question before the purchase decision is made. When the run ends, the scene waits for the next scenario — it does not run the shift.
What twinzo adds
Live RTLS positions, IoT readings and production KPIs sit on the 3D model people already recognise. ERP, MES, WMS and SCADA keep owning orders, OEE and inventory. Supervisors stop walking the aisle to find a tugger; they open the twin.
That live spatiotemporal history is what later AI and predictive layers train on. Without it, “predictive twin” demos stay slideware.
Five steps from average to best in class
From siloed systems to one live plant model. Each step raises maturity and value — and keeps what you already built.
00
Starting point
Average factory
Siloed systems
- Data in separate tools
- No shared place-and-time picture
01
Correlation
Position analytics
See where everything moves
- Live 3D map of every asset
- Heatmaps show where time goes
02
Causality
Logistics orchestration
Every move gets a purpose
- Call material by button, scan or MES
- Every trip linked to its order
03
Scale
End-to-end logistics
From supplier to an assembly line, on one map
- GPS hands over to indoor RTLS
- Gaps between sites measured
04
AI analytics
AI for intralogistics
Understand why, then act
- AI finds causes in position data
- Targets idle time and empty travel
05
The big vision
Operational digital twin
Best in class is here — the whole plant, live, in one place
- MES, WMS, ERP, SCADA in one model
- History feeds predictive operations
Start small. Prove value at every step. Nothing you build early gets thrown away.
Who it is for
Plant Manager
A business case
Idle minutes and recovered output in dollars — not another dashboard to maintain.
IT / OT
One spatial model
Connectors into systems you already trust. Cloud, on-prem or hybrid.
What happens next
Benchmark your Factory
A 30-minute assessment against plants of similar size.
Scope a 2–4 week pilot
One hall or flow, with success metrics agreed up front.
Connect the systems of record
The ones that already own the truth: ERP, MES, WMS, SCADA.
Check the approach before the assessment
A short read from the engineers who run these deployments.
-
Start with the decision, not the technology →
Match accuracy and scope to the logistics job before you buy sensors.