Digital twin development at iUNI means building a VR-ready virtual replica of a factory, plant, warehouse or facility at 1:1 scale from CAD data, then connecting it to live sensor data so it can be used for planning, training, monitoring and virtual commissioning. iUNI is the Wartens group's digital twin and immersive company, and a typical project takes 12 to 24 weeks.
This brief sets out the features, the four deliverables, how the build is usually phased, what the twin will not do, and how to decide whether you need one. Details are taken from iUNI's digital twin development page.
What a digital twin is, and what it is not
The Digital Twin Consortium defines a digital twin as a virtual representation of real-world entities and processes, 'synchronized at a specified frequency and fidelity'. The synchronisation is what separates a twin from a 3D model or a one-off simulation. If the virtual line does not know what the real line is doing, it is a model, which may be all you need. We unpack that distinction in virtual commissioning vs digital twin vs simulation.
Key features of an iUNI digital twin
- 1:1 scale replicas from CAD data, so layouts, clearances and reach are measured, not estimated.
- Real-time IoT data integration and visualisation, showing the live state of equipment on the model.
- Predictive maintenance and anomaly detection, using the same data stream.
- What-if scenario simulation, to test a layout change, a new product mix or a different shift pattern before committing.
- A web-based 3D viewer that needs no software installation.
- Integration with SCADA, PLC and ERP systems, so the twin reads from the systems you already run.
What you get: the four deliverables
| Deliverable | What it is | Who uses it |
|---|---|---|
| Digital twin platform with 3D viewer | The model itself, viewable in a browser | Engineering, operations, management |
| IoT integration layer | Connections to sensors, SCADA, PLC and ERP data | Controls and IT/OT teams |
| Analytics and reporting dashboard | Trends, anomalies and what-if results | Operations and maintenance |
| VR/AR visualisation module | Immersive walk-throughs and training views | Training, safety, sales |
Use cases that justify the build
iUNI lists five use cases: factory and production line monitoring, virtual commissioning rehearsal, building management and energy optimisation, infrastructure lifecycle management, and supply chain visualisation. The strongest business cases usually combine two of them. A twin built first for virtual commissioning of a new line can be connected to live data after start-up and become the operations twin, so the modelling cost is paid once.
That reuse is where the Wartens group's background matters. Because the team are automation engineers first, the twin is driven by the same control logic the plant will run, and virtual commissioning tests the real PLC program against the model before anyone goes to site.
Why digital twins will be the backbone of industry in the future
Siemens Knowledge Hub
This video is hosted on YouTube. Playing it loads content from Google, which may set cookies.
Watch on YouTubeHow a 12 to 24 week build is usually phased
The phases below are indicative. The real plan depends on scope, the quality of your CAD data and how many systems the twin must connect to.
- Discovery (one to two weeks). Agree the use cases, the area in scope, the data sources and the success measures. Collect CAD, layouts, P&IDs, tag lists and network drawings.
- Model build (four to eight weeks). Convert and clean CAD into a 1:1 model at the fidelity the use case needs. Behaviour is added for equipment the twin must animate.
- Integration (three to six weeks). Connect the IoT layer to sensors, SCADA historians, PLCs or ERP, usually through standard interfaces such as OPC UA. Map every tag.
- Analytics and views (two to four weeks). Build the dashboard, anomaly rules and what-if scenarios, plus the VR/AR module.
- Acceptance and handover (one to two weeks). Test against the agreed measures, train users and hand over documentation.
What to have ready before you start
Two data points help the integration phase. If your plant already exposes data over OPC UA, connection is much faster. And if you supply equipment into Germany, structuring the twin's data for the Asset Administration Shell on request can save rework later. For the reference model behind manufacturing twins, see our guide to ISO 23247.
Measuring whether the twin is earning its keep
Agree the measures in discovery, not after go-live. For a monitoring twin, that might be the time from a fault occurring to the right person seeing it, or the share of unplanned stops with a recorded cause. For a commissioning twin, count software defects found before site against a comparable earlier project. For an energy twin, compare consumption per unit before and after the changes it suggested. Each measure needs a baseline taken before the twin is live, otherwise the result is an opinion.
What a digital twin will not do
- It will not fix poor data. A twin shows what the sensors report. If instruments are uncalibrated or tags are mislabelled, the twin will faithfully display the wrong thing.
- It does not replace the control system. The twin reads from SCADA and PLCs; it is not a safety system and should not be in the control path.
- It needs an owner. Plants change. Without someone updating the model after modifications, a twin drifts away from reality within months.
Cost and next step
iUNI publishes price ranges on its pricing page; at the time of writing, digital twin development is listed from 拢20,000 per project, with the final figure depending on scope. Every Wartens engagement starts with a written scope, schedule and price.
To discuss a twin for your line, plant or building, start with our digital twins and virtual commissioning service, see NIST's digital twin research for background, or send us your layout and use cases.
