The problem: the data exists, the answer does not
An offshore platform, a hospital or a distribution hub already produces more data than anyone reads. The building management system logs every setpoint. The maintenance system holds every work order. Cameras cover every deck and corridor. The ERP knows what each hour of output is worth.
Each system shows one piece. None of them tells the operations director which single constraint is holding the whole site back, or what it costs per day. So the morning meeting compares spreadsheets, and the most expensive problem stays where it is.
In the demo scenario of a gas asset, one dehydration constraint holds production below nomination. The twin names it and prices it at about $142k a day deferred. The figure is modelled on demo data, but the pattern is real on most sites: one bottleneck, many systems, nobody joining them.
Who an industrial digital twin is for
- Operations and asset directors. on offshore energy, process and utility sites who need one live picture instead of six dashboards.
- HSE leads. who want safety checks running on every camera, all shift, not on a sample walk.
- Hospital, data centre and logistics hub operators. where downtime, comfort or throughput carries a direct cost.
- Hotel and office owners. running FM contracts, energy and comfort targets across one or many buildings.
What the twin does
A platform like Synopton works in four steps.
It turns existing cameras into sensors. Computer vision runs on the cameras the site already owns. It checks hard hats and high-visibility clothing, watches exclusion zones and flags work at height, with every detection tied to a camera, a time and a location on the model.
It joins every system into one live model. Building management, SCADA, maintenance, ERP, IoT sensors and CCTV feed one 3D twin of the asset, so a reading is always attached to the equipment it comes from.
It answers in plain English, with sources. A manager asks "why are we below nomination?" and gets the binding constraint, the numbers behind it and the record each number comes from. Answers come only from the twin's own data.
It shows its working. Every figure is labelled measured, modelled, published or assumed. That label is what lets a director take the answer into a board meeting.
What it connects to
Read-only connectors to BMS, SCADA, CMMS, ERP, IoT platforms and CCTV. Nothing is written back into operational systems. The twin runs in your cloud tenant, on your own servers or hosted, and the same platform covers ten kinds of asset, from offshore platforms and warehouses to hotels and hospitals.
How Codset builds digital twins
We build digital twins end to end: the data platform underneath, the 3D model, computer vision on existing cameras, the AI layer that answers with citations and the dashboards leadership reads. Our data and digital twins service covers the whole build, and cloud, security and sovereign deployment covers sites with strict hosting rules.
Two platforms built by the Codset team show the range. Systwin combines policy governance, HSE compliance and real-time monitoring for energy operators. Gemini Atlas turns data from many sites into one executive briefing.
Most engagements start with a pilot on one asset: connect the systems in scope, build and calibrate the twin, then run real decisions through it. For sector examples, see the logistics digital twin and the hotel twin pilot plan.
What it costs
A dedicated Product team starts at €13,900 a month on a 12-month term, and a Startup team at €7,900 a month. The team configurator prices any mix of roles, and the code and IP built for you belong to you from the first commit.
Find your leak
Tell us about the site: the asset type, the systems you run and the question you cannot answer today. Write to info@codset.com or send a brief, and we reply with a pilot plan for one asset.
Questions buyers ask
What is an industrial digital twin?
A live model of a site or asset that combines sensor, camera, maintenance and business data in one place. It shows what is happening now, explains why and lets you test a change before you make it.
Do we need new sensors or cameras?
Usually not. Most twins start from the cameras and systems a site already has. Computer vision turns existing camera feeds into safety and operations sensors.
Does the twin change settings in our control systems?
No. Connectors are read-only. The twin recommends, people decide, and nothing is written back into SCADA or the BMS.
Are the figures from a real site?
The $142k a day figure comes from a modelled demo scenario. A pilot replaces it with your own data.
Can it run on our own infrastructure?
Yes. It can run in your cloud tenant, on your own servers or as a hosted service.