The problem: every winter arrives as a surprise
Patients wait hours in the emergency department for a bed. Wards overflow in the first week of January. A critical medicine runs short while suppliers are closed for the holidays. A scanner fails at the worst moment, and nobody saw the warning signs in the building data.
The information to see all of this coming already exists. It sits in the EHR, the patient administration system, the lab, pharmacy, the theatre list, the rota and the building systems. Each answers its own question. None of them forecasts the peak and tells the operations team what to do about it.
What the twin finds in a demo hospital
In a demo hospital with illustrative figures, the twin shows:
- Peak on 2 January: 104% of beds. A holiday occupancy forecast with an 80% interval. With the plan the twin recommended, the peak drops to 93%.
- Four critical medicines short by New Year. Alteplase, an IV antibiotic, salbutamol and O-negative blood, against forecast use while suppliers are closed. Order by 12 December.
- 18 more beds next January. Bed demand for 2027 by month, with the drivers: more patients over 75, the flu season and a new day unit.
- MRI-2 reaches its helium quench line in 12 days. Repair and top-up cost €22.4k, against €444k and 10 days out of service if it quenches.
Who a hospital digital twin is for
- Chief operating officers and site directors. who own capacity, flow and the winter plan.
- Bed managers and emergency department leads. who need a forecast they can act on, not last year's spreadsheet.
- Pharmacy and estates teams. watching medicines, the cold chain and critical equipment.
- Hospital groups and clinic networks. comparing sites on the same measures.
How it works
Connect. Read-only links to the EHR, PAS and ADT, the lab, pharmacy, the theatre list, the rota, the BMS and the cameras you already have.
Model. One patient-flow model and a 3D twin of every ward, theatre, fridge and scanner, built from your drawings.
Monitor. Beds, the emergency department, theatres, the cold chain and critical equipment, live, on one screen.
Forecast. Occupancy next week and next holiday, medicines over supplier closures, beds needed next year and equipment runways, each with its drivers and an interval. Forecasts are backtested on your own past winters before anyone relies on them.
Explain. Ask why Ward 7A will overflow on 2 January, in plain language, and the answer cites the twin's own data.
Decide. Try a plan as a what-if, approve it as a decision record with an owner, and track the measured result.
Privacy, systems and installation
The twin connects to the EHR over HL7 FHIR, PAS and ADT over HL7 v2, PACS over DICOM, and to the lab, pharmacy, theatre list, rota, nurse call, RTLS, BMS, cameras, cold-chain sensors, SAP, Microsoft 365 and Power BI. Connectors are read-only, and nothing is written back to clinical systems.
Analytics run on pseudonymised data, with single sign-on, role-based access and an audit log of every view and decision. It runs in your Azure, AWS or Google Cloud tenant, on your own servers, fully air-gapped, or hosted by us.
How Codset builds it
We build hospital digital twins end to end through our data and digital twins and AI and agentic systems services, with sovereign deployment where data must stay inside your perimeter. A pilot runs for 90 days: connect in weeks one and two, build and calibrate until the twin is usable from week six, then run real decisions through it to week twelve. Start by early autumn and your team plans the winter peak with it.
A named delivery lead from the same team supports it after the pilot, and the code and IP built for you are yours. The free hospital twin pilot plan sets out what we connect and what you get by week six.
Plan winter before it arrives
Tell us about the hospital: the number of beds, the systems you run and where the pressure builds. Write to info@codset.com, send a brief or download the pilot plan.
Questions buyers ask
Does the twin touch patient care or clinical systems?
No. Connectors are read-only. Accepting a recommendation drafts a decision record for people to agree, and nothing is written back.
How is patient data protected?
Analytics run on pseudonymised data, with single sign-on, role-based access and an audit log. Your information governance lead is part of the pilot from day one.
How accurate are the forecasts?
Each forecast is backtested on your own past winters and shows its drivers and an 80% interval. When reality differs, the twin records it and the next forecast learns from it.
Are the figures from a real hospital?
They come from a demo hospital and are illustrative. A pilot replaces them with your own data.