Every hospital knows the feeling. Ambulances queue outside, the waiting room fills, trolleys line the corridor and the emergency department is full of patients who have already been admitted but have nowhere to go. Crowding is one of the most visible signs of pressure in any health system — and one of the most damaging, for patients and staff alike. A hospital digital twin offers a different approach: see the surge coming, understand where the real constraint sits, and act hours earlier.
Why ED crowding is a whole-hospital problem
It is tempting to treat emergency department crowding as an emergency department problem. In reality, the front door is usually where pressure becomes visible, not where it starts. The most common cause of crowding is access block: patients who need admission cannot move to a ward because inpatient beds are full. Those beds are full because discharges happen late in the day, because patients wait for diagnostics or transport, or because elective admissions compete for the same capacity.
This is exactly the kind of interconnected system a digital twin is designed to model. If you are new to the concept, our guide What is a hospital digital twin? explains how spatial models, live data and simulation fit together.
How the twin forecasts demand
Emergency attendances are not random. They follow strong patterns by hour of day, day of the week and season, with additional effects from weather, public holidays, local events and infectious disease activity. A digital twin learns these patterns from your own historical data and combines them with live signals to forecast:
- Arrivals to the emergency department over the next hours, by acuity where possible.
- Admissions likely to result from those arrivals, by specialty.
- Bed availability across the hospital, based on expected discharges and transfers.
- The gap between the two — the point at which boarding is likely to begin.
Crucially, the twin does not forecast the emergency department in isolation. It connects predicted admissions to the wards, critical care beds and theatres that will need to absorb them, which is what turns a forecast into something operational teams can act on.
From forecast to action
A prediction is only valuable if it changes what people do. The twin translates forecasts into recommended actions, matched to your hospital’s escalation plans:
- Prioritising discharges on the wards that will receive the most admissions, and supporting earlier-in-the-day discharge.
- Moving medically fit patients to a discharge lounge to free beds sooner.
- Opening escalation or surge capacity before the queue forms rather than after.
- Adjusting the next day’s elective schedule when emergency demand is expected to peak.
- Redeploying staff, porters and cleaning teams to where turnaround will matter most.
Because the twin understands the physical layout of the hospital, recommendations are specific: not just “free up medical beds”, but which ward, which bays and by when.
The shift that matters: moving from reacting to a full department to preparing for one. Even a few hours of warning gives teams time to discharge, transfer and staff up calmly instead of in crisis mode.
Simulating surge scenarios
Beyond day-to-day forecasting, the twin is a safe place to rehearse. Planners can test questions that are difficult or risky to answer in the real hospital:
- What happens if winter respiratory illness arrives two weeks earlier than usual?
- How would a major incident affect bed availability over the following 48 hours?
- If we close a ward for refurbishment, where should displaced activity go?
- Would an additional assessment area or a new discharge pathway actually reduce boarding?
Each scenario can be run many times with different assumptions, giving leaders a range of likely outcomes rather than a single guess.
What teams see day to day
For site managers and command-centre teams, the twin becomes a shared operational picture: current occupancy across the hospital, predicted pressure for the coming hours, and a prioritised list of actions. For ward leaders, it highlights which expected discharges matter most today. For executives, it provides an honest view of risk across the whole site. Everyone works from the same information, which reduces the number of phone calls and meetings needed to coordinate a response.
The data behind it
Emergency department surge prediction relies on near-real-time admission, discharge and transfer events, bed states, attendance data and, ideally, staffing information. Our guide to data integration for hospital digital twins explains how these sources connect securely and without disrupting clinical systems.
Measuring impact
Agree a small set of metrics before you start, and measure them consistently. Useful measures include:
- Time from decision to admit to arrival on a ward.
- Total boarding hours in the emergency department.
- The proportion of patients who leave without being seen.
- The share of discharges completed before midday.
- The number of escalation events and how early they were triggered.
Our guide on building the business case for a hospital digital twin explains how to set baselines and present results to your board.
Getting started
Most hospitals begin with the pathway between the emergency department and acute medical wards, where the link between arrivals and bed demand is strongest. Once forecasting is trusted there, the model naturally extends to surgical pathways — see optimising operating theatre utilisation with simulation — and to the wider hospital.
Want to see how surge prediction would work in your hospital? Book a walkthrough with our team, or explore more use cases and guides.
