Imagine an emergency department with 16 treatment bays. Patients arrive, receive care and leave. If the department begins to accumulate patients, what should we change? Add more bays? Increase observation capacity? Make more hospital beds available?
Each option sounds reasonable. But how do we know which one will make the greatest difference?
The difficulty is that an emergency department does not operate as a collection of independent rooms. Patients move through a sequence of stages, and each stage depends on the resources available at the next. A change in one part of the system can improve performance, leave another problem untouched or shift pressure elsewhere.
To explore these interactions, I developed the DFU simulation model as part of my Master’s Final Project in Hospital Engineering and Architecture. The model has 55 configurable parameters and represents the patient’s journey through the emergency department minute by minute, from arrival to departure.
I used it to conduct eight controlled experiments, each examining a different factor affecting the department’s operational capacity.
The aim was not to predict exactly what will happen in every hospital. It was to understand how the system behaves when specific conditions change.
The experiments produced seven lessons.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 1. Time matters as much as the number of patients
Suppose an emergency department receives 10 patients per hour and has 16 treatment bays. If demand remains constant, should the department’s performance remain stable?
Not necessarily.
In the first experiment, I varied the average time patients occupied a treatment bay while keeping demand and the number of bays constant.
As occupancy time increased, patients began to accumulate. The same number of arrivals had to be accommodated by the same number of resources, but those resources became available less frequently.
This is a fundamental distinction between physical capacity and effective capacity. Counting the treatment bays tells us how many resources exist. It does not tell us how frequently those resources can serve incoming patients.
The implication is straightforward: understanding capacity requires examining not only how many resources are available, but also how long patients occupy them.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 2. Adding treatment bays can improve performance without solving every problem
If treatment bays are under pressure, adding more seems like the obvious response. And the second experiment confirmed that this intervention can help.
I kept demand at 10 patients per hour and average bay occupancy at 90 minutes, then varied the number of treatment bays from 8 to 24.
As the number of bays increased, average length of stay decreased and bay occupancy fell.
But another result emerged. More patients could move through the treatment bays without necessarily finding enough space to continue through the rest of the department. In the simulated configurations with 12 to 24 bays, the corridor reached its maximum capacity.
The expansion improved performance in one part of the system, but congestion persisted elsewhere.
This does not mean that adding bays is a mistake. It means that the effect of an expansion depends on what happens after patients leave those bays.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 3. Observation is part of the department’s overall capacity
What if the problem is not simply the number of treatment bays, but the ability to accommodate patients after the initial treatment phase?
The third experiment examined this question by varying observation capacity while keeping demand and the number of treatment bays constant.
Increasing observation capacity reduced average length of stay and the number of patients left without space because of overcrowding. However, bay occupancy increased slightly, and the corridor continued to reach its maximum capacity.
Once again, improving one part of the system did not eliminate every capacity problem.
The lesson is not that observation capacity matters more than treatment-bay capacity. It is that the two cannot be planned independently. Patients need to move between stages of care, and the resources available at each stage influence the operation of the whole department.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 4. The daily patient total hides the pressure created by arrival patterns
Imagine two days with exactly 240 patients arriving at the emergency department.
On the first day, arrivals are distributed evenly, at 10 patients per hour. On the second, arrivals are concentrated into periods of 6 and 14 patients per hour.
The total is identical. The hourly distribution is not.
In the fourth experiment, average length of stay increased from 4 hours and 51 minutes in the uniform-arrival scenario to 5 hours and 47 minutes in the scenario with concentrated arrivals.
Why? Because the timing of arrivals changes how many patients compete for resources at the same time.
Quieter periods do not automatically undo the pressure created during busier periods. Patients who have already accumulated must still progress through the department.
The implication for planning is important: daily activity figures are useful, but they can conceal the periods when a service faces its greatest operational pressure.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 5. Patients do not place the same demands on resources
The number of patients arriving is only part of the picture. The care they need also affects how long they occupy the department’s resources.
The fifth experiment introduced four modelled patient profiles, with treatment-bay examination times ranging from 25 minutes for mild cases to 100 minutes for severe cases. These were parameters used in the simulation, not universal clinical standards.
The model compared different patient compositions under different treatment-bay configurations.
In one scenario, with 15% mild patients, 20% moderate patients, 50% multimorbid patients and 15% severe patients, average length of stay was 6 hours and 22 minutes with 8 treatment bays, compared with 4 hours and 25 minutes with 16 bays.
Yet the number of episodes classified as having no space available in the emergency department remained at 205 in both configurations.
The result illustrates two important points. Patient composition influences the workload placed on the system, and an improvement in one performance measure does not guarantee an improvement in every other measure.
Planning capacity therefore requires more than counting arrivals. It also requires understanding the resource requirements represented by the patients being treated.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 6. The exit from the emergency department matters too
What happens when patients are ready to leave the emergency department for hospital admission, but no inpatient bed is available?
The sixth experiment examined the effect of changing the number of beds available for hospitalisation.
Increasing inpatient-bed capacity reduced the number of patients left without space to continue their journey. However, average length of stay changed very little in the scenario analysed.
At first, this may seem contradictory. If more patients can move into hospital, why does average length of stay barely change?
Because different indicators describe different aspects of system performance. A change can reduce a specific form of congestion without producing a large change in the average time patients spend in the department.
The result also highlights the importance of the hospital beyond the emergency department itself. The ability to accommodate patients after they leave emergency care can influence whether they can continue through the system.
An emergency department cannot be understood entirely in isolation from the processes and resources that determine its patients’ next destination.
La maestría como actitud, la maestría como actitud, la maestría como actitud
Lesson 7. The system must be observed over time, and the patient’s journey must continue beyond the treatment bay
A final question remains: are we looking at the department for long enough, and are we following the patient far enough, to understand what is happening?
The seventh experiment examined the effect of the simulation horizon. The eighth examined what happened after patients left the treatment bay but remained within the care process.
These experiments addressed two limitations of a static view of emergency capacity.
First, congestion can develop progressively. A department may appear manageable at one moment while patients accumulate as demand, resource occupancy and available space interact over time.
Second, a patient leaving a treatment bay has not necessarily left the emergency department. The patient may still need to complete another stage of care, and that stage requires an appropriate place within the service.
In the eighth experiment, for example, 44 patients left the treatment bay but continued within the care process during the 24-hour simulation horizon. Of these, 23 passed through the corridor. In the longer simulation horizons, 47 patients left the bay and remained within the process, with 23 passing through the corridor.
These are simulated results, not observations from a real hospital. They illustrate why measuring treatment-bay occupancy alone can miss part of the patient’s journey and part of the department’s operational workload.
The implication is architectural as well as operational: the department must provide for the stages that connect treatment with the patient’s eventual departure.
La maestría como actitud, la maestría como actitud, la maestría como actitud
What connects these seven lessons?
The experiments examined different variables, but they point towards the same conclusion.
Capacity is not determined by a single number. It emerges from the interaction between patient arrivals, resource occupancy, patient characteristics, available spaces and the ability to move patients from one stage of care to the next.
Adding treatment bays can help. Increasing observation capacity can help. Making more inpatient beds available can help. But the effect of each intervention depends on the conditions of the system in which it operates.
This is where simulation becomes useful.
A configurable model allows us to change one parameter, hold other conditions constant and observe how the system responds under a defined scenario. It helps us investigate questions before committing to changes in a real department.
That does not make every simulated result a universal prediction. The findings depend on the model’s assumptions and parameters, and their application to a particular hospital requires appropriate data and further assessment.
The value lies in making the questions more precise: where is capacity being lost, what is preventing patients from progressing, and which intervention is likely to address the constraint that matters?
La maestría como actitud, la maestría como actitud, la maestría como actitud
From counting spaces to understanding the system
The eight experiments began with specific questions about treatment-bay occupancy, the number of bays, observation capacity, arrival patterns, patient composition, inpatient beds, simulation duration and the continuity of the patient’s journey.
Together, they suggest a different way to approach emergency department planning.
Instead of starting with the number of rooms or beds a department should contain, we should first understand the patients it serves and the care process they need to complete. We can then examine how the available spaces and resources support that process, where patients accumulate and what prevents them from moving forward.
Only then can we make better-informed decisions about capacity and design.
The purpose of simulation is not to predict every future event. It is to understand how a system behaves, identify its vulnerabilities and make better-informed decisions before changing it.
About this research
This article draws on a four-month research project conducted as part of my Master’s Final Project in Hospital Engineering and Architecture at the University of Cádiz, Spain, which received a grade of 9.5/10.
The research examines demographic trends, hospital designs and functional plans, emergency care processes, and the perspectives of emergency department professionals gathered through surveys. It culminates in the development of a digital twin simulation model designed to explore multiple scenarios and analyse how different factors affect emergency department operational capacity.