Staffing and patient assignment were planned from experience and spreadsheets. When demand shifted, the team found out on the day — through queues, overtime and missed targets. They wanted to know what a change would do before they made it.
Case study — Healthcare
A digital twin for patient flow and staff assignment.
How a healthcare operations team began testing staffing decisions on a simulated week before making them on a real one.
Client
Healthcare operator (name withheld)
Disciplines
Simulation, machine learning, software
Engagement
Feasibility study, build, support
Delivered
Simulation model, priority and assignment models, dashboard
Fig. 1 — The ward corridor where most of the waiting happened.
01The problem
02Our approach
We started with a fixed-price feasibility study on historical activity data, to check whether a simulation could reproduce a real week closely enough to trust. Once it could, we added priority and assignment models on top, so the team could compare staffing plans side by side against the same simulated demand.
“We stopped arguing about staffing plans and started testing them.”
Operations lead, client team — quote to be confirmed
PLAN A · CURRENT
42 min
avg wait
PLAN B · +1 NURSE 2–6 PM
27 min
avg wait
Plan B saves 6 hrs of overtime a week
Fig. 2 — The staffing simulator: two plans tested on the same simulated week.
03What we delivered
A simulation model of patient flow across units, priority and assignment models for staff allocation, and a dashboard where planners can run a scenario and read the result in operational terms: waiting time, utilisation and overtime.
04The result
Planners now test a staffing change against a simulated week before rolling it out, and the conversation has moved from opinion to evidence. Replace this paragraph with the measured outcome once the client has approved the figures.