A spreadsheet only one expert understands
We turn it into a model anyone on the team can run.
Mathematical optimisation & AI for planning and allocation
The best plan within your rules. And the proof.
acmedis builds optimisation and AI models for decisions that must be both optimal and defensible, in healthcare, housing and the public sector. Senior specialists, a first working model on your own data in three weeks, and reasoning you can trace to your rules.
- Senior specialists do the work
- First model in 3 weeks
- Reasoning you can show
Why acmedis
Skilled, trusted, and fast enough to start.
Skilled: a plan you can inspect
We are optimisation specialists who also use modern AI. The plan follows your rules, the assumptions are written down, and where the solver can measure it you see how close the plan sits to the best feasible solution.
Trusted: transparent by design
Every outcome is traceable to your data, rules and assumptions, so you can explain it to a board, regulator or council. Scope and price are agreed before we start. If it will not pay off, we say so.
Agile: seniors, weeks, no platform to buy
No junior team handed the work after the pitch. The founders build the model, show progress every two weeks, and connect it to the systems you already use.
Where we help
Where we make the difference
Decisions you must defend
We make the trade-offs and scenarios visible for a board, regulator or council.
Rules your standard software cannot hold
We build the optimisation engine that can, alongside the systems you already use.
An AI pilot that never reached daily work
We connect AI to a real decision, with acceptance criteria before go-live.
Services
What we do
Optimisation is the core. AI and data are how the model works in daily practice.
Decision optimisation
Rosters, capacity, maintenance programmes, portfolio and budget choices. We model your real rules and targets and compute the best feasible plan.
AI in the decision chain
Language models that read requests, reports and contracts, and forecasts of demand and capacity. Always feeding a decision, never a chatbot on the side.
An independent AI check
A short assessment of your data, systems and obligations, including under the EU AI Act, so you invest where it pays off.
Approach
How we work
A fixed first step. Then build only if it pays off.
1
Quick Scan
Three weeks, fixed price. A working mini-model on your data, a business case in euros and hours, and an honest go or no-go.
2
Build
Two-week sprints. Visible progress every sprint, and acceptance criteria agreed before go-live.
3
Operate & improve
We monitor, retrain and adjust as your organisation changes.
Comparison
How we compare
| Standard planning software | Large consultancies | acmedis | |
|---|---|---|---|
| First result | After selection and implementation | After a discovery phase | Working model on your data in 3 weeks |
| Unusual rules | Within what the product supports | Depends on the team | Modelled exactly, rule by rule |
| Who does the work | Implementation team | Often mixed seniority | The senior specialists you meet |
| Explainability | Varies by product | Varies | Assumptions and trade-offs written down |
| Your systems | Inside the product | Varies | Connects to what you already use |
Sectors
We lead with the sectors we know from the inside: healthcare, housing and the public sector.
Healthcare
OR, bed and outpatient capacity, strategic staffing and scenarios, working alongside your existing rostering system.
Talk about your contextHousing & real estate
Multi-year maintenance and sustainability programmes, portfolio strategy, and allocation you can explain.
Talk about your contextPublic sector
Budget and subsidy allocation, inspection planning, and capacity for public services.
Talk about your context
Start small, then expand
New clients start with a Quick Scan. Advisory hours stay available later, for teams that already work with us.
Start here
Quick Scan
3 weeks
- Fixed price, agreed up front
- Working mini-model on your data
- Business case and an honest go or no-go
Project
Scoped
- Fixed price per milestone
- Two-week sprints
- Acceptance criteria before go-live
Operate & improve
Ongoing
- Monitoring and retraining
- Model changes as you change
- For a model already in use
Bring us your hardest planning or allocation puzzle.
In 30 minutes we will tell you honestly whether optimisation can improve on your current plan, and what it would take. You work with Niels Maijers and Jip Maijers directly.