Optimisation that stays practical.
Qatalyst converts operational decisions into ready-to-solve optimisation models, selects the most suitable solver, and returns explainable outputs. Classical optimisation is the default. Hybrid and quantum options are used selectively when scale and coupling justify it.
Automatic model generation
Turn objectives, constraints, and operational rules into optimisation models without manual maths.
Solver selection
Choose classical, hybrid, or quantum back-ends based on structure, runtime, hardware, and cost.
Explainable decisions
Return decisions with clear constraint checks, trade-offs, and auditable outputs.
How it works
Qatalyst runs as a decision layer. It ingests operational inputs, builds a model, solves it, and publishes actions you can implement.
- Capture objectives, constraints, and policies (capacity, time windows, power limits, compatibility rules)
- Build a structured optimisation model from the captured rules
- Solve with the best-fit back-end (classical first, hybrid or quantum when beneficial)
- Return actionable outputs with constraint checks and traceability
What you provide
You do not need to provide equations. You provide operational inputs in the format you already use.
- Demand, jobs, arrivals, or bookings
- Resource and capacity limits (chargers, lanes, staff, bays, power caps)
- Time windows and deadlines
- Operational policies and priorities
- Cost and emissions preferences (optional)
Model types we support
Qatalyst supports the optimisation structures commonly used in operations. Where appropriate, models can be mapped into quadratic forms suitable for specialised solvers, including quantum-ready back-ends.
Assignment and allocation
Allocate jobs to resources under capacity and compatibility rules.
Scheduling
Sequence tasks with deadlines, setup times, and limited resources.
Routing and network flow
Route vehicles or goods through networks with time windows and congestion constraints.
Solver strategy
Qatalyst uses classical optimisation for most deployments, then escalates only when needed.
- Classical (default) for reliable day-to-day scheduling and planning
- Hybrid when the problem is large, tightly coupled, or must be solved repeatedly under disruption
- Quantum-ready when the model structure and runtime constraints justify specialised quadratic solvers
When quantum is actually useful
Quantum is not required for most problems. It becomes interesting in rare, high-stress scenarios.
- Many independent actors competing for limited shared capacity
- Highly coupled decisions where local optimisation creates system-level congestion
- Time-critical re-optimisation where classical methods become slow or unstable
- Quadratic penalty structures that map naturally to specialised solvers
What the system returns
Qatalyst outputs decisions your team can implement immediately, alongside the reasoning and constraint checks needed for trust.
Decision plan
Assignments, schedules, and recommendations for the next planning horizon.
Constraint checks
Clear validation showing which constraints bind and why the plan is feasible.
Trade-offs
What changed, what improved, and what was traded (time, cost, energy, emissions).
Ready to test on your operational data?
Share a small sample of demand, constraints, and site limits. We will map your decision problem and propose an approach, classical first, with hybrid or quantum options only where they add value.