Logistics
Real-Time Route OptimizationMaturity: in development
QUBO/annealing approaches for dynamic routing under traffic, capacity and timing constraints.
in development — Actively being built or tested. Results are provisional.
What this is
The problem
Vehicle routing is textbook-hard, and the textbook version is the easy one. Real routing carries time windows, capacity limits, driver hours, traffic that changes mid-route, and orders that arrive after the plan is fixed.
Where the current approach strains
Most operations re-plan on a schedule — overnight, or hourly — because re-planning is expensive. Between re-plans, the plan degrades. Dispatchers absorb the difference manually, and the quality of the day depends on who is on the desk.
What we are exploring
QUBO and annealing formulations for the re-optimisation step, aiming at the case where a good answer now beats an optimal answer in twenty minutes. Constraint handling matters more here than raw solution quality.
What would have to be true
A formulation carrying the real constraints, not a simplified variant. A comparison against the operator's incumbent solver, on their data, with their constraint set. And an honest accounting of solution quality against wall-clock time, since that is the actual trade being made.
Where it applies
Related
Demand Forecasting
Quantum-enhanced machine-learning concepts for complex demand patterns and supply planning.
Warehouse Management
Slotting, picking, labour, inventory and order-flow optimisation.
Last-Mile Delivery Enhancements
Dispatch, vehicle routing, delivery windows, energy and customer-priority optimisation.