Workshop Description
For fleet operations directors, logistics technology leads, and automotive R&D teams evaluating quantum computing for combinatorial optimisation. Covers QUBO formulations for vehicle routing, quantum algorithms for fleet scheduling, EV charging network planning, and an honest assessment of what current hardware can and cannot solve.
Vehicle routing with real-world constraints (time windows, capacity limits, driver hours, multi-depot assignment) is NP-hard. Classical solvers handle fleets of hundreds of vehicles using heuristics, but solution quality degrades as constraint complexity grows. Quantum approaches encode these problems as QUBO formulations and solve them via quantum annealing (D-Wave) or variational gate-based algorithms (QAOA). Published results from Volkswagen (traffic flow optimisation in Lisbon, 2019), BMW (production scheduling), and the Toyota/QC Ware collaboration show quantum approaches matching classical solvers for small fleet sizes (15 to 40 vehicles) with the expectation of advantage at larger scales on future hardware. EV charging network optimisation is a particularly promising domain: the combinatorial complexity of station placement, demand prediction, and grid integration maps naturally to quantum optimisation formulations. This workshop does not cover quantum-enhanced reinforcement learning for real-time autonomous driving decisions, which remains a speculative research concept with no demonstrated automotive application.
What participants cover
- Vehicle routing problem complexity: why fleet logistics with real-world constraints is NP-hard and where classical heuristics fall short
- QUBO formulations for VRP: encoding time windows, capacity limits, multi-depot assignment, and driver hours as quantum-native problem representations
- Quantum algorithms for fleet scheduling: QAOA, VQE, and quantum annealing applied to automotive logistics problems
- Hands-on comparison: solving a fleet routing problem on D-Wave annealer and IBM gate-based simulator, benchmarked against classical OR-Tools
- EV charging network planning: station placement, demand forecasting, V2G scheduling, and grid load balancing as combinatorial optimisation
- Hardware reality: fleet sizes solvable today (15-40 vehicles with constraints), quantum-inspired classical solvers for near-term deployment, and the fault-tolerant timeline