D-Wave Collaborates with Volkswagen to Optimize Traffic Flow Using Quantum Annealing
April 14, 2016
Volkswagen and D-Wave Test Quantum Traffic Models for Urban Logistics
Urban logistics had become a growing challenge by 2016, with megacities like Beijing, London, and São Paulo facing mounting congestion from both passenger and commercial vehicles. Seeking to address this using next-generation computation, Volkswagen announced a strategic research project with D-Wave Systems on April 14, 2016. Their shared goal: to investigate how quantum annealing could be used to optimize traffic and fleet logistics in real time.
The partnership marked one of the first known cases of a major automaker testing quantum computing techniques outside of theoretical simulation—and the implications for city logistics, emissions reduction, and delivery scheduling were profound.
“We’re trying to rethink mobility at the quantum level,” said Dr. Martin Hofmann, Chief Information Officer of the Volkswagen Group at the time. “This is about predicting and avoiding traffic bottlenecks before they arise—not reacting after they’ve already cost time and fuel.”
The Quantum Angle: Why Annealing?
Unlike general-purpose quantum computers, which in 2016 remained largely experimental, D-Wave’s systems were based on quantum annealing, a process well-suited for solving complex optimization problems involving vast numbers of variables. City traffic flows, delivery vehicle routes, signal timing, and driver behavior—these all represent combinatorially explosive challenges.
D-Wave’s quantum annealers were specifically designed to tackle such problems using an Ising model representation, transforming logistical optimization into an energy minimization exercise.
Volkswagen’s team believed that by encoding routing variables into D-Wave’s system, they could identify optimal traffic configurations faster than traditional algorithms—especially during rush hours, when dynamic rerouting is critical.
Target Use Case: Urban Logistics and Fleet Delivery
Though the project gained attention for its potential in commuter mobility, Volkswagen and D-Wave emphasized the implications for urban freight movement and last-mile delivery. In congested cities, commercial fleets often face delivery windows, restricted zones, and fuel penalties—all of which must be considered in real-time routing.
Volkswagen’s quantum experiment modeled:
Delivery vehicle paths across neighborhoods
Warehouse dispatch timings
Fleet-wide traffic impact predictions
Alternating signal controls
Initial simulations focused on Beijing and Hamburg, two cities with vastly different logistics infrastructures but comparable levels of congestion.
“The objective isn’t just smoother rides for consumers—it’s about smarter routes for delivery vehicles, taxis, and ride-hailing fleets,” explained Dr. Florian Neukart, Volkswagen's lead data scientist on the project. “Quantum annealing lets us test hundreds of thousands of route permutations faster than classical methods.”
Methodology: From Classical to Quantum
To build the model, Volkswagen first collected large-scale traffic datasets from Hamburg’s municipal transport agency and fleet telematics in Beijing. These datasets included real-time congestion patterns, GPS logs, vehicle speeds, and time-stamped delivery events.
Next, the data was transformed into QUBO (Quadratic Unconstrained Binary Optimization) form—the native format required by D-Wave’s quantum annealing hardware.
Key logistics variables included:
Time windows for deliveries
Number of packages per route
Road capacity constraints
Emissions caps
Traffic light phasing
These were encoded into the QUBO model and submitted to a 1,000+ qubit D-Wave 2X system located in Burnaby, British Columbia.
Within seconds, the system returned route configurations designed to minimize delivery time, traffic interference, and fuel consumption—all while satisfying constraints.
Early Results: Promising Gains for Fleet Logistics
Volkswagen reported that in pilot simulations, the quantum annealing-based routing:
Reduced delivery fleet travel times by an average of 10–15%
Lowered estimated fuel usage by up to 9%
Significantly improved route diversity, minimizing traffic “clumping” effects
The simulations showed that even modest reductions in travel time per vehicle could result in dramatic cost and carbon savings when applied across a fleet of 1,000+ delivery vans.
For example, shaving five minutes per route across a fleet could reduce CO₂ emissions by tens of thousands of kilograms annually in a single city.
Integrating with Smart City Initiatives
Volkswagen’s quantum work came at a time when German cities and EU infrastructure planners were investing heavily in “smart mobility” initiatives. Hamburg, in particular, had launched ITS (Intelligent Transport Systems) pilot zones, outfitted with connected traffic lights, vehicle sensors, and predictive congestion alerts.
The quantum model was designed to plug into these systems, enabling real-time rerouting of delivery trucks and autonomous service vehicles. Combined with 5G and V2X (vehicle-to-everything) communication, the system promised a logistics network that could adjust routes proactively—before congestion occurred.
“This isn’t about replacing traffic control systems,” said Neukart. “It’s about augmenting them with quantum capabilities to see further ahead.”
Impact Beyond Mobility: Logistics Efficiency as a Competitive Edge
While the partnership was led by Volkswagen’s mobility and IT teams, the potential logistics applications were clear to industry observers. With retail giants like Amazon and Alibaba rapidly building urban fulfillment networks, real-time route optimization had become a competitive differentiator.
Quantum routing, if integrated into fleet management platforms, could:
Reduce labor hours by shrinking delivery windows
Improve SLA compliance by hitting exact delivery timeframes
Cut costs on fuel, maintenance, and idling penalties
Increase throughput during peak shopping seasons
D-Wave’s system, while still in its early stages, offered a hint of what logistics operations might look like in five to ten years: more predictive, adaptive, and energy-efficient.
Volkswagen’s Continued Quantum Investments
Following the April 2016 announcement, Volkswagen continued to invest in quantum computing R&D. It opened a dedicated Data:Lab in Munich and began working with other quantum startups, including Google Quantum AI and later Xanadu.
By 2019, Volkswagen had successfully demonstrated real-time quantum-based traffic flow optimization on nine buses in Lisbon, Portugal—an extension of this early Hamburg work.
While critics noted that quantum annealers could still be simulated on classical supercomputers in some cases, proponents argued that early adoption was about building expertise, not perfection.
D-Wave’s Strategy: Real-World Use Cases First
For D-Wave, the Volkswagen collaboration validated its unique go-to-market approach: targeting practical, optimization-heavy applications long before other quantum vendors had commercially available hardware.
By focusing on real business problems—fleet routing, supply chain scheduling, risk modeling—D-Wave carved out an early leadership position in applied quantum logistics.
Its “quantum as a service” model also helped demystify quantum access, allowing enterprise customers to experiment without investing in exotic hardware or quantum physicists.
Conclusion
The April 2016 collaboration between Volkswagen and D-Wave marked a critical milestone in the convergence of quantum computing and urban logistics. By targeting the complex problem of fleet optimization and real-time traffic management, the partnership moved quantum technologies from the lab into the logistics trenches.
With cities growing larger, fleets getting smarter, and emissions regulations tightening, the need for breakthrough optimization is only intensifying. Quantum annealing, once seen as a niche academic tool, is now at the center of a broader shift in how we model movement—of people, packages, and platforms.
As Volkswagen’s early work with D-Wave showed, the road to quantum-powered logistics may not be far off. It’s already being mapped.