Canadian National Railway and Xanadu Launch Quantum Freight Scheduling Research in Toronto
June 6, 2022
Quantum Tracks: Applying Quantum Algorithms to Rail Logistics
On June 6, 2022, Canadian National Railway (CN) and Toronto-based quantum computing firm Xanadu officially launched a groundbreaking collaboration to explore how quantum optimization could reshape freight scheduling and capacity planning in the rail industry. The announcement, made in Toronto, marks a historic milestone as Canada’s first railway-sector quantum logistics pilot. It not only underscores the growing momentum of the country’s quantum ecosystem but also signals the beginning of a transformative journey in North American supply chain orchestration.
CN, one of North America’s most extensive rail operators, manages a network of over 30,000 kilometers spanning Canada and parts of the United States. This immense logistical footprint requires coordination of thousands of freight cars, hundreds of locomotives, and multiple intermodal terminals, all under tight scheduling constraints. Traditionally, these challenges have been addressed with classical optimization methods. However, as the supply chain faces increasing pressures from congestion, environmental mandates, and evolving customer expectations, CN is turning to quantum computing as a next-generation solution.
Xanadu, founded in 2016, has emerged as a global leader in photonic quantum computing. Its open-source software platform, PennyLane, is already widely adopted by researchers and developers building hybrid quantum-classical applications. By partnering with CN, Xanadu is positioning itself as a pioneer in bringing quantum technologies directly into heavy industries like rail transportation.
Project Scope and Research Goals
The joint initiative between CN and Xanadu aims to tackle three of the most pressing problems in freight rail logistics:
Train departure scheduling – Determining optimal departure times and track allocation when yard and mainline availability are limited.
Intermodal cargo flow optimization – Coordinating container transfers between rail, trucking fleets, and port facilities to ensure smooth cargo handoffs.
Freight car allocation and sequencing – Assigning cars to specific trains while minimizing idle time and maximizing energy efficiency.
These tasks are highly combinatorial in nature, meaning that as variables scale, the number of possible solutions grows exponentially. For instance, even a mid-sized scheduling problem can involve billions of possible configurations, making them extremely challenging for classical solvers.
The CN–Xanadu team is exploring the use of Quantum Approximate Optimization Algorithm (QAOA) and Variational Quantum Eigensolvers (VQE) to tackle these problems. While today’s quantum hardware is not yet at large-scale fault tolerance, hybrid approaches allow partial solutions to be generated by quantum processors and refined with classical computing.
Key Partners and Technical Contributions
The project is structured as a multi-stakeholder collaboration involving industry, academia, and quantum technology providers:
Canadian National Railway (CN): Provides anonymized historical and real-time freight data, operational constraints, and logistics expertise.
Xanadu: Supplies quantum hardware through its photonic quantum processors, as well as the PennyLane software platform for hybrid algorithm design.
University of Toronto: Contributes academic expertise, validation frameworks, and graduate-level research talent to strengthen algorithmic testing.
By combining operational knowledge with quantum expertise, the partnership aims to produce solutions that are not only scientifically innovative but also directly applicable to real-world rail logistics.
Why Rail Logistics Needs Quantum Help
Rail freight scheduling is notoriously complex. Operators must coordinate track time, locomotive power, train crew availability, and intermodal interfaces. Disruptions—such as weather delays, maintenance outages, or unexpected surges in cargo—further complicate planning.
Challenges include:
Freight car assignment: Efficiently allocating thousands of cars without creating empty hauls.
Yard management: Avoiding bottlenecks where trains must be broken down and reassembled.
Network synchronization: Ensuring arrival and departure times align across multiple hubs.
Traditional heuristics and brute-force solvers often fail at large scale or take too long to adapt when disruptions occur. Quantum algorithms, however, can explore solution spaces more effectively by leveraging quantum mechanical properties like superposition. This could allow rail operators to find better schedules, faster, and more reliably than before.
Early Experiments and Benchmarks
As part of their initial research, CN and Xanadu modeled a simplified corridor scheduling problem with constraints such as:
Limited siding capacity
Overlapping train paths
Car-type-specific dwell time targets
By applying QAOA through PennyLane and running experiments on Xanadu’s X-Series photonic quantum processors, the team achieved:
10% reduction in average train wait times compared to classical heuristics
Improved yard utilization during simulated peak congestion
Faster rescheduling when disruptions like delays or equipment faults were introduced
While these results are early-stage and based on scaled-down models, they provide meaningful evidence that quantum optimization can deliver advantages over legacy methods.
Strategic Significance for Canada’s Quantum Ecosystem
The CN–Xanadu partnership is not occurring in isolation. Canada has made substantial investments in quantum research, highlighted by the federal National Quantum Strategy launched in early 2022. Initiatives such as the Quantum Valley Ideas Lab in Waterloo and partnerships across universities have positioned Canada as a global hub for quantum development.
By bringing quantum into the rail sector, CN is not only solving pressing operational challenges but also anchoring industrial applications of quantum in Canada. This reinforces the country’s ambition to lead in both scientific research and commercialization.
As CN’s Chief Innovation Officer, Michelle Novak, remarked during the announcement:
“Rail logistics is undergoing digital transformation. We believe quantum optimization will be one of the pillars of intelligent supply chain orchestration in the next decade.”
Xanadu’s Photonic Edge
Unlike many competitors pursuing superconducting or trapped-ion systems, Xanadu is developing photonic quantum computers, which use particles of light (photons) to perform quantum operations. These systems offer several advantages:
Operate at room temperature, unlike cryogenic systems
Modular scalability for large quantum networks
Natural integration with fiber-optic and cloud infrastructure
For this project, Xanadu deployed a 24-qumode version of its X-Series processor, allowing CN’s logistics problems to be modeled directly in variational circuits. The system’s API access enabled seamless integration into CN’s existing Python-based logistics workflows.
Broader Industry Implications
The significance of this research extends beyond Canada’s borders. Globally, railways are seen as ideal candidates for quantum optimization due to:
Fixed infrastructure networks, reducing variables compared to road freight
High-volume, high-density scheduling needs
Long planning horizons, compatible with hybrid batch optimization models
As more railways move toward precision scheduled railroading (PSR) and multimodal orchestration, the ability to adapt rapidly to disruptions will be crucial. Quantum tools can provide operators with decision-support systems that are both faster and more resilient.
Roadmap and Next Steps
The CN–Xanadu partnership has outlined a phased roadmap:
Phase 1 (Q2–Q4 2022): Model development and simulation with real CN datasets
Phase 2 (2023): Prototype integration with CN’s yard and scheduling tools
Phase 3 (2024+): Deployment testing on larger, next-generation quantum hardware
The partners are also considering publishing open benchmarks for rail optimization, inviting collaboration across the logistics and quantum communities. Such benchmarks would accelerate progress by creating standard problem sets for researchers and technology developers.
Conclusion
The June 2022 launch of CN and Xanadu’s quantum freight scheduling research represents a milestone not just for Canada but for the global logistics sector. For CN, it opens the possibility of a new era of intelligent, adaptive scheduling that improves efficiency, reduces delays, and strengthens resilience across its network. For Xanadu, it provides an industrial proving ground for its photonic quantum technology.
Most importantly, the initiative demonstrates how quantum computing can move beyond theory and research labs into real-world applications. As railways, ports, and logistics operators worldwide face mounting pressures, quantum optimization may soon become a defining tool of next-generation supply chains.
By combining Canada’s strengths in both logistics and quantum research, CN and Xanadu have positioned themselves at the forefront of a technological revolution—one that could reshape how freight moves across continents in the years to come.