Built for logistics operators and software providers who need high-quality solutions within tight planning windows. TQrouting combines AI and advanced mathematical optimization to solve large-scale vehicle routing problems.
Request a place in the Early Access program. Bring your hardest instance: your stops, your constraints, your fleet. We’ll support you in running it.
Places are limited and selected for fit. Built for teams planning hundreds to tens of thousands of stops under dense constraints, and for software providers embedding routing into their own platform.
When networks scale and constraints multiply, standard solvers degrade. Every percent of routing inefficiency adds cost: fuel, time, empty miles, maintenance, and emissions, eroding the service levels you've committed to.
TQrouting turns that scale into your advantage.
Built for where standard tools degrade: thousands to tens of thousands of stops, dense constraint sets, and tight planning windows.
Delivers best-in-class routes within seconds to minutes, enabling live re-planning as conditions change. Consistently outperforms commercial and open-source solvers under strict operational time limits and constraints.
Scales from thousands to tens of thousands of stops by intelligently decomposing large problems and solving them in parallel for speed without loss of quality.
Handles time windows, capacities, vehicle availabilities, EV energy limits, and more out of the box. Custom constraints plug in as modules, with no rebuild of the core engine.
For when your delivery plan keeps changing faster than your team can react, so you can re-optimize routes instantly as new orders, delays, or disruptions happen.
Deployed entirely within your own infrastructure. Your data and routing logic stay inside your environment at all times.
One optimization engine across very different operations. Its modular core adapts to each fleet's constraints without rebuilds, keeping routes efficient and cutting the driver time and fuel that weak planning wastes.
High-performance routing for thousands of daily stops, dense time windows and complex multi-depot parcel and postal networks.
Plans booked-slot grocery delivery around tight delivery windows, high stop density and perishable goods which leave no room for late routes.
Plans energy-feasible routes considering battery range, charging coordination and state-of-charge constraints across electric and mixed fleets.
Optimizes fleet scheduling across multi-depot retail and wholesale networks, supporting complex distribution needs and diverse vehicle types.
Optimizes picker routing and scheduling for high-throughput warehouses and distribution centers, improving efficiency across fulfillment operations.
Optimizes territory and visit scheduling around skills, time windows, break requirements and service-level commitments.
We're opening TQrouting to a small group of teams. Bring a real routing problem: your stops, your constraints, your fleet. Get one month complimentary access with direct support from the team who built it. Places are limited and we select for fit.
Request a place in the Early Access program. Bring your hardest instance: your stops, your constraints, your fleet. We’ll support you in running it.
Places are limited and selected for fit. Built for teams planning hundreds to tens of thousands of stops under dense constraints, and for software providers embedding routing into their own platform.
TQrouting is benchmarked against commercial and open-source solvers as well as state-of-the-art algorithms on the public CVRPLib benchmark, the standard reference for evaluating vehicle routing algorithms. Performance is strongest on large-scale and constraint-dense instances under operational time limits.
Leading solution quality on the largest CVRPLib BKS Challenge instances.
Set during the CVRPLib BKS Challenge and verified by the academic community.
Solved at scale on standard CPU infrastructure, inside operational time limits.
For large-scale problems, TQrouting intelligently breaks down complex instances into manageable subproblems solved in parallel, combining machine learning and advanced optimization for speed and quality. The engine is modular by design: new business rules activate as constraint modules without rebuilding the solver. The architecture is designed so that quantum solvers can augment classical optimization modules as quantum hardware matures: classical performance today, with a path to future acceleration.
“Vehicle routing is one of the most studied problems in combinatorial optimization. For over six decades, researchers from mathematics, computer science and operations research have produced algorithms that reach near-optimal solutions in acceptable time, even at large scale. But most are designed for one clean VRP variant. Real operations are messier: multiple objectives, and a rich set of soft and hard constraints that change from one customer to the next.
What is actually needed is a solver flexible enough to absorb all of it, modular enough to fit each customer's business logic, and robust enough to run in production, while still performing at the same level as state-of-the-art academic algorithms. We built TQrouting to close that gap: between what the academic VRP community has achieved and what a real operation requires.”
AI is reaching even the most conservative operational sectors. Fleets are automating planning, dispatch, and execution faster than ever, which creates a new problem: the optimization core has to keep up. Routing under diverse constraints, strict time limits, and tens of thousands of stops requires a specialized engine built for the problem, not a general-purpose model.
TQrouting is that engine.
If a few of these sound like your operation, TQrouting is built for you.
A strong fit if you have:
Its modular architecture supports, among others, time windows, capacities, multi-depot, EV energy feasibility, etc. Should you need custom constraint logic, please reach out to discuss edge cases specific to your operation.
It depends on how you work with us.
As an engine: you license TQrouting and integrate it into your own systems, mapping your fleet data, orders, and constraints into its structured input format, with onboarding support for enterprise deployments.
As a delivered solution: Terra Quantum configures and integrates a ready-to-run routing solution for your specific use case. Timeline depends on scope and your technical environment.
TQrouting is optimized for operational efficiency, including efficient use of hardware. It delivers competitive performance on standard CPU servers without requiring specialized hardware. Deployment is on-premises or in your private cloud.
Yes. Mixed fleets are handled natively in a single optimization run. EVs with range and charging constraints are processed alongside conventional vehicles, each governed by their own operational rules.
TQrouting's decomposition architecture is designed so that quantum solvers can replace or augment classical optimization modules as quantum hardware matures. Performance improvements from quantum computing can be integrated without rebuilding the system. Today, TQrouting delivers state-of-the-art performance on classical hardware.
Whether you're running a large-scale delivery network or building routing capabilities into your platform, TQrouting delivers the performance you need. Talk to our routing team to discuss your requirements.
Request a demo session with our routing team. We solve a large-scale instance and walk through what the engine produced: route quality, runtime, and how each constraint was handled.
For teams evaluating a routing engine to license and deploy in their own environment, whether you run the fleet yourself or build routing into a platform your customers use.
