City-wide coordination. The missing layer in urban mobility.
TAP is not a navigation app. It's a coordination layer — one that distributes demand across the network so every commuter moves faster, together. When traffic is coordinated, everyone wins.
Every commuter makes a rational decision in isolation. Thousands of rational decisions create city-wide congestion.
TAP changes the level of the problem. Instead of optimising one journey at a time, it coordinates movement across the entire network — distributing demand across routes and departure times so the city flows better for everyone.
Traffic jams are a collective problem with a collective solution. TAP jointly optimizes thousands of individual commutes — maximizing the utilization of existing infrastructure.
TAP passively observes commuter movements — where people go, when they travel, which routes they take, where congestion happens. No manual input required. The system builds a living picture of the city's demand from real behaviour.
TAP aggregates anonymised trip data city-wide and leverages reinforcement learning to solve a large scale variant of the Traffic Assignment Problem. Unlike static optimisation, RL operates under real-world conditions — partial adherence, shifting demand, noisy data — converging on better solutions as it gathers more real-world data.
TAP delivers data-driven nudges to commuters at the right moment — a suggested route or adjusted departure window, sent as a push notification before a trip they regularly make. Every recommendation satisfies a strict fairness constraint: no commuter receives a suggestion that leaves them worse off than without TAP. Solutions are pareto-optimal by design. Individual journeys improve. The network flows better.
System-optimal routing distributes demand across routes and departure times so the whole network flows faster
Fewer vehicles idling in congestion means measurably lower CO₂ — without changing infrastructure
TAP is a pure software layer. Capacity improvements come from coordination, not construction
A 30-minute flexible arrival window, spread intelligently across commuters, can eliminate peak congestion entirely — maximizing utilization of existing infrastructure
Changing commuter behaviour doesn't require grand interventions. TAP learns each commuter's regular patterns and delivers targeted, high-value suggestions — the moments where a small shift in route or timing produces an outsized improvement for the network. One well-timed nudge to the right person on the right road can cascade into city-wide congestion relief.
Road capacity is expensive, slow, and politically difficult. Most cities already have unused capacity hiding in plain sight.
TAP unlocks that capacity with software. No new roads. No mass surveillance. No behaviour change campaigns. Coordinated nudges that shift demand toward system-optimal flow.
Every commuter independently selects the fastest route. Demand concentrates on the same roads at the same times. The network overloads in predictable places. Cities respond with infrastructure investment, restrictions, or costly traffic management systems.
Commuters receive personalised, data-driven suggestions that improve their own journey while reducing congestion for everyone. Cities gain a software layer that coordinates individual movement into system-optimal flow.
This is grounded in decades of traffic theory. Wardrop's user equilibrium — where every driver picks the fastest route for themselves — is provably suboptimal for the network as a whole. The gap between user-optimal and system-optimal flow is known as the Price of Anarchy. TAP is built to close it.
Measurable traffic improvements without physical infrastructure investment. TAP is a pure software layer on top of the existing road network.
Aggregated, anonymised commuter data gives city planners ground truth on how people actually move — not how roads were designed to be used.
Time saved, emissions reduced, network utilisation — all tracked and reportable. TAP Ops provides a live dashboard for traffic analysts and decision-makers.
No surveillance. Trip data is anonymised at the edge, never sold, never used for advertising. GDPR-compliant from day one. Fully auditable.
We are now speaking with cities that want to test a software-first approach to congestion before investing in expensive infrastructure. The ideal pilot city has recurring congestion, flexible commuter patterns, and a willingness to measure outcomes openly.
If your city is ready to coordinate traffic instead of simply managing it, we should talk.
Explore a TAP pilot for your city →We are a research-driven team building the coordination layer cities have always needed but never had.
Founded by researchers in Traffic Engineering and Machine Learning, TAP is the product of years of work on the Traffic Assignment Problem — now finally deployable at city scale.