50,000 free calls a month, card-free. Get an API key →

Solutions / Trucking & Logistics

The router refuses the bridge before your driver ever sees it.

Height, weight, ADR and hazmat restrictions are enforced inside the route search itself, not bolted on afterwards. Ask for the rationale and get an honest answer back: which constraint was binding, and why.

Why MapMap for fleets

Routing that takes your vehicle seriously.

Dimensional and dangerous-goods restrictions shape the search itself, and everything downstream of the route, from dispatch to fuel, rides the same engine.

Diagram of a route line diverting around a low bridge arch, then around a tunnel cross-section marked with a restricted clearance zone
Enforced, not penalised

Dimensional and ADR restrictions built in

Height, weight, width, axle load and hazmat class are excluded from the search, not merely penalised: the router does not find the low bridge and then avoid it, it never sees it as a candidate. ADR tunnel categories A to E are enforced the same way, and Directive 96/53/EC defaults fill in any dimension you do not supply, erring towards over-restriction rather than a lucky guess.

A real truck route near Manchester with a marker where a declared 4.9 m height constraint was binding, and a card explaining the rationale
Honest answers

Ask why, get a rationale

Set rationale to true and the response names which constraint was binding on that route: a height limit here, an ADR tunnel category there. It is a routing engine, not a compliance certificate, so check local signage on the day, but it will not put a 44-tonne artic under a 3.9 m arch.

Illustration of safety camera and weather report cards pinned along a route line
Along the route

Fuel, cameras and weather along the route

Cheapest fuel along the route is ranked by real engine-computed detour cost through the MCP server, from statutory open-data price feeds with a staleness flag. Safety cameras, jurisdiction-gated, and weather along the route ride the same along-route search.

A real optimised five-stop delivery round across London on the MapMap map, with a card listing the solver's visit order and cumulative times
Dispatch

Multi-vehicle optimisation

One call plans multiple vehicles against time windows and capacities, backed by the matrix workhorse. This is a real optimised round from that call: five London stops reordered by the solver, 18.6 km, 37 minutes, back at the depot. EU drivers' hours are a stated single-shift approximation of EC 561/2006: a planning aid, not a tachograph replacement.

Optimisation docs
the whole stack, inside your own network
$ docker compose up -d --build
→ tiles · geocoder · router · guidance · territories
→ 0 external map dependencies
Self-hosted

Your fleet data never leaves your network

Docker distribution, signed territory packages, and a procurement pack with an SBOM and security questionnaire ready to go. No per-vehicle metering to a third party, no telemetry leaving your network unless you send it.

Ink illustration of a lorry cab with a chat bubble, its route threading between a low bridge and a tunnel
Agent-first

Built for the TMS your engineers are writing

A hosted MCP server, one-step installs for Claude, Cursor and VS Code, and a card-free API key in one call. Your engineers can be routing trucks against real ADR data inside an hour, not a sales cycle.

Known limit: ETAs come from the routing engine's speed model, not a live-traffic feed. Ask MapMap what the road network supports, not what today's congestion is doing to it, and pair the router's rationale with your own knowledge of local restrictions before it goes near a live load.

Agent-first

We will build your dispatch proof of concept, fast.

The low-bridge capture app in our showcase, four thousand seeded UK bridges with a truck-versus-car route comparison, was built by autonomous AI agents in under a day on this platform. Give us your vehicle dimensions, depots and a handful of real deliveries, and we will build a working proof of concept of your routing and dispatch problem the same way: fast, on real data, running on MapMap.