AI · Innovations

Intelligence behind mobility networks

Blixar’s AI platform designs routes, builds timetables and driver rosters, tracks every vehicle and connects city sensors, so transit agencies and fleet operators move more people with fewer resources.

Book a demo See the platform
  • Networks of 50 to 1,000+ buses
  • Diesel, hybrid and battery-electric
  • Cloud or on-premise

Portland, Oregon71 bus lines, 5 MAX lines · route data TriMet

Products and capabilities

AI that plans the network, runs the fleet and tracks every bus.

All capabilities
NPProduct 01 · Plan

AI Network & Route Planning

Atlas

Enhance the route network, widen access to public transport and connect the first and last mile, with the riders, kilometres and buses every change costs and saves.

Explore AI Network & Route Planning See it on a map
  • Route network enhancerMerge, shorten, split or divert routes; trunk-and-feeder and express overlays proposed with a cost line each.
  • Public transport accessibilityWalking catchments by district, and the people and homes left outside a 500 m walk of any service.
  • First & last mile connectivityFeeder links and on-demand transit scored where riders have no other way to reach the network.
  • Corridor & overlap analysisWhere routes run on top of each other, measured in the direction that decides a merge.
  • Stop optimisationWhich stops earn their place, with a kept stop always within walking distance of any removed.
  • AI planning assistantAsk the network questions in plain language and see the answer drawn on the map.
SRProduct 02 · Schedule

Scheduling & Rostering

Schedura

The fewest buses, chargers and drivers that can run a timetable, with a lower bound beside every answer.

  • Vehicle blocking & fleet sizing
  • Electric charging planner
  • Crew scheduling
  • Rostering & daily operations
Explore
FTProduct 03 · Operate

Fleet Telematics

GPStracs

Live location, driver behaviour and vehicle health for every bus, from 50+ tracker brands.

  • Live fleet map
  • Driver behaviour
  • Energy & vehicle health
  • Depot & geofence events
Explore

Built on

How the engine works

Generative & language AI

Generative AILarge language modelsAI agentsRetrieval-augmented generationNatural-language querying

Machine learning & perception

Machine learningPredictive analyticsRidership analyticsComputer visionEdge AI

Optimisation & simulation

Mathematical optimisationOperations researchGeospatial AIDigital twinsSystem Dynamics simulationExplainable AI

AI Network & Route Planning

See every route, and where they run on top of each other.

Routes that share a corridor carry the same riders twice and run buses nobody needs. Blixar finds the overlap, measures it in the direction that matters, and proposes the leaner network with its cost printed beside it.

Explore the product
Network map: routes 12, 31 and M4 share a corridor through the city centre; routes 7 and 22 cross it. SHARED CORRIDOR3 ROUTES · 3.2 KM1231M4722
  • Stop proposed for removal

Measured, not promised

What route and bus optimisation delivers.

Results from a real bus corridor: six local routes, one industrial area, three months of stop ridership. The biggest savings come from setting a kilometre target; every saving is shown with what it costs riders.

See the worked example

Kilometres saved

−39%

route-kilometres, with a kilometre target

Before163.1 km
After99.9 km

63 km of route taken out of one corridor by shortening four routes, with none withdrawn.

Without a kilometre target: 163.1 → 148.0 km, 9% fewer.

Buses saved

−32%

peak buses on one corridor

Before13 buses
After8.8 buses

The same corridor run with about four fewer buses at the peak.

Without a target: 13 → 11 buses, 15% fewer.

Passenger demand met

99.5%

of daily riders keep their stop (no target)

Before181 stops
After124 stops

Fewer, better-placed stops while almost every rider keeps their bus.

With the kilometre target, 58 riders a day (18%) walk further to a kept stop; the engine reports this itself.

First & last mile

≤ 600 m

walk to a kept stop

BeforeEvery rider
Afterwithin 600 m

A stop is only removed if another stop stays within walking distance.

Where the walk is too far, feeder or on-demand links are scored instead.

And in scheduling

124 / 124 buses against the timetable’s lower bound on a 1,545-trip depot

−6% driver duties for the same driving (1,601 → 1,501)

Example engagements by fleet and region

Electric bus transition

Size the electric fleet before you buy it.

Blixar models the energy of every trip, the state of charge of every bus and the posts at every site, and schedules charging inside the timetable, not after it.

  • Depot and terminal chargers counted and queued separately
  • Charging placed inside layovers
  • The evening peak your grid connection must carry
Plan the switch
Charging load: posts in use by site every quarter hour and energy going into batteries each hour

Fields of Consulting

Advice for everyone who plans, runs, funds or hosts public transport.

How we consult
Transport authoritiesNetwork design, tenders and reportingTransport agenciesStandards, data and multi-operator oversightEnvironment authoritiesEmissions baselines and zero-emission plansLand and network authoritiesDepots, terminals, corridors and charging sitesCities and municipalitiesMobility plans, coverage and city servicesTransport operatorsFewer buses and duties, fairer rostersEnergy and utility providersDepot grid connections and charging loadHow an engagement worksData audit, baseline, options, decision

What would it cost to run your timetable electric?

Send us a GTFS feed or a timetable export. We will run the first scenario with you: buses, chargers and drivers, with the lower bound beside every answer.

Book a demo Explore the platform