AI-Powered Trip Optimisation
Automatically assign consignments to vehicles with a built-in VRP solver, and preview the result before committing.
Inside the product
Stop spending hours building route plans manually. Transportial's built-in Vehicle Routing Problem (VRP) solver analyses every open consignment, every available vehicle, driver capacity, time windows, and EU driving regulation constraints, then produces an optimised plan in seconds.
One search instead of a stack of tools
A typical planning stack answers one question with many tools. A route optimiser orders the stops, a driving-time check runs afterwards, ADR lives in a spreadsheet, a load planner packs the trailer, a dock screen books slots, an empties desk finds containers, and a carrier portal handles what is left. Each answer can be right on its own while the combined plan fails: at the dock, where three trucks arrive for two doors; at the terminal, where no empty container is waiting; or at the trailer door, where the frozen goods are in the chilled compartment.
The Transportial planner decides these together. While it builds a tour it chooses the vehicle, driver and trailer or carrier, checks each driver's clock, packs the compartments, times the stops around the docks, and sources the empties, so a change in one place is weighed against all the others. The plan it hands over is one a planner can save and dispatch.
The same engine plans in the optimiser, in map planning and on the API as POST /vrp/solve, so what developers test is what planners get.
Optimisation is a tool for planners, not a replacement
Transportial treats the optimiser's output as a proposal. It produces tours quickly against explicit rules, then hands control back: the planner sees every trip on the map, on the board and as a 3D load plan, moves what local knowledge says should move, and only then saves. That is why the output goes through the same trips panel as manual drag-and-drop planning.
The plan board guardrails run on the saved result as well. Driver-hour, capacity, ADR and conflict alerts do not care whether a human or an algorithm built the trip.
Built on published research
Two parts of the planner adapt methods from operations research. Docks follow A. Bombelli and S. Fazi (2022), "The ground handler dock capacitated pickup and delivery problem with time windows: A collaborative framework for air cargo operations", Transportation Research Part E 159, 102603. In the paper's experiments, uncoordinated plans queued for docks for up to about three hours and some became infeasible, while coordination cut travel time by 2 to 19%. Our adaptation keeps every plan the search holds dock-feasible: instead of the paper's penalised overbooking, a tour waits or the tour that can shift gives way.
Empty equipment follows S. Fazi, S.K. Choudhary and J.X. Dong (2023), "The multi-trip container drayage problem with synchronization for efficient empty containers re-usage", European Journal of Operational Research 310(1), 343–359. In the paper's real cases near Tilburg and 's-Hertogenbosch, its plans were 25 to 43% cheaper than the planners' own, with fewer trips. Our adaptation runs inside our large neighbourhood search instead of branch-and-price, checks stock only when a unit is taken (the paper's critical times), and prices depot fetches as detours.
Those results are the papers' own, not measurements of Transportial. We built on the published work, which implies no endorsement by its authors.
Why the engine is pluggable
No single optimiser fits every operation, and some platforms already license one. Transportial runs the engine as a platform setting, so a platform can plan on the Transportial planner or on PTV OptiFlow, xTour, OSRM or CloverOpt without changing how planners work. Compartments, docks, equipment stock, scheduled chains and the custom objective are decided by the Transportial planner; the other engines plan road tours with their own options.
The same applies to routing. The route calculator behind the optimiser can be PTV, HERE, Google, OSRM or the self-hosted GraphHopper, and routing work can be offloaded asynchronously to worker nodes over Kafka when many routes must be computed at once.
How it works
Select the scope
Open the Optimizer tab and pick the open consignments and the vehicles it may use; set operating intervals, max driving period and max period, and keep the platform objective or choose another one for this run.
Run
Press "Optimize!". The selected engine plans tours against capacity, compartments, docks, equipment stock and the other rules, and reports back with proposed trips and unplanned consignments.
Inspect
Each proposed trip shows its stops, distance, duration, cost, emissions and dock waits. Switch from Map to Load plans to see each trip packed in 3D. Unplaced consignments list the reason so you can relax a constraint or add capacity.
Adjust by hand
Drag a consignment to another trip, reorder stops on the map, or pin an assignment you do not want changed, then recalculate.
Save and dispatch
Save all trips. They appear on the boards with plan status and their load plan, ready to confirm and send to drivers.
What it does
What one search decides
Most planning stacks split one decision over a route optimiser, a driving-time check, an ADR sheet, a load planner, a dock screen, an empties desk and a carrier portal. The Transportial planner decides these together, in one search, so the plan it hands over can run. The same engine plans in the optimiser, in map planning and on the API.
Every resource per tour
Each tour gets a vehicle, a driver and, when it needs one, a trailer, or a carrier's capacity contract instead. Every driver has one EU 561/2006 driving-time clock across all their tours, starting from what they already drove. Trailers are picked from a pool by type, attributes, temperature and room, and fetched from where they stand.
Carriers next to your own fleet
Capacity contracts are priced by your own pricing modules and limited by their lanes, service areas, operating hours and order cut-offs, so a subcontractor only gets work it can actually carry.
Every mode in one plan
Road, rail, inland waterway, sea and air in one plan, plus door-to-door chains over scheduled services: truck to the terminal, a train, barge, ship or aircraft, and truck to the door, with cut-offs and capacity per departure.
Compartments
Multi-temperature reefers, tank compartments and food-grade holds. A consignment only goes into a compartment that fits its temperature, dangerous goods and foodstuff status, and each compartment's capacity is checked over what is on board at the same moment.
Dangerous goods on the load
ADR 1.1.3.6 points are added up per tour, incompatible classes travel apart, and foodstuffs never share a compartment with toxic or infectious classes (6.1 and 6.2 by default, after ADR 7.5.4).
Docks, ramps and berths
A site plans no more vehicles at once than its docks can handle, counting other trips already there and booked dock timeslots. A vehicle waits for a free dock as real waiting time, or a tour with slack gives way.
Empty containers and pool equipment
An export needs an empty container at pickup, and an import frees one after delivery. The planner reuses a freed unit on the same tour (a street-turn), or takes one from stock or a depot, priced as a detour. Stock is tracked over time across all tours and never planned below zero.
Your definition of "best"
Weigh distance, driving time, tour time, waiting, money, CO₂e, tours, resources, lateness and unplanned consignments, and rank them in priority levels: first plan everything, then fewest tours, then cost. The result shows the objective per component, per trip and in total.
The load, visible before you save
A 3D load plan for every proposal
Every proposed trip gets a 3D load plan you can open before saving: the optimiser has a Map | Load plans switch, and map planning has a Load plan button per proposed trip. It shows the loading order stop by stop, whatever engine planned the trip.
Kept with the trip
Saving the trip keeps its load plan, so the yard sees the same picture the planner approved. See 3D load planning.
You choose how strict packing is
The Load planning setting is Off, Verify or Enforce. Off shows the plans without letting them change trips, Verify moves goods that do not fit, and Enforce makes packing a hard rule during the search. It needs the 3D load planning module.
Choose the engine that fits the job
The engine is a platform setting. Everything above is the Transportial planner; the other engines plan road tours with their own options, and their trips still get load plans.
Transportial planner
A large neighbourhood search: it starts from a nearest-neighbour plan, then repeatedly takes parts of it apart and rebuilds them, retrying consignments it could not place, within a time budget. Every insertion passes the same checks: compatibility, capacity per segment, 3D packing when enforced, time windows, docks and tour resources. An atomic search with relocate, swap, merge, split and 2-opt moves is available too.
PTV OptiFlow
PTV Developer's route optimisation with configurable routing profiles, traffic mode, cost per hour, kilometre, stop, route and unused vehicle, outsourcing costs, compactness, duration balancing, planning horizon, driver break rules and ADR or high-value vehicle attributes.
PTV xTour
Tour planning on PTV xServer for platforms that still run it (xServer 1 reaches end of life on 1 September 2026; OptiFlow is the successor).
OSRM and CloverOpt
OSRM solves single-vehicle trip ordering (TSP); CloverOpt is a cloud VRP service for platforms that prefer it. With CloverOpt CLAIRE, routes and loads are planned together and its load plans arrive on the trips the same way.
What the optimiser respects
Rules you set the strictness of
In the Transportial planner every rule can block, warn or be off. A warned rule costs something in the plan and is listed on the tour, and every consignment left out gets a reason with a count of how many resources each rule refused.
Vehicle operating intervals
Default operating windows per vehicle, a maximum driving period and a maximum total period per route.
Route shape limits
Maximum stops per route and maximum colli per route, plus toggles to ignore location, consignment or action constraints when you want a rough first cut.
Optimisation objective
The Transportial planner optimises distance, time, CO₂, cost, fewest assets or a custom weighted objective, set per platform and overridable per run. PTV, OptiFlow, OSRM and CloverOpt expose distance, time, CO₂ and cost.
Constraints from the data
Time windows, weight, size, speed, sensor value, fuel type, vehicle type, emission standard, equipment type and route constraints on consignments and locations, each enforced or preferred.
Preview, adjust, commit
Preview before saving
The optimiser returns proposed trips and a list of consignments it could not place. Each trip shows its dock waits, where its empties come from and go to, and its share of the objective. Nothing is saved until you press "Save all trips".
Edit the result
Open any generated trip on the map, add or reorder stops, reverse the route and recalculate; the plan board shows the effect on ETAs and plan status.
Driver-hour aware ETAs
Optional EU 561/2006 planning inserts breaks and daily rests into the route and into the ETA chain, so promised times already contain the driver's rest.
Recommendations for one consignment
For a single consignment, the recommendation engine scores vehicles, drivers, equipment, capacity contracts and existing trips with weighted rules (proximity, driving time, ADR, licence, language, schedule conflicts, headroom) and explains each score.
Technical details. The optimiser engine is a platform setting. Routing and optimisation work is offloaded asynchronously to worker nodes over Apache Kafka. PTV OptiFlow, PTV xTour, OSRM and CloverOpt expose distance, time, CO₂ and cost objectives; the Transportial planner optimises distance, time, CO₂, cost, fewest assets or a custom weighted objective.
Try it free first
Works with
Frequently asked questions
Which optimisation engines are available?
The Transportial planner, PTV OptiFlow, PTV xTour (xServer 1, retiring September 2026), OSRM and CloverOpt. The engine is a platform setting. Compartments, docks, equipment stock, scheduled chains and the custom objective belong to the Transportial planner.
Does the optimiser respect driving-time rules?
The Transportial planner gives each driver one EU 561/2006 clock across all their tours, starting from what they already drove, and only plans tours a driver can do within it. PTV OptiFlow applies driver break rules directly. For any engine, the optional EU 561/2006 planner inserts breaks and daily rests into routes and ETA chains, and the built-in alerts flag insufficient remaining drive time on the saved trip.
Can I keep some assignments fixed?
Yes. Pinned assignments and constraints marked as enforced are kept; the optimiser works around them. You can also ignore location, consignment or action constraints for a rough first draft.
What happens with consignments the optimiser cannot place?
They are listed as unplanned with the reason, so you can add capacity, relax a window or plan them by hand on the board.
What does the Transportial planner optimise?
What you tell it to. Choose distance, time, CO₂, cost or fewest assets, or a custom objective that weighs distance, driving time, tour time, waiting, money, CO₂e, tours, resources, lateness and unplanned consignments. The named objectives are presets of those same weights. PTV, OptiFlow, OSRM and CloverOpt expose distance, time, CO₂ and cost.
Does it work for a single new order arriving mid-day?
Use the recommendations on the consignment: it scores existing trips by time overlap, headroom in weight, volume and load meters, pickup proximity and ADR compatibility, and lets you apply the best combination in one click.
Can I see how a proposed trip is loaded before I save it?
Yes, with the 3D load planning module. Every proposed trip gets a load plan: switch the optimiser from Map to Load plans, or press Load plan on a proposed trip in map planning. It shows the loading order stop by stop, whatever engine planned the trip, and stays with the trip when you save.
Does the optimiser know how many docks a site has?
Yes. A site's capacity is the sum of its operational loading docks, and stops of other trips and requested or confirmed dock timeslots count against it. When all docks are taken, a vehicle waits as real waiting time, or a tour with slack gives way. Separate loading and unloading docks are not modelled yet: all operational docks of a site count together.
Can it plan empty containers?
Yes. Set the equipment stock in a location's site details; leave the units empty and the planner counts the operational, not-full equipment of that type last seen on the site. A consignment's equipment flow, a unit it needs at pickup or frees after delivery, comes in through the consignment API. The planner reuses a freed unit on the same tour where it can, otherwise takes one from stock or a depot as a priced detour, and never plans a stock below zero. Types match exactly, and a stock visit shows in the trip's times and distance rather than as its own stop.
Can I decide what "best" means for my operation?
Yes. Choose Custom in the optimiser settings and give each component a weight, such as per km, per hour of driving, per tour or per consignment left out. Priority levels rank them, for example first plan everything, then fewest tours, then cost. You can override the objective per run, and the result shows what each component contributed per trip and in total.
Concepts & terminology
- Vehicle routing problem (VRP)Glossary
- Assigning a set of consignments to a fleet of vehicles and ordering the stops so total distance, time, cost or CO₂ is minimised under capacity and time-window constraints.
- Objective
- What the engine minimises. The Transportial planner offers distance, time, CO₂, cost, fewest assets and a custom objective; other engines offer distance, time, CO₂ and cost.
- 2-opt
- A route-polishing move that reverses a segment of a route when doing so shortens it; part of the planner's atomic search.
- Operating interval
- The window in which a vehicle may work during an optimisation run, with a maximum driving period and total period.
- Constraint
- A rule on a consignment, location or trip (time window, weight, size, vehicle type, emission standard, equipment type, sensor value, route) marked as enforced or preferred.
- Recommendation
- A scored, explained candidate (vehicle, driver, equipment, capacity contract or existing trip) for one consignment.
- EU 561/2006Glossary
- The driving-time regulation whose breaks and rests can be inserted into routes and ETAs during optimisation.
- Street-turn
- Reusing the container an import frees for an export on the same tour, instead of returning it empty to a depot and fetching another one.
- Dock capacity
- How many vehicles a site can load or unload at once: the sum of its operational docks, ramps, berths or sidings. Other trips and booked dock timeslots count against it.
- Compartment
- A part of a load space, counted from the front wall and separated by a bulkhead, with its own length, capacity, temperature and rules for dangerous goods and foodstuffs.
- Custom objective
- Your own weights on distance, driving time, tour time, waiting, money, CO₂e, tours, resources, lateness and unplanned consignments. The named objectives are presets of the same weights.
- Priority level
- A rank in the custom objective. A higher level always outweighs everything below it, for example fewest unplanned consignments before fewest tours before cost.
Related features
Interactive Plan BoardA drag-and-drop Gantt planning board that gives planners complete visibility over every resource.
Fleet & Resource ManagementFull lifecycle management for every vehicle, driver, and piece of equipment in your operation.- Multi-Modal RoutingRoute every shipment by road, sea, air, or rail, with a truck-grade routing engine built into the platform.