Route and tour optimisation
Free route and tour optimiser for FTL and LTL: set stops, calculate the optimal sequence and save kilometres, driving time, CO₂ and cost — including utilisation in loading metres. Real-time transparency, sustainability and AI combined. Pro with live traffic and telematics on request.
Route optimization in 30 seconds
The short film shows step by step how many individual stops become a short, drivable tour – and why every kilometre saved lowers time, CO₂ and cost at once.
Every tour starts with many stops – loading, unloading, return trip.
⚙ Settings & assumptions ▾
What's behind it: the Vehicle Routing Problem
The question of the cheapest order across all stops is a classic optimisation problem — the Vehicle Routing Problem (VRP). Dantzig and Ramser described it in 1959 as the “Truck Dispatching Problem” for fuel deliveries. It generalises the Travelling Salesman Problem and is NP-hard: the provable optimum can no longer be computed at real-world sizes, so one works with good approximation methods.
Our calculator first builds a starting route and then improves it with local search: 2-opt reverses crossed edges, Or-opt moves individual stops. Larger metaheuristics such as Tabu Search, Simulated Annealing or ALNS often reach within 0.5 to 1 percent of the optimum in research, even with thousands of stops. Optimisation runs on real road kilometres (OSRM), not straight-line distance.
The tool embeds several VRP variants: time windows (VRPTW), capacity by loading metres (CVRP), backhaul instead of empty running (Backhauling) as well as round trip or open tour (OVRP). The result is a very good, not provably optimal tour — the 5 to 30 percent savings cited by software vendors are an industry figure, not a promise; what matters is the specific lane.
Where AI helps — and where caution is due
AI complements these methods above all by learning from operational data: more accurate geocoding and address analysis, forecasting of service and travel times, more robust plans under traffic and demand swings, and a better allocation of vehicles and drivers.
Caution remains warranted: AI needs plenty of clean data, is compute-intensive and often a black box — it is hard to trace why a route is recommended. Over-reliance on the model or on historical data can mislead when conditions change. AI is decision support, not an autopilot.
Sources: Dantzig & Ramser 1959 · Toth & Vigo, SIAM · Wikipedia: Vehicle Routing Problem (CC BY-SA) · 0.5–1 %: Vidal et al. 2014 · AI points after an interview with An de Wispelaere (PTV Logistics)
Real-time transparency, AI and sustainability – thought through together
Every kilometre avoided saves at the same time time, CO₂ and money. The free calculator shows the principle; the full effect comes from three levers working together:
Real-time transparency
- Dynamic adjustmentReacts immediately to traffic, roadworks or weather to shorten driving times.
- Live trackingPrecise arrival times (ETAs) for customers and dispatch.
- Stock and status updatesShows vehicle and load status without gaps.
Artificial intelligence
- Pattern recognitionCalculates optimal sequences from historical and current data.
- Machine learningContinuously improves forecasts for traffic flow and energy demand.
- AutomationRelieves dispatchers through fully automatic route recalculation.
Sustainability
- Emission reductionLowers fuel consumption and CO₂ output through less empty running.
- Electric mobilityPlans charging stops and ranges of electric vehicles intelligently.
- Resource efficiencyMinimises wear and bundles deliveries optimally.
Places are geocoded via address search (Nominatim / OpenStreetMap); the distances are real road km from OpenStreetMap routing (OSRM), on which the sequence is optimised as well (no straight-line distance). Live traffic, telematics tracking, AI forecasts and EV charge planning are additional functions (on request).
FTL & LTL: two roads, one goal – full tours
Whether full or part load: optimisation, real time and AI all pay into the same goal – more payload per kilometre driven. The lever simply sits in different places. Switch between the modes above.
LTL Part load & groupage
Many small shipments share one vehicle. The gain comes from bundling, a sensible stop sequence and high utilisation.
- Consolidation of several orders on one groupage tour
- Optimal stop sequence including time windows
- Utilisation in loading metres instead of empty space
- Cost per shipment falls with every additional stop
FTL Full load & direct traffic
One order fills the vehicle – the lever sits between the trips: short repositioning and a return load instead of empty kilometres.
- Direct traffic without transhipment
- The sequence of several legs minimises repositioning
- A return load avoids empty running
- Assign the right lane and carrier to each load
Artificial intelligence & machine learning
The free calculator shows the sequence. Beyond that, planning and dispatch can be sharpened with methods from AI, machine learning and operations research. That is available as Pro and Premium services (on request) — explained here only, not part of the free calculator:
Predictive analytics
Volume and demand forecasts for fluctuating utilisation: internal and external data sources are used to estimate which capacities (fleet, slots, staff) will likely be needed.
ETA forecast per mode of transport
Forecast arrival times across road, rail and water — including alerts on disruptions and checks of alternative carriers or routes, so order and delivery timing fit better.
Automated dispatching
Dynamic real-time route planning: current fill levels and forecast demand feed in automatically, respecting working hours, prioritisation and vehicle capacity — especially for spontaneous, customer-driven transports.
EV range & driving-energy forecast
For electric vehicles, range matters: a model of topography, weather, weight and trip sequence estimates consumption, so tours are neither planned too short nor stretched too far at risk.
Image analysis for damage detection
Image recognition inspects load carriers and vehicles on photo or video footage and flags damage early at intake — less manual effort, fewer unnoticed follow-up costs.
Network flow analysis
Multi-commodity optimisation across the whole network: production sites, warehouses, transhipment points and special requirements (dangerous goods, refrigeration) are considered together to find cost-optimal quantities while meeting delivery times.
Multimodal network planning
Assemble transport chains across several modes: check goods flows for consolidation and suitable transhipment points — often more complexity, but lower costs and more resource-efficient transports.
Supply chain analytics
Data mining and business analytics on your own logistics structure: find bottlenecks, reveal potential for change — and equally point out where no optimisation is needed. A basis for strategic decisions.
These methods are available as Pro and Premium services on request; they are not included in the free calculator and are only explained here. Where figures would be named, they carry their method and provenance with them.
Optimisation with live traffic, telematics and AI
- Real road km and live traffic instead of a straight-line estimate
- Time windows, driving hours and vehicle restrictions
- AI forecast of driving time, demand and delay risk
- Fleet-wide optimisation and automatic re-dispatch
- Return-load matching (FTL) and consolidation (LTL)
- CO₂ and cost reporting per tour and customer
Open-source project. The free calculator stays free; the additional functions are provided exclusively on request and after approval – no automatic access.