Automatic dispatching and scheduling of a fleet via REST API
(passengers, freight, service orders)
Fleet tour-planning, tour-optimization or sometimes also called fleet orchestration solves the task, e.g. 10 heterogeneous vehicles (3.5t, 40t etc. in a fleet) with 100 transportation orders; which transportation orders should be delivered by which vehicles, in which order, taking into account capacity constraints, such as volume, payload, service times as well as driving and rest periods of the drivers. The product is suitable for delivering previously defined freight or service orders in a cost-minimized manner for the transport company (if capacities are too low, transportation orders are also rejected, i.e. they must then be scheduled for another time). Weights and thus loaded- and empty-kilometers are taken into account, as well as heterogeneous vehicles in terms of costs and their physical properties (e.g. consumption but also loading volume/payload). It is suitable for cargo, passenger and service tasks on trucks, vans and cars.
Many to many with random pick up and delivery stops. Depot to customer, customer to depot, customer to customer in one optimization run.
The AI-supported determination of the ETA and thus the calculation of the driving time gives you exact predictions about the arrival times and thus allows you to plan much more realistically. The system uses the value-based routing engine to evaluate the transportation orders, but also to resolve them into routes. The results can thus be exported as turn-by-turn navigation to a carrier app or navigation device (e.g. our own navigation app). The optimized solutions can be imported into your existing system, e.g. into the plan board of the TMS via REST-API, and the rest of the system can continue to be used as before. (Dynamic rescheduling and overnight optimization are in development and coming soon.)
(Suitable also for individual vehicles as tour planning for a pure sequence problem)
Builds on the Value-based Routing Engine. For more information please click on the link.
planning time reduction
Support your dispatcher and have more time for other tasks. An average time saving in fleet tour-planning is between 90-95%.
Let the computers do the tasks they are good at, namely calulation. Plan in minutes instead of hours.
With computer-optimized tour-planning, cost reductions of between 15-35% can be expected. Save money. In most cases, the investment pays for itself in the first month. They use the Value-based Routing Engine to evaluate the tours and stops. In addition to more reliable tours, they also have better timed stops, which means that more stops can be completed per tour on average.
Convince yourself and test the system with our DEMO frontend.
■ Daily scheduling/fleet route planning/route optimization with detailed turn-by-turn routing
■ Can be used for freight orders, service orders and passengers
■ Round trips (same start and end stop), (multiple) pickup and delivery (pick up ‘n delivery)
■ optimized loading order
■ Time window (e.g. opening times, ramp times), payload, free loading area/seats, etc
■ (LKW) Germany map material, driver profiles and tolls
■ Working time restrictions with driving and rest times for drivers (e.g. a break of 40 minutes after 240minutes of total 480 minutes working time)
■ Different loading and unloading times or service times for each stop separately
■ Calculation of ETA (expected time of arrival) by our AI model
■ Cost calculation for each tour (heterogeneous vehicle fleets, e.g. cars, 3.5t, 40t)
■ Cost-optimized dispatching, not shortest/fastest routes with loaded and empty kilometers
■ Goods to be transported can be freely defined (payload, volume, parking spaces, seats, units, etc.)
■ calculates the optimum number of vehicles for daily operational planning
Optimization is a complex problem. Already for 1 vehicle with 10 stops there are over 3 million possible solutions that have to be calculated and compared, with 11 stops over 39 million and with 15 stops already over 1.3 trillion (1300 billion). Further constraints such as volume restrictions, payload limitation, minimizing costs, max. working time of drivers and above all random pick up n delivery with multiple depot trips etc. complicate the problem exponentially. And even in the simplest use case, in which all deliveries are driven out of one depot, we are already moving beyond the quadrillion to septillion possible solutions with a fleet of 2 vehicles with 25-31 stops. Only complex heuristics and algorithms can help here; in a test by a car manufacturer for robo-taxis, even a quantum computer needed about 20 minutes of computing time to plan “only” 2 vehicles and a total of “only” 7 stops.
Load- and empty-kilometers
With the introduction of weight-based planning, tours can suddenly turn around. Loaded and empty kilometers and their corresponding different costs can make a big difference.