AI route planning: Three options for a new order
Three vehicles are available for a new transport order. Option A has the shortest route. Option B avoids an empty run later. Option C keeps a critical time slot free for another order. In precisely these situations, AI route planning should not simply select the nearest vehicle, but rather evaluate several impacts on the daily schedule simultaneously.
The more vehicles, orders, and restrictions are considered, the greater the number of plausible assignments. Factors taken into account in dispatching include current position, vehicle category, capacity, driver qualifications, time window, priority, travel time, and already scheduled orders.
Three plausible assignments with different consequences
A short journey can be useful, but it might not be the most economical option for the entire day. The vehicle might not have a suitable follow-up job later. Another vehicle could initially drive a few kilometers further, thus preventing a subsequent empty trip.
An optimization logic therefore does not only evaluate the current order in isolation. It also considers the planned subsequent trips and checks how a new assignment affects the rest of the day's schedule.
What data does AI-powered route planning require?
Good proposals depend directly on the available data. The following are particularly relevant for operational planning:
- current vehicle position
- Vehicle type and capacity
- available drivers and qualifications
- Pick-up and delivery locations
- Time windows and priorities
- open and already planned orders
- Industry-specific resources such as skips or materials
A custom-developed scheduling solution can combine this information in such a way that the optimization logic works with the actual rules of the operation.
When a delay changes the best option
The assessment is not static throughout the day. If a vehicle is delayed, the ETA and potential follow-up orders are affected. A previously sensible allocation can therefore become less attractive.
The documented use cases provide for precisely this dynamic replanning: New ETAs, affected follow-up orders and alternative vehicles are considered together before a new proposal is created.
Open deliveries, pickups and return trips can also be analyzed together to reduce unnecessary kilometers.
AI as a suggestion logic for scheduling
The dispatcher can see which option is selected and what consequences it has for the daily schedule. AI or optimization logic thus primarily takes over the time-consuming evaluation of numerous combinations and provides transparent options for operational decision-making.
Whether the deployment is worthwhile can then be assessed using key figures such as planning effort, empty run rate, punctuality or vehicle utilization.
Jointly examine a real-life planning scenario
Together, in a 30-minute demo, we will see how Comitas combines vehicle positions, capacities, time windows and open orders, and where AI or optimization logic can meaningfully support your dispatching.
