Optimization API: Optimization algorithms account for fuel costs and energy consumption

We just released an improved version of our optimization algorithms. They can now consider fuel costs, i.e. load dependent costs, also. We assume that the higher the capacity utilisation the more energy the vehicles consume. Internally, it is modeled as a markup to the distance dependent costs. Thus, there is now a slight preference to unload heavy items not later than necessary and to avoid unnecessary driving with a full truck load. This might impact the orientation of the route.

NOTE that this only makes a difference if one of your size dimensions is greater than zero. Additionally, it is important to specify explicitly whether your service is a pickup or a delivery. By default, it is a general service which if it has a size dimension > 0 behaves like a pickup. You can specify it by setting the type of your service like this:

     "id": "service-id",
     "type": "pickup",
     "name": "meaningful-name", 
     "address": {
       "location_id": "service-location-id",
       "lon": 9.999,
       "lat": 53.552
     "duration": 3600,
     "size": [ 1 ], 
     "time_windows": [ 
            "earliest": 0,
            "latest": 3600
            "earliest": 7200,
            "latest": 10800
     "required_skills": ["drilling-machine", "water-level"],
     "allowed_vehicles": ["technician_peter","technician_stefan"]

You can also find the documentation of how to set the type of service here.

1 Like