-
|
In the PyVRP paper from 'INFORMS Journal on Computing', the following has been written:
Are the writers referring to Nagata et al.'s work from 2010 (https://doi.org/10.1016/j.orl.2009.04.006)? If yes, it is not clear to me how I can implement that in PyVRP from the paper or documentation. Any help in this regard would be appreciated. Also, I would love to contribute to PyVRP for multi-objective optimisation (i.e. Number of routes AND Total distance), but I was wondering if it is even possible or feasible to contribute this. |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 2 replies
-
Not on purpose. This idea has been (re)invented many times.
I'd run PyVRP with a given fleet size until you find a first feasible solution, or a time/iteration limit has been reached. If the solution is feasible, reduce the fleet size by one (or to the number of vehicles used in the feasible solution minus one, whichever is smaller) and try again. If the time/iteration limit has been reached, return the fleet size you know results in a feasible solution. |
Beta Was this translation helpful? Give feedback.
-
|
You might also appreciate #638. |
Beta Was this translation helpful? Give feedback.
-
|
@N-Wouda Any timelines for the release of 0.10 version, through which we can implement minimise_fleet? Is there any other way we can download that version ((pip install --upgrade pyvrp) installs 0.9)? |
Beta Was this translation helpful? Give feedback.
-
|
Your best bet would be to install from source for now (explained here). I expect we'll release v0.10 late next month, but that depends a bit on my currently fairly limited availability. |
Beta Was this translation helpful? Give feedback.
Not on purpose. This idea has been (re)invented many times.
I'd run PyVRP with a given fleet size until you find a first feasible solution, or a time/iteration limit has been reached. If the solution is feasible, reduce the fleet size by one (or to the number of vehicles used in the feasible solution minus one, whichever is smaller) and try again. If the time/iteration limit has been reached, return the fleet size you know results in a feasi…