Hi all,
I am using the VRP tool in order to create new routes for newspapers trucks.
In order to allow the software to choose the best trucks mixture, I created a large "bank" of vehicles (I have a few options).
My problem is with the Start depot. Since it is a printing house, I can start upload the papers at 01:30, but then only a small part of the papers are ready. It takes another 3 hours to finish the total amount. Since all the work must be finished before 06:30, what they are doing now, is to upload the trucks for the remote places first, and then slowly converge to the closer distribution points.
Does anybody have an idea how can I limit the depot capacity to supply?
Regards,
Asaf.
Hi Asaf,
Very interesting problem! This isn't one I have tried to solve before but here is an idea for you. It might take some trial and error to get the timing and truck distribution to work well for you so definitely experiment a little.
If the depot starts sorting the papers at 1:30 and then continues until it is done at 3:30 you could try putting several depots at the exact same location with different starting times and split the capacities up among the depots based on the loading rate. So for example if you had one depot start at 1:30, one at 2:00, one at 2:30 and so on you could then divide the number of newspapers by the number of depots you create. The trucks would then need to be divided out among the different depots also but it sounds like you have a lot of flexibility in your trucks. Maybe think about how many of each truck type could be filled in that half hour (or whatever time frame you use) and divide the trucks so that for each depot they have the max number of each type of truck if only that one type was decided to be filled during that time.
Let me know how you end up solving the problem or if you need any other ideas.
Thanks,
Heather
Hi Heather,
Thank you very much for responding.
I have thought more or less, about the same work around. As you mentioned, our main concern is about how would it damage the optimization..
We introduced the problem to the end customer, and hopefully it will be good enough for him as well.
Regards,
Asaf.