Dmitri,
One unanswered question is to equalize the bin weights as much as possible
subject to tighest packing. Below is a solution. There's probably a more
elegant formulation.
In any event, you now have solutions to special cases of your more general
multiobjective problem.
-- minimize maximu
Jeff,
Wow, thanks a lot! That is exactly what I was looking for.
Best regards,
Dmitri
On 27/01/2013, at 14:17, Jeffrey Kantor wrote:
> Dmitri,
>
> This is a problem with two objectives which, depending on problem data, may
> conflict with each other. So the best we can do is to find tradeo
Dmitri,
This is a problem with two objectives which, depending on problem data, may
conflict with each other. So the best we can do is to find tradeoffs
between the conflicting objectives. Here's one solution where alpha is a
parameter expressing the relative significance of the two objectives.
s
Hi Jeff,
Thanks, yes that would be preferable packing. My goal is to pack items as
"tightly" as possible while preserving even (within some predefined margin)
weight distribution between bins.
Best regards,
Dmitri
On 27/01/2013, at 13:40, Jeffrey Kantor wrote:
> Hi Dmitri,
>
> This could be
Hi Dmitri,
This could be done through the objective function. But before going there,
I'm wondering a bit
about the solution you're showing for your example. Wouldn't you want move
item 7 from Bin 3 to Bin 2 to meet your primary objective?
Jeff
On Sun, Jan 27, 2013 at 1:34 PM, Dmitri Goutnik
Hello everyone,
I modified bpp.mod to pack items in predefined number of bins of unlimited
capacity so that weight is (approximately) evenly distributed between bins. In
addition to weight, each item has an "order" or "position". Is there a way to
write an additional model constraint so that it
