[BCP] semi-continuous, sos1, ... variables

[David Catteeuw]

Hi Ted,

Thanks for offering your help!

DIP seems promising indeed.

I have a cutting-stock problem where the demands that must be satisfied have a predefined width and a _total_ length. In other words, you may/must cut several elements to satisfy a demand but you can freely choose the length of each element (within some reasonable range).

The cutting patterns are 2D guillotine cuts whose order is fixed: first cut to fix the elements' length, then cut to determine the elements' width. 
You can cut in 2 stages: the elements just obtained can be cut into elements of smaller width but equal length.

I decided to tackle the problem as follows:
- a pattern is a tree of widths (a tree to represent 2 stage cutting).
- the master problem decides over which length to apply each pattern. These decision variables are semi-continuous.
- the subproblem finds the best pattern (or several good ones)

I wonder whether DIP would be a better option for my problem.

Thanks,
david.

PS: It seems that the documents that projects.coin-or.org/Dip <http://projects.coin-or.org/Dip> refers to are currently not accessible. I'll have a closer look at DIP as soon as possible.

> On 19 Mar 2016, at 16:24, Ted Ralphs <tkralphs at lehigh.edu> wrote:
> 
> Hi David,
> 
> I can probably help you a bit with BCP, but it would be useful if you could describe a little more about what you're trying to do exactly, since there may be better options, such as DIP (projects.coin-or.org/Dip <http://projects.coin-or.org/Dip>), which is actively developed. It's been a long time since I really delved into BCP, so it might take some time to answer your questions if you want to continue with BCP. 
> 
> Cheers,
> 
> Ted
> 
> On Sun, Mar 13, 2016 at 4:46 AM, David Catteeuw <david.catteeuw at yazzoom.com <mailto:david.catteeuw at yazzoom.com>> wrote:
> Hi all,
> 
> Do I understand correctly that semi-continuous, SOS variables, etc. must be dealt with manually?
> 
> In other words, in BCP you cannot do something like:
>         BCP_var_algo(BCP_SemiContinuousVar, cost_coefficient, lower_bound, upper_bound);
> to create a variable whose value is either between lower_bound and upper_bound, or 0.
> 
> Thanks,
> david.
> _______________________________________________
> BCP mailing list
> BCP at list.coin-or.org <mailto:BCP at list.coin-or.org>
> http://list.coin-or.org/mailman/listinfo/bcp <http://list.coin-or.org/mailman/listinfo/bcp>
> 
> 
> 
> -- 
> Dr. Ted Ralphs
> Professor, Lehigh University
> (610) 628-1280
> ted 'at' lehigh 'dot' edu
> coral.ie.lehigh.edu/~ted <http://coral.ie.lehigh.edu/~ted>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/bcp/attachments/20160320/19c948ea/attachment.html>