Re: [Help-glpk] Sizing the problem...

[Jeffrey Kantor]

Hi Michael,

No invective here!

Could you say a bit more about the problems that are not finding a solution?  For example, the type of problem you're trying to solve, and some metrics regarding problem size, the number of integer and binary variables?  

Jeff

On Sun, Feb 3, 2013 at 10:48 AM, Michael Bramley <address@hidden> wrote:

Hello:

 

We are using Linux glpk v4.48.

 

This may be a school boy question, but I’ll ask and beg indulgence from those on the list.

 

We have a number of LP/MIP problems that range in size.  Most are quite manageable in glpk (yeah!!), while a small subset of others test the limits of glpk to either find a solution or run for days without a solution.  Indeed, the annoying bit is that some problems just seem to run for days and days, while CPLEX seems to find an answer in minutes.

 

Is there a way to determine this limit in advance, i.e. can we look at the problem and say that generates X conditions (or something else) and thus is not suited for glpk in its current state?

 

Is there a way that we could build this into glpk as a pre-processing option?

 

Looking for any ideas/help, and may even respond to invective. :)

MB

 

 

The contents of this message and any attachments to it are confidential and may be legally privileged.
If you have received this message in error you should delete it from your system immediately and advise the sender.
dunnhumby may monitor and record all emails. The views expressed in this email are those of the sender and not those of dunnhumby.

_______________________________________________
Help-glpk mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/help-glpk