[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Help-glpk] help with .bv. model
From: |
Andrew Makhorin |
Subject: |
Re: [Help-glpk] help with .bv. model |
Date: |
Wed, 7 Mar 2007 07:23:57 +0300 |
> I am trying to count shift calendar problem, in the attachment you can see my
> model
>
> after some seconds i got optimal solution and trying to find integer
> solution, but after hours i have only this kind of output:
> +6434641: mip = 1.920000000e+02 >= 1.900000000e+02 1.0% (125068; 112713)
> ^^^rising ^^^ not going down anymore
>
> is it ok? do i have some problem in model?
This means that your model is hard for glpk b&b solver.
If a suboptimal solution (i.e. 192) is sufficient for you, you can
limit the solution time with the command-line option `--tmlim nnn';
note that the exact optimum is either 190 or 191 or 192.
Also you can try using the options `--intopt' or `--cuts'; sometimes
this helps.