[SCIP] Question About IIS Functionality in SCIP

[Ameya Munagekar]

Hi Mark,

Thanks for the step-by-step breakdown. Super helpful—I’ll get right on it
and try out the process you shared. Looking forward to when it’s released
to PySCIPOpt.

Thank you for getting back! Cheers!

Best,
Ameya

On Mon, 27 Jan, 2025, 15:18 Mark Turner, <turner at zib.de> wrote:

> Dear Ameya,
>
>
> You should now be able to create an IIS with SCIP!
>
> Things that you'd need to do:
>
> - Clone the current master branch from the GitHub
>
> - Build SCIP
>
> - Load your infeasible problem in the SCIP shell with command `read
> path_to_instance_file` (you can simply write out the instance from your
> python script)
>
> - Run the command `iis`
>
> - Then either run `write iis path_to_new_instance_file` or `display iis`
> to see an IIS of your original instance.
>
> It will take a couple of months before this feature is in the official
> SCIP release, and therefore also won't be available via PySCIPOpt until
> then. After that date there's going to be some nice Python function like
> `computeIIS` or `generateIIS`.
>
>
> Cheers and have a good week,
>
> Mark
>
> On 1/12/25 12:20, Kamp, Dominik wrote:
> > Dear Ameya,
> >
> > MIP-DD (https://github.com/scipopt/MIP-DD) on the develop branch
> provides some kind of IIS-heuristic, which can simply be set up by linking
> SCIP to it and adding option -o feasible when calling.
> >
> > However, since it is rather designed to investigate the underlying
> solver instead of the given problem, you might want to wait for the
> IIS-plugin in SCIP announced in an earlier message on this list, which
> could be available soon on the master branch of SCIP.
> >
> > Best regards,
> >
> > Dominik
> >
> >> On 12. Jan 2025, at 07:50, Ameya Munagekar <amunagekar at gmail.com>
> wrote:
> >>
> >> Hi SCIP Team,
> >>
> >> I hope you're doing well!
> >>
> >> I'm reaching out to check if SCIP has a feature for identifying an
> Irreducible Inconsistent Subsystem (IIS),
> >> I'm working on a project in python where I need to analyze
> infeasibility in optimization models, and I was wondering if SCIP can help
> with this directly or if there's a recommended workaround, any pointers or
> advice would be really helpful.
> >>
> >> Thanks in advance for your time!
> >> Regards,
> >> Ameya
> >> _______________________________________________
> >> Scip mailing list
> >> Scip at zib.de
> >> https://listserv.zib.de/mailman/listinfo/scip
> >
> > _______________________________________________
> > Scip mailing list
> > Scip at zib.de
> > https://listserv.zib.de/mailman/listinfo/scip
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.zib.de/pipermail/scip/attachments/20250127/5155d6db/attachment.html>