[Cbc] Restricting computation time for infeasible problems

[Christopher Gross]

Hi John,

thanks for getting back to me.

I had a look at the CBC options and I cannot find setMaximumSeconds. Using
it in interactive mode gives me an error message: "No match for
setMaximumSeconds - ? for list of commands".

I do, however, already use the sec(onds) option in my CMPL file: "%opt cbc
sec 120". Maybe this is what you meant? As I said, this works fine if the
problem is feasible, however for infeasible problems it sometimes takes very
long, much longer than the 120 seconds I specified. My assumption is that
there is something going on which is not regularly interrupted to check for
the current running time.

Maybe I can send you some data to replicate the behavior. I will check and,
if possible, get back to you. Thanks in advance!

Best regards,

Christopher Gross

--
Christopher Gross, M.Sc.
Lehrstuhl für Health Care Operations/Health Information Management
UNIKA-T, Neusässer Straße 47, 86156 Augsburg

Tel.: +49 821 598-6454

-----Original Message-----
From: Cbc [mailto:cbc-bounces at coin-or.org] On Behalf Of John Forrest
Sent: Thursday, October 8, 2015 16:46
To: cbc at list.coin-or.org
Subject: Re: [Cbc] Restricting computation time for infeasible problems

Christopher,

The maximum time can easily be set - setMaximumSeconds.

Infeasible problems often take much longer.  If you want to send me some
data, I can have a quick look to see if there is something simple that can
be done.

John Forrest
On 06/10/15 15:44, Christopher Gross wrote:
> Hi,
>
> I am using CMPL with CBC as a backend in a software that is used 
> without me being present. The users are able to enter data into the 
> software which will then generate a data file and solve the model I 
> created. Currently, the software is in the testing stage so users 
> expect timely feedback. I have therefore limited the solution time to 
> 120 seconds which gives quite good solutions so far.
>
> Sometimes, users enter data that leads to infeasibilities in the model.
> Often, this results in a quick failure (within a few seconds) with an 
> error message. Sometimes, however, it might take very long for CBC to 
> fail. In the most recent instance of this problem it took the solver 
> more than 20 minutes of processing at which point the users called me 
> and asked me to kill the process so they could try again with different
data.
>
> Obviously, it is undesirable to have CBC processing for such a long time.
> Ideally, I would like to always terminate after 120 seconds and then 
> give the user the intermediate result, whether a solution has been found
or not.
>
> Why is CBC sometimes taking so long when presented with an infeasible 
> problem? And what can I do to limit this time?
>
> Thanks and best regards,
>
> Christopher
> --
> Christopher Gross, M.Sc.
>
> Lehrstuhl für Health Care Operations/Health Information Management
>
> Universitäres Zentrum für Gesundheitswissenschaften am Klinikum 
> Augsburg - UNIKA-T, Neusässer Straße 47, 86156 Augsburg
>
> Wirtschaftswissenschaftliche Fakultät, Universität Augsburg, 
> Universitätsstraße 16, 86159 Augsburg
>
> Tel.: +49 821 598-6454
> E-Mail: christopher.gross at unikat.uni-augsburg.de
> Web: http://www.wiwi.uni-augsburg.de/bwl/brunner/
>
>
> _______________________________________________
> Cbc mailing list
> Cbc at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/cbc
>
>

_______________________________________________
Cbc mailing list
Cbc at list.coin-or.org
http://list.coin-or.org/mailman/listinfo/cbc