[Symphony] (file) format for bi-objective problems

[Ted Ralphs]

My apologies for the delay in responding. I recently extended COIN-OR's
parse for problems in LP format so that it is capable of understanding
multiple objectives. If you just simply add another objective to an LP file
in the obvious way and pass it to SYMPHONY with the -L flag, it will solve
it as a multi-objective problem, e.g.,

./symphony -L s1c50w01.lp

If it's working, the start of the output should be something like:

*Using FIFO search order...*
*Using augmented Chebyshev weight 0.00001000*

*****************************************************
*****************************************************
*Now solving with gamma = 1.0 tau = 0.0*
*****************************************************
*****************************************************

Then, at the very end, you should get output like:

*****************************************************
*****************************************************
*Displaying non-dominated solution values and breakpoints*
*****************************************************
*****************************************************

*First Objective: -18224.000 Second Objective: -17584.000 Range: 1.008593 -
1.000      000*
*First Objective: -18256.000 Second Objective: -20568.000 Range: 0.714815 -
1.008      593*
*First Objective: -17916.000 Second Objective: -20629.000 Range: 0.766990 -
0.714      815*

etc.


With that said, the multiobjective capability now seems to be broken in
trunk and possibly in later releases in the 5.6 series. I haven't had a
chance to track down the problem, but it seems to be working in 5.6.2 at
least.

Cheers,

Ted



On Mon, Apr 4, 2016 at 6:18 AM, Sebastian Schenker <schenker at zib.de> wrote:

> Dear all,
>
> on https://projects.coin-or.org/SYMPHONY it is written that Symphony can
> be used to solve bi-objective MILPs. In the FAQ it is mentioned that it
> "will read AMPL/GMPL files using GLPKs parser, or MPS file using COIN-OR's
> parser".
>
> How does Symphony read bi-objective problems? Do you use an extended file
> format?
> I am not aware that AMPL and MPS (directly) supports several criteria.
>
> best,
> Sebastian
>
> --
> Sebastian Schenker
> Research Assistant
>
> Zuse Institute Berlin
> Department of Mathematical Optimization
> Tel.: +49 (0)30 84185-223
> Email: schenker at zib.de
>
> TU Berlin, Institute of Mathematics
> Combinatorial Optimization & Graph Algorithms group
> Tel.: +49 (0)30 314-78657
> Email: schenker at math.tu-berlin.de
> _______________________________________________
> Symphony mailing list
> Symphony at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/symphony
>



-- 
Dr. Ted Ralphs
Professor, Lehigh University
(610) 628-1280
ted 'at' lehigh 'dot' edu
coral.ie.lehigh.edu/~ted
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/symphony/attachments/20160414/07c29a00/attachment.html>