MIP Initial Solution set up but not initial GAP

Archiver · December 8, 2012, 12:56am

Dear All,

I’m using GAMS with CPLEX to solve a MIP problem. I have set up an initial solution for one binary variable V_X(i,m) like this=

V_X.L(i,m)=1;

The thing is that is not showing any GAP, not even after an hour, and the logic tells me it should, am I right?

Or could be that this initial solution is not feasible?

Here are the CPLEX parameters I’m using:

$onecho > cplex.opt
threads 4
nodefileind 0
polishafterintsol 1
workmem = 3000
mipemphasis 1
$offecho

I’m new in this modeling world so, please be patient .

Thank you very much.

–
To view this discussion on the web visit https://groups.google.com/d/msg/gamsworld/-/kQaTE8GIRisJ.
To post to this group, send email to gamsworld@googlegroups.com.
To unsubscribe from this group, send email to gamsworld+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/gamsworld?hl=en.

Archiver · December 9, 2012, 5:49pm

Andres,

When I use a GAMS or solver feature for the first time I start with a smaller model whose solution and/or behavior is familiar. In your case, I’d suggest starting with the rotdk model from the GAMS model library: it’s nontrivial but won’t make you wait hours either. Solve it and save the solution using the savepoint facility, then modify the model to load the solution and use it. The log should indicate if you’ve got it working, and you’ll know what to look for in your model’s log.

BTW, you should google “GAMS mipstart” first, for example it turned up this nugget:

http://support.gams.com/doku.php?id=solver:using_the_mipstart_-_option

-Steve

On Fri, Dec 7, 2012 at 6:56 PM, AndrÃ©s Bego wrote: