[Coin-discuss] Construct a MIP problem

[Spencer Fung]

Well, I have read the code of OsiCpxSolverInteger.cpp, I think the problem
is caused by the function setInteger(). Since It will change the column type
by CPXchgctype() only when the problem is MIP, i.e. probtypemip_ is true.
But it will never be true before invoking the branchAndBound() function.

Is it a bug in there?

Spencer

-----Original Message-----
From: coin-discuss-bounces at list.coin-or.org
[mailto:coin-discuss-bounces at list.coin-or.org] On Behalf Of Ted Ralphs
Sent: Saturday, March 19, 2005 1:31 AM
To: Discussions about open source software for Operations Research
Subject: Re: [Coin-discuss] Construct a MIP problem

John J Forrest wrote:
> 
> Looking OsiCpx it looks as if it sets coltype_ in OsiCpx correctly but 
> if it is not a MIP in Cplex does not tell Cplex. Using writeMpsNative 
> should work.

Yes, the problem is that CPLEX will not store integrality information 
unless it thinks the problem is a MIP. By default, problems loaded into 
Osi are considered LPs. It is only when branchAndBound() is called that 
the problem type is set to MIP. This is an issue specific to CPLEX---I 
believe all of the other solvers will store integrality information even 
when the problem is being solved as an LP.

Ted
-- 
Dr. Ted Ralphs
Assistant Professor
Industrial and Systems Engineering
Lehigh University
(610)758-4784
tkralphs at lehigh.edu
www.lehigh.edu/~tkr2
_______________________________________________
Coin-discuss mailing list
Coin-discuss at list.coin-or.org
http://list.coin-or.org/mailman/listinfo/coin-discuss