Originally posted by: SystemAdmin

Hi everybody,

I was wondering if anyone has seen this before that while models is infeasible cplexstatus is OptimalTol?
In the following i first write the lp file then solve the problem and finally ask for the status.

Warning:  Output names have been modified to conform to LP format. Infeasibility row 
'id47867':  0 <= -59702.3. Presolve time =    0.00 sec. OptimalTol

any comment appreciated.
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

Can you reproduce the behavior with the interactive CPLEX using the LP file that you dumped?
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

The thing is that the optimzier reports infeasible.
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

Yes, I understand. My question is whether you can see the same in the interactive optimizer, namely that the presolver claims to have found an infeasible row, while the final MIP status message claims optimality within tolerance.
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

No, the interactive optimizer says simply infeasible.
I cleaned up the VS project an I cannot even reproduce it again in C++.
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: EdKlotz

> Shahin G wrote:
> No, the interactive optimizer says simply infeasible.
> I cleaned up the VS project an I cannot even reproduce it again in C++.

From interactive CPLEX, try running the conflict refiner to get an explanation
of the infeasibility. After you get the infeasible result, do

CPLEX> conf
CPLEX> write myconf.clp

Quit interactive CPLEX and examine the clp file. It should list a minimal conflict that hopefully will explain the infeasibility. Once we have an
explanation of the infeasibility, we may be able to figure out what is happening
regarding the OptimalTol status.
#DecisionOptimization
#MathematicalProgramming-General