Originally posted by: RohithKumar

We are using Concert Technology (C++) in one of our application with cplex 12.1 version.
And, we have declared an IloNumVarArray for a bool type variable as below.

U = IloNumVarArray(env,umin,umax, ILOBOOL);

After solving, when we try to query the results of this variable using getValues method, we get the following error.

"Concert exception caught: IloExtractable 16017 IloNumVarI has not been extracted by IloAlgorithm 0x1f1b33e0"
How can this error be resolved? Is there any other method to query the results?

Also for the same variable U if umin equals umax then all such variables would be converted into a General variable (i.e. GENERALS refer to integer variables with default bounds 0 and +infinity). Is this is bug? How do we resolve this issue?

Regards
Rohith
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

> RohithKumar wrote:
> We are using Concert Technology (C++) in one of our application with cplex 12.1 version.
> And, we have declared an IloNumVarArray for a bool type variable as below.
>
> U = IloNumVarArray(env,umin,umax, ILOBOOL);
>
> After solving, when we try to query the results of this variable using getValues method, we get the following error.
>
> "Concert exception caught: IloExtractable 16017 IloNumVarI has not been extracted by IloAlgorithm 0x1f1b33e0"
> How can this error be resolved? Is there any other method to query the results?

This may seem like a silly question, but does every entry of the U array whose value is being request in getValues appear in either a constraint or the objective function in the model being solved? The "not extracted" message usually occurs when you are trying to get the value of a variable that did not actually appear in the model (perhaps an auxiliary variable you intended to use in a related model, or perhaps an entry in an array that for some reason was skipped over in the model).
>
> Also for the same variable U if umin equals umax then all such variables would be converted into a General variable (i.e. GENERALS refer to integer variables with default bounds 0 and +infinity). Is this is bug? How do we resolve this issue?

I don't understand. If umin = umax, then the "variables" are effectively constants. Why would it then be a bug to treat them as GENERAL, so long as the model correctly bounded them?

/Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: RohithKumar

Hi Paul,
"This may seem like a silly question, but does every entry of the U array whose value is being request in getValues appear in either a constraint or the objective function in the model being solved?"
Yes, every value of U variable is available within the Objective Function itself.

Regards
Rohith
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

You might try giving every variable a name. After solving, run getValue against each variable in a loop, catching any Concert exceptions and, in the catch block, printing out the name of the variable whose value is being checked. That will at least tell you which variable is triggering the exception.

The message suggests that, for some reason, CPLEX does not consider the variable to be part of the model you just solved.

/Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#CPLEXOptimizers
#DecisionOptimization