Originally posted by: SystemAdmin

Hi all,

There is probably a very simple solution to this but I am new to CPLEX and the Concert library so please forgive my ignorant question.

I'm trying to solve an ILP in Java. At a certain point I'm using

IloNumVar[] x = cplex.numVarArray( 1, 0, 1 );

But this produces a variable within the range 0-1: 0 <= x <= 1.
What do I do if I want x to be able to be only 0 ór 1: x \in {0,1}?

I already tried using IloIntVar but then I get: "ilog.cplex.CpxException: CPLEX Error 1017: Not available for mixed-integer programs." so that's probably not it.

Any pointers are much appreciated, thanks!
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

Try IloCplex.boolVarArray() or IloCplex.intVarArray() to create integral variables.
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

Thanks Daniel,

I already tried IloCplex.intVarArray() but I saw that I still was using a numVarArray somewhere else, and that causes problems when trying to retrieve variables with a range.

Thanks!
#CPLEXOptimizers
#DecisionOptimization

Originally posted by: SystemAdmin

> 3ckhardt wrote:
>
> I already tried IloCplex.intVarArray() but I saw that I still was using a numVarArray somewhere else, and that causes problems when trying to retrieve variables with a range.

It should not. This is the Java API, correct? The following sample code mixes integer variables into a numVarArray and works fine:

IloCplex cplex = new IloCplex();
IloNumVar[] array = cplex.numVarArray(3, 0, 100);
array[0] = cplex.numVar(0, 100, "x");
array[1] = cplex.boolVar("y");
array[2] = cplex.numVar(0.0, Double.MAX_VALUE, "z");
//...
if (cplex.solve()) {
  double[] v = cplex.getValues(array);  // gets values for x, y and z
  //...
}

Regarding error 1017, I doubt that has to do with the array declarations; more likely you asked for something (such as reduced costs) that exist for LPs but not for MIPs.

/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