I am fairly new to Java OPL interface and I am being faced with a problem about how to obtain the values of the decision variables after CPLEX solves it. More particularly, I have this code which integrates an OPL model file and a data file:
String inDataFile = "F:\\UPS\\OPL\\data_mod_6.dat";
String modfileloc = "F:\\UPS\\OPL\\model2.mod";
IloOplFactory.setDebugMode(false);
IloOplFactory oplF = new IloOplFactory();
IloOplErrorHandler errHandler = oplF.createOplErrorHandler(System.out);
IloOplModelSource modelSource = oplF.createOplModelSource(modfileloc);
IloCplex cplex = null;
try {
cplex = oplF.createCplex();
} catch (IloException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
IloOplSettings settings = oplF.createOplSettings(errHandler);
IloOplModelDefinition def=oplF.createOplModelDefinition(modelSource,settings);
IloOplModel opl=oplF.createOplModel(def,cplex);
//IloOplDataElements DataElements = opl.makeDataElements();
IloOplDataSource dataSource=oplF.createOplDataSource(inDataFile);
opl.addDataSource(dataSource);
opl.generate();
if (cplex.solve()){
IloNumMap sw1 = opl.getElement("sw").asNumMap();
double [] sw = null;
int sw_length = sw1.getSize();
for (int i=0; i< sw_length; i++)
{
sw[i] = sw1.get(i+1);
}
}
Now sw here is an "double" vector decision variable which is obtained from the cplex.solve() step. However it gives me an error saying:
Exception in thread "main" ilog.concert.IloException: Type dvar float+ expected for element "sw"
at ilog.opl.lang.wrapJNI.IloOplElement_cpp_asNumMap(Native Method)
Please help me as to whether it's a declaration error or am I missing some of the libraries for OPL (I have only the oplall.jar files from the CPLEx installation kit)
Regards,
Praveen
If sw
is a dvar float+
, then you rather use asNumVarMap
rather than asNumMap
.
You should have a look at examples in
opl\examples\opl_interfaces\java