Decision Optimization

Hello, 
the scheduling model I wrote cannot be solved to optimality due to the large number of constraints and variables. Therefore I have decided to proceed by scheduling a subset of jobs at every iteration, save the solution and use it in the next iteration. 
In my .mod file I have 

{int} J = {1,2};
dvar float+ t[j in J, i in I]; 

and all the constraints go over the set J.
then the simplified version of my .cpp file, where I believe there is the issue, is

IloOplRunConfiguration rc(env,"./dynamic.mod",/dynamic.dat");
IloInt job=3;
while (job <= jobMax) 
  {
    rc.getOplModel().generate();
    rc.getCplex().solve();
    rc.getOplModel().getElement("J").asIntSet().add(job);
    job++;
  }

Before saving the solution at each iteration I wanted to test this...which it does not work! It returns an error the first time I tried to add a job to the set of jobs J

I also tried to iterate as in the example mulprod_main.cpp without any luck...

Any ideas??

Thanks

Sam

Hello Samuele,

               First, I think that you have to put J = {1,2}; in your .dat file and {int} J = ...; in your .mod file.

And after, you can retry to use the mulprod_main.cpp, i-e create an

IloOplRunConfiguration rcIteration(rc.getOplModel().getModelDefinition(), dataElements); in your loop.

Where dataElements  is defined as follow: IloOplDataElements dataElements = rc.getOplModel().makeDataElements(); before your loop.

Best regards

Hello, 
the scheduling model I wrote cannot be solved to optimality due to the large number of constraints and variables. Therefore I have decided to proceed by scheduling a subset of jobs at every iteration, save the solution and use it in the next iteration. 
In my .mod file I have 

{int} J = {1,2};
dvar float+ t[j in J, i in I]; 

and all the constraints go over the set J.
then the simplified version of my .cpp file, where I believe there is the issue, is

IloOplRunConfiguration rc(env,"./dynamic.mod",/dynamic.dat");
IloInt job=3;
while (job <= jobMax) 
  {
    rc.getOplModel().generate();
    rc.getCplex().solve();
    rc.getOplModel().getElement("J").asIntSet().add(job);
    job++;
  }

Before saving the solution at each iteration I wanted to test this...which it does not work! It returns an error the first time I tried to add a job to the set of jobs J

I also tried to iterate as in the example mulprod_main.cpp without any luck...

Any ideas??

Thanks

Sam

Hello, 
the scheduling model I wrote cannot be solved to optimality due to the large number of constraints and variables. Therefore I have decided to proceed by scheduling a subset of jobs at every iteration, save the solution and use it in the next iteration. 
In my .mod file I have 

{int} J = {1,2};
dvar float+ t[j in J, i in I]; 

and all the constraints go over the set J.
then the simplified version of my .cpp file, where I believe there is the issue, is

IloOplRunConfiguration rc(env,"./dynamic.mod",/dynamic.dat");
IloInt job=3;
while (job <= jobMax) 
  {
    rc.getOplModel().generate();
    rc.getCplex().solve();
    rc.getOplModel().getElement("J").asIntSet().add(job);
    job++;
  }

Before saving the solution at each iteration I wanted to test this...which it does not work! It returns an error the first time I tried to add a job to the set of jobs J

I also tried to iterate as in the example mulprod_main.cpp without any luck...

Any ideas??

Thanks

Sam

The problem seems to be in the way you're trying to modify the set J after generating the model. In CPLEX, once a model has been generated, it cannot be directly modified. Instead, you would have to regenerate the model with the new set J.

Here's an example of how you could do it:

IloInt job=3;while (job <= jobMax) {  rc.end();  rc = IloOplRunConfiguration(env,"./dynamic.mod",/dynamic.dat");  rc.getOplModel().getElement("J").asIntSet().add(job);  rc.getOplModel().generate();  rc.getCplex().solve();  job++;}

In this code, rc.end(); ends the current configuration before creating a new one with the updated set J. Then, the model is generated and solved as before.

I hope this helps."

Hello, 
the scheduling model I wrote cannot be solved to optimality due to the large number of constraints and variables. Therefore I have decided to proceed by scheduling a subset of jobs at every iteration, save the solution and use it in the next iteration. 
In my .mod file I have 

{int} J = {1,2};
dvar float+ t[j in J, i in I]; 

and all the constraints go over the set J.
then the simplified version of my .cpp file, where I believe there is the issue, is

IloOplRunConfiguration rc(env,"./dynamic.mod",/dynamic.dat");
IloInt job=3;
while (job <= jobMax) 
  {
    rc.getOplModel().generate();
    rc.getCplex().solve();
    rc.getOplModel().getElement("J").asIntSet().add(job);
    job++;
  }

Before saving the solution at each iteration I wanted to test this...which it does not work! It returns an error the first time I tried to add a job to the set of jobs J

I also tried to iterate as in the example mulprod_main.cpp without any luck...

Any ideas??

Thanks

Sam