Originally posted by: Filipe Costa

Hi all,

Before I had model that I run on docplexcloud from Decision Optimization. On dropsolve I can download the output as a json (Great!!).

However I added to the model a Main script block and now I cannot download anymore on dropsolve the solution as a json - the status is "Failed- System Error".

I know that I have solutions because in the log file I can visualize it.

What I need to add in order to have the result in a json file at dropsolve?

Hi,

you could have a look at docplexcloud APIs : https://developer.ibm.com/docloud/documentation/docloud/

Models with main work fine on docplexcloud

I managed to dropsolve the model I shared at https://www.ibm.com/developerworks/community/forums/html/topic?id=022856ea-463e-4ded-80d3-970a5aa38882&ps=25

regards

Originally posted by: Filipe Costa

Yes, But using that code no solution is directly returned.

On the optimization studio the status of the job is "Processed (Unknown)", instead of "Processed (Feasible solution)".

At drop solve, This is what I see IMG1.

As you can see there is no option to download the result (Before I had the main block I was able do download the result IMG2).

Regards

Hi,

with the CPLEX OPL IDE in order to get the results back from docplexcloud you could use a tuple set

For instance if I run in the IDE

using CP;
 
 int nbKids=300;
    float costBus40=500;
    float costBus30=400;
     
    dvar int+ nbBus40;
    dvar int+ nbBus30;
     
    minimize
     costBus40*nbBus40  +nbBus30*costBus30;
     
    subject to
    {
     ctKids:40*nbBus40+nbBus30*30>=nbKids;
    }
    
    tuple t
    {
    int b40;
    int b30;    
    }
    
    {t} res;

    execute
    {
    writeln("nbBus40 = ",nbBus40);
    writeln("nbBus30 = ",nbBus30);
    res.add(nbBus40,nbBus30);
    }

    main
    {
    thisOplModel.generate();

    cp.solve();
    thisOplModel.postProcess();

    //now 350 kids instead of 300
    writeln("now 350 kids instead of 300");
        
    thisOplModel.ctKids.LB=350;

    cp.solve();
    thisOplModel.postProcess();

    // no more than 4 buses 40 seats
    writeln("no more than 4 buses 40 seats");

    thisOplModel.nbBus40.UB=4;

    cp.solve();
    thisOplModel.postProcess();

   

   

     

    }

then if I run in the cloud and right click to get back the results I will see

regards

Originally posted by: Filipe Costa

Thanks for the answer.

But If at the beginning  you have this declaration:

thisOplModel.generate();
  var def  = thisOplModel.modelDefinition;
  var data = thisOplModel.dataElements;
  var data2 = thisOplModel.dataElements;
  var parm = thisOplModel.settings;

How would you do the first step using var opl1 = new IloOplModel(def, cp); and for the second step  using :

var cp2 = new IloCP();
var opl2 = new IloOplModel(def, cp2)

?

Regards,

Hi,

if I adapt that to your model, this gives

 using CP;
 
 tuple t
 {
 float a;
 float b;
 }
 
 {t} res={};
 
 dvar int x;
 
 subject to
 {
 x==2;
 }
 
 
 
 main{
 
 var o1=0;
 var o2=0;
 
  //thisOplModel.generate();
  var source = new IloOplModelSource("sub.mod");
 
  var def = new IloOplModelDefinition(source);
  var data = new IloOplDataElements();
  var parm = thisOplModel.settings;
  var opl1 = new IloOplModel(def,cp);
  data.use_delay=1;
  data.use_bal=1;
  data.use_isi_grade=1;
 
   data.use_setup=0;
  data.use_isi_grade2=0;
  opl1.addDataSource(data);
 
  opl1.generate();
  //  STEP 1
  writeln("STEP 1: Minimizing Bal,Delay,ISI");
  cp.param.TimeLimit = 50;
  cp.param.NoOverlapInferenceLevel = "Extended";
  cp.solve();   
  writeln(opl1.printSolution());
  o1=cp.getObjValue();
 
  var cp2 = new IloCP();
  var opl2 = new IloOplModel(def, cp2);
 
 
  data.use_setup=1;
  data.use_isi_grade2=1;
  opl2.addDataSource(data);
  opl2.generate();
 
 

 

 
  //  STEP 2
  writeln("STEP 2: Minimizing regular machines usage");
  cp2.param.TimeLimit = 5400;
  cp2.param.NoOverlapInferenceLevel = "Extended";
  cp2.solve();
  o2=cp2.getObjValue();  
  writeln(opl2.printSolution());
  opl1.end();
  opl2.end();
 
  var opl=thisOplModel;
  opl.generate();
  cp.solve();
  opl.postProcess();
 
 thisOplModel.res.add(o1,o2);
}

regards