Try our new documentation site (beta).


Multiobj.java


/* Copyright 2023, Gurobi Optimization, LLC */

/* Want to cover three different sets but subject to a common budget of
   elements allowed to be used. However, the sets have different priorities to
   be covered; and we tackle this by using multi-objective optimization. */

import gurobi.*;

public class Multiobj {
  public static void main(String[] args) {

    try {
      // Sample data
      int groundSetSize = 20;
      int nSubsets      = 4;
      int Budget        = 12;
      double Set[][] = new double[][]
      { { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 },
        { 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1 },
        { 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0 },
        { 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0 } };
      int    SetObjPriority[] = new int[] {3, 2, 2, 1};
      double SetObjWeight[]   = new double[] {1.0, 0.25, 1.25, 1.0};
      int e, i, status, nSolutions;

      // Create environment
      GRBEnv env = new GRBEnv("Multiobj.log");

      // Create initial model
      GRBModel model = new GRBModel(env);
      model.set(GRB.StringAttr.ModelName, "Multiobj");

      // Initialize decision variables for ground set:
      // x[e] == 1 if element e is chosen for the covering.
      GRBVar[] Elem = model.addVars(groundSetSize, GRB.BINARY);
      for (e = 0; e < groundSetSize; e++) {
        String vname = "El" + String.valueOf(e);
        Elem[e].set(GRB.StringAttr.VarName, vname);
      }

      // Constraint: limit total number of elements to be picked to be at most
      // Budget
      GRBLinExpr lhs = new GRBLinExpr();
      for (e = 0; e < groundSetSize; e++) {
        lhs.addTerm(1.0, Elem[e]);
      }
      model.addConstr(lhs, GRB.LESS_EQUAL, Budget, "Budget");

      // Set global sense for ALL objectives
      model.set(GRB.IntAttr.ModelSense, GRB.MAXIMIZE);

      // Limit how many solutions to collect
      model.set(GRB.IntParam.PoolSolutions, 100);

      // Set and configure i-th objective
      for (i = 0; i < nSubsets; i++) {
        GRBLinExpr objn = new GRBLinExpr();
        String vname = "Set" + String.valueOf(i);

        for (e = 0; e < groundSetSize; e++)
          objn.addTerm(Set[i][e], Elem[e]);

        model.setObjectiveN(objn, i, SetObjPriority[i], SetObjWeight[i],
                            1.0 + i, 0.01, vname);
      }

      // Save problem
      model.write("Multiobj.lp");

      // Optimize
      model.optimize();

      // Status checking
      status = model.get(GRB.IntAttr.Status);

      if (status == GRB.INF_OR_UNBD ||
          status == GRB.INFEASIBLE  ||
          status == GRB.UNBOUNDED     ) {
        System.out.println("The model cannot be solved " +
                 "because it is infeasible or unbounded");
        System.exit(1);
      }
      if (status != GRB.OPTIMAL) {
        System.out.println("Optimization was stopped with status " + status);
        System.exit(1);
      }

      // Print best selected set
      System.out.println("Selected elements in best solution:");
      System.out.println("\t");
      for (e = 0; e < groundSetSize; e++) {
        if (Elem[e].get(GRB.DoubleAttr.X) < .9) continue;
        System.out.print(" El" + e);
      }
      System.out.println();

      // Print number of solutions stored
      nSolutions = model.get(GRB.IntAttr.SolCount);
      System.out.println("Number of solutions found: " + nSolutions);

      // Print objective values of solutions
      if (nSolutions > 10) nSolutions = 10;
      System.out.println("Objective values for first " + nSolutions);
      System.out.println(" solutions:");
      for (i = 0; i < nSubsets; i++) {
        model.set(GRB.IntParam.ObjNumber, i);

        System.out.print("\tSet" + i);
        for (e = 0; e < nSolutions; e++) {
          System.out.print(" ");
          model.set(GRB.IntParam.SolutionNumber, e);
          double val = model.get(GRB.DoubleAttr.ObjNVal);
          System.out.print("      " + val);
        }
        System.out.println();
      }
      model.dispose();
      env.dispose();
    } catch (GRBException e) {
      System.out.println("Error code = " + e.getErrorCode());
      System.out.println(e.getMessage());
    }
  }
}

Try Gurobi for Free

Choose the evaluation license that fits you best, and start working with our Expert Team for technical guidance and support.

Evaluation License
Get a free, full-featured license of the Gurobi Optimizer to experience the performance, support, benchmarking and tuning services we provide as part of our product offering.
Academic License
Gurobi supports the teaching and use of optimization within academic institutions. We offer free, full-featured copies of Gurobi for use in class, and for research.
Cloud Trial

Request free trial hours, so you can see how quickly and easily a model can be solved on the cloud.

Search

Gurobi Optimization