Fixanddive.java


/* Copyright 2020, Gurobi Optimization, LLC */

/* Implement a simple MIP heuristic.  Relax the model,
 sort variables based on fractionality, and fix the 25% of
 the fractional variables that are closest to integer variables.
 Repeat until either the relaxation is integer feasible or
 linearly infeasible. */

import gurobi.*;
import java.util.*;

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

    // Comparison class used to sort variable list based on relaxation
    // fractionality

    class FractionalCompare implements Comparator<GRBVar> {
      public int compare(GRBVar v1, GRBVar v2) {
        try {
          double sol1 = Math.abs(v1.get(GRB.DoubleAttr.X));
          double sol2 = Math.abs(v2.get(GRB.DoubleAttr.X));
          double frac1 = Math.abs(sol1 - Math.floor(sol1 + 0.5));
          double frac2 = Math.abs(sol2 - Math.floor(sol2 + 0.5));
          if (frac1 < frac2) {
            return -1;
          } else if (frac1 == frac2) {
            return 0;
          } else {
            return 1;
          }
        } catch (GRBException e) {
          System.out.println("Error code: " + e.getErrorCode() + ". " +
              e.getMessage());
        }
        return 0;
      }
    }

    if (args.length < 1) {
      System.out.println("Usage: java Fixanddive filename");
      System.exit(1);
    }

    try {
      // Read model
      GRBEnv env = new GRBEnv();
      GRBModel model = new GRBModel(env, args[0]);

      // Collect integer variables and relax them
      ArrayList<GRBVar> intvars = new ArrayList<GRBVar>();
      for (GRBVar v : model.getVars()) {
        if (v.get(GRB.CharAttr.VType) != GRB.CONTINUOUS) {
          intvars.add(v);
          v.set(GRB.CharAttr.VType, GRB.CONTINUOUS);
        }
      }

      model.set(GRB.IntParam.OutputFlag, 0);
      model.optimize();

      // Perform multiple iterations. In each iteration, identify the first
      // quartile of integer variables that are closest to an integer value
      // in the relaxation, fix them to the nearest integer, and repeat.

      for (int iter = 0; iter < 1000; ++iter) {

        // create a list of fractional variables, sorted in order of
        // increasing distance from the relaxation solution to the nearest
        // integer value

        ArrayList<GRBVar> fractional = new ArrayList<GRBVar>();
        for (GRBVar v : intvars) {
          double sol = Math.abs(v.get(GRB.DoubleAttr.X));
          if (Math.abs(sol - Math.floor(sol + 0.5)) > 1e-5) {
            fractional.add(v);
          }
        }

        System.out.println("Iteration " + iter + ", obj " +
            model.get(GRB.DoubleAttr.ObjVal) + ", fractional " +
            fractional.size());

        if (fractional.size() == 0) {
          System.out.println("Found feasible solution - objective " +
              model.get(GRB.DoubleAttr.ObjVal));
          break;
        }

        // Fix the first quartile to the nearest integer value

        Collections.sort(fractional, new FractionalCompare());
        int nfix = Math.max(fractional.size() / 4, 1);
        for (int i = 0; i < nfix; ++i) {
          GRBVar v = fractional.get(i);
          double fixval = Math.floor(v.get(GRB.DoubleAttr.X) + 0.5);
          v.set(GRB.DoubleAttr.LB, fixval);
          v.set(GRB.DoubleAttr.UB, fixval);
          System.out.println("  Fix " + v.get(GRB.StringAttr.VarName) +
              " to " + fixval + " ( rel " + v.get(GRB.DoubleAttr.X) + " )");
        }

        model.optimize();

        // Check optimization result

        if (model.get(GRB.IntAttr.Status) != GRB.Status.OPTIMAL) {
          System.out.println("Relaxation is infeasible");
          break;
        }
      }

      // Dispose of model and environment
      model.dispose();
      env.dispose();

    } catch (GRBException e) {
      System.out.println("Error code: " + e.getErrorCode() + ". " +
          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