Maximum Flows

In the following sections, you get an example of a maximum flow (max flow) problem.

A max flow example

The problem is defined by the following graph, which represents a transportation network:

network flow graph

You want to transport material from node 0 (the source) to node 4 (the sink). The numbers next to the arcs are their capacities — the capacity of an arc is the maximum amount that can be transported across it in a fixed period of time. The capacities are the constraints for the problem.

A flow is an assignment of a non-negative number to each arc (the flow amount) that satisfies the following flow conservation rule:

The max flow problem is to find a flow for which the sum of the flow amounts for the entire network is as large as possible.

The following sections present a programs to find the maximum flow from the source (0) to the sink (4).

Import the libraries

The following code imports the required library.

Python

import numpy as np

from ortools.graph.python import max_flow

C++

#include <cstdint>
#include <vector>

#include "ortools/graph/max_flow.h"

Java

import com.google.ortools.Loader;
import com.google.ortools.graph.MaxFlow;

C#

using System;
using Google.OrTools.Graph;

Declare the solver

To solve the problem, you can use the SimpleMaxFlow solver.

Python

# Instantiate a SimpleMaxFlow solver.
smf = max_flow.SimpleMaxFlow()

C++

// Instantiate a SimpleMaxFlow solver.
SimpleMaxFlow max_flow;

Java

// Instantiate a SimpleMaxFlow solver.
MaxFlow maxFlow = new MaxFlow();

C#

// Instantiate a SimpleMaxFlow solver.
MaxFlow maxFlow = new MaxFlow();

Define the data

You define the graph for the problem with three arrays, for the start nodes, end nodes, and capacities of the arcs. The length of each array equals the number of arcs in the graph.

For each i, arc i goes from start_nodes[i] to end_nodes[i], and its capacity is given by capacities[i]. The next section shows how to create the arcs using this data.

Python

# Define three parallel arrays: start_nodes, end_nodes, and the capacities
# between each pair. For instance, the arc from node 0 to node 1 has a
# capacity of 20.
start_nodes = np.array([0, 0, 0, 1, 1, 2, 2, 3, 3])
end_nodes = np.array([1, 2, 3, 2, 4, 3, 4, 2, 4])
capacities = np.array([20, 30, 10, 40, 30, 10, 20, 5, 20])

C++

// Define three parallel arrays: start_nodes, end_nodes, and the capacities
// between each pair. For instance, the arc from node 0 to node 1 has a
// capacity of 20.
std::vector<int64_t> start_nodes = {0, 0, 0, 1, 1, 2, 2, 3, 3};
std::vector<int64_t> end_nodes = {1, 2, 3, 2, 4, 3, 4, 2, 4};
std::vector<int64_t> capacities = {20, 30, 10, 40, 30, 10, 20, 5, 20};

Java

// Define three parallel arrays: start_nodes, end_nodes, and the capacities
// between each pair. For instance, the arc from node 0 to node 1 has a
// capacity of 20.
// From Taha's 'Introduction to Operations Research',
// example 6.4-2.
int[] startNodes = new int[] {0, 0, 0, 1, 1, 2, 2, 3, 3};
int[] endNodes = new int[] {1, 2, 3, 2, 4, 3, 4, 2, 4};
int[] capacities = new int[] {20, 30, 10, 40, 30, 10, 20, 5, 20};

C#

// Define three parallel arrays: start_nodes, end_nodes, and the capacities
// between each pair. For instance, the arc from node 0 to node 1 has a
// capacity of 20.
// From Taha's 'Introduction to Operations Research',
// example 6.4-2.
int[] startNodes = { 0, 0, 0, 1, 1, 2, 2, 3, 3 };
int[] endNodes = { 1, 2, 3, 2, 4, 3, 4, 2, 4 };
int[] capacities = { 20, 30, 10, 40, 30, 10, 20, 5, 20 };

Add the arcs

For each start node and end node, you create an arc from start node to end node with the given capacity, using the method AddArcWithCapacity. The capacities are the constraints for the problem.

Python

# Add arcs in bulk.
#   note: we could have used add_arc_with_capacity(start, end, capacity)
all_arcs = smf.add_arcs_with_capacity(start_nodes, end_nodes, capacities)

C++

// Add each arc.
for (int i = 0; i < start_nodes.size(); ++i) {
  max_flow.AddArcWithCapacity(start_nodes[i], end_nodes[i], capacities[i]);
}

Java

// Add each arc.
for (int i = 0; i < startNodes.length; ++i) {
  int arc = maxFlow.addArcWithCapacity(startNodes[i], endNodes[i], capacities[i]);
  if (arc != i) {
    throw new Exception("Internal error");
  }
}

C#

// Add each arc.
for (int i = 0; i < startNodes.Length; ++i)
{
    int arc = maxFlow.AddArcWithCapacity(startNodes[i], endNodes[i], capacities[i]);
    if (arc != i)
        throw new Exception("Internal error");
}

Invoke the solver

Now that all the arcs have been defined, all that remains is to invoke the solver and display the results. You invoke the Solve() method, providing the source (0) and sink (4).

Python

# Find the maximum flow between node 0 and node 4.
status = smf.solve(0, 4)

C++

// Find the maximum flow between node 0 and node 4.
int status = max_flow.Solve(0, 4);

Java

// Find the maximum flow between node 0 and node 4.
MaxFlow.Status status = maxFlow.solve(0, 4);

C#

// Find the maximum flow between node 0 and node 4.
MaxFlow.Status status = maxFlow.Solve(0, 4);

Display the results

Now, you can display the flow across each arc.

Python

if status != smf.OPTIMAL:
    print("There was an issue with the max flow input.")
    print(f"Status: {status}")
    exit(1)
print("Max flow:", smf.optimal_flow())
print("")
print(" Arc    Flow / Capacity")
solution_flows = smf.flows(all_arcs)
for arc, flow, capacity in zip(all_arcs, solution_flows, capacities):
    print(f"{smf.tail(arc)} / {smf.head(arc)}   {flow:3}  / {capacity:3}")
print("Source side min-cut:", smf.get_source_side_min_cut())
print("Sink side min-cut:", smf.get_sink_side_min_cut())

C++

if (status == MaxFlow::OPTIMAL) {
  LOG(INFO) << "Max flow: " << max_flow.OptimalFlow();
  LOG(INFO) << "";
  LOG(INFO) << "  Arc    Flow / Capacity";
  for (std::size_t i = 0; i < max_flow.NumArcs(); ++i) {
    LOG(INFO) << max_flow.Tail(i) << " -> " << max_flow.Head(i) << "  "
              << max_flow.Flow(i) << "  / " << max_flow.Capacity(i);
  }
} else {
  LOG(INFO) << "Solving the max flow problem failed. Solver status: "
            << status;
}

Java

if (status == MaxFlow.Status.OPTIMAL) {
  System.out.println("Max. flow: " + maxFlow.getOptimalFlow());
  System.out.println();
  System.out.println("  Arc     Flow / Capacity");
  for (int i = 0; i < maxFlow.getNumArcs(); ++i) {
    System.out.println(maxFlow.getTail(i) + " -> " + maxFlow.getHead(i) + "    "
        + maxFlow.getFlow(i) + "  /  " + maxFlow.getCapacity(i));
  }
} else {
  System.out.println("Solving the max flow problem failed. Solver status: " + status);
}

C#

if (status == MaxFlow.Status.OPTIMAL)
{
    Console.WriteLine("Max. flow: " + maxFlow.OptimalFlow());
    Console.WriteLine("");
    Console.WriteLine("  Arc     Flow / Capacity");
    for (int i = 0; i < maxFlow.NumArcs(); ++i)
    {
        Console.WriteLine(maxFlow.Tail(i) + " -> " + maxFlow.Head(i) + "    " +
                          string.Format("{0,3}", maxFlow.Flow(i)) + "  /  " +
                          string.Format("{0,3}", maxFlow.Capacity(i)));
    }
}
else
{
    Console.WriteLine("Solving the max flow problem failed. Solver status: " + status);
}

Here is the output of the program:

Max flow: 60

  Arc    Flow / Capacity
0 -> 1    20  /  20
0 -> 2    30  /  30
0 -> 3    10  /  10
1 -> 2     0  /  40
1 -> 4    20  /  30
2 -> 3    10  /  10
2 -> 4    20  /  20
3 -> 2     0  /   5
3 -> 4    20  /  20
Source side min-cut: [0]
Sink side min-cut: [4, 1]

The flow amounts across each arc are displayed under Flow.

Complete programs

Putting it all together, here are the complete programs.

Python

"""From Taha 'Introduction to Operations Research', example 6.4-2."""
import numpy as np

from ortools.graph.python import max_flow


def main():
    """MaxFlow simple interface example."""
    # Instantiate a SimpleMaxFlow solver.
    smf = max_flow.SimpleMaxFlow()

    # Define three parallel arrays: start_nodes, end_nodes, and the capacities
    # between each pair. For instance, the arc from node 0 to node 1 has a
    # capacity of 20.
    start_nodes = np.array([0, 0, 0, 1, 1, 2, 2, 3, 3])
    end_nodes = np.array([1, 2, 3, 2, 4, 3, 4, 2, 4])
    capacities = np.array([20, 30, 10, 40, 30, 10, 20, 5, 20])

    # Add arcs in bulk.
    #   note: we could have used add_arc_with_capacity(start, end, capacity)
    all_arcs = smf.add_arcs_with_capacity(start_nodes, end_nodes, capacities)

    # Find the maximum flow between node 0 and node 4.
    status = smf.solve(0, 4)

    if status != smf.OPTIMAL:
        print("There was an issue with the max flow input.")
        print(f"Status: {status}")
        exit(1)
    print("Max flow:", smf.optimal_flow())
    print("")
    print(" Arc    Flow / Capacity")
    solution_flows = smf.flows(all_arcs)
    for arc, flow, capacity in zip(all_arcs, solution_flows, capacities):
        print(f"{smf.tail(arc)} / {smf.head(arc)}   {flow:3}  / {capacity:3}")
    print("Source side min-cut:", smf.get_source_side_min_cut())
    print("Sink side min-cut:", smf.get_sink_side_min_cut())


if __name__ == "__main__":
    main()

C++

// From Taha 'Introduction to Operations Research', example 6.4-2."""
#include <cstdint>
#include <vector>

#include "ortools/graph/max_flow.h"

namespace operations_research {
// MaxFlow simple interface example.
void SimpleMaxFlowProgram() {
  // Instantiate a SimpleMaxFlow solver.
  SimpleMaxFlow max_flow;

  // Define three parallel arrays: start_nodes, end_nodes, and the capacities
  // between each pair. For instance, the arc from node 0 to node 1 has a
  // capacity of 20.
  std::vector<int64_t> start_nodes = {0, 0, 0, 1, 1, 2, 2, 3, 3};
  std::vector<int64_t> end_nodes = {1, 2, 3, 2, 4, 3, 4, 2, 4};
  std::vector<int64_t> capacities = {20, 30, 10, 40, 30, 10, 20, 5, 20};

  // Add each arc.
  for (int i = 0; i < start_nodes.size(); ++i) {
    max_flow.AddArcWithCapacity(start_nodes[i], end_nodes[i], capacities[i]);
  }

  // Find the maximum flow between node 0 and node 4.
  int status = max_flow.Solve(0, 4);

  if (status == MaxFlow::OPTIMAL) {
    LOG(INFO) << "Max flow: " << max_flow.OptimalFlow();
    LOG(INFO) << "";
    LOG(INFO) << "  Arc    Flow / Capacity";
    for (std::size_t i = 0; i < max_flow.NumArcs(); ++i) {
      LOG(INFO) << max_flow.Tail(i) << " -> " << max_flow.Head(i) << "  "
                << max_flow.Flow(i) << "  / " << max_flow.Capacity(i);
    }
  } else {
    LOG(INFO) << "Solving the max flow problem failed. Solver status: "
              << status;
  }
}

}  // namespace operations_research

int main() {
  operations_research::SimpleMaxFlowProgram();
  return EXIT_SUCCESS;
}

Java

package com.google.ortools.graph.samples;
import com.google.ortools.Loader;
import com.google.ortools.graph.MaxFlow;

/** Minimal MaxFlow program. */
public final class SimpleMaxFlowProgram {
  public static void main(String[] args) throws Exception {
    Loader.loadNativeLibraries();
    // Instantiate a SimpleMaxFlow solver.
    MaxFlow maxFlow = new MaxFlow();

    // Define three parallel arrays: start_nodes, end_nodes, and the capacities
    // between each pair. For instance, the arc from node 0 to node 1 has a
    // capacity of 20.
    // From Taha's 'Introduction to Operations Research',
    // example 6.4-2.
    int[] startNodes = new int[] {0, 0, 0, 1, 1, 2, 2, 3, 3};
    int[] endNodes = new int[] {1, 2, 3, 2, 4, 3, 4, 2, 4};
    int[] capacities = new int[] {20, 30, 10, 40, 30, 10, 20, 5, 20};

    // Add each arc.
    for (int i = 0; i < startNodes.length; ++i) {
      int arc = maxFlow.addArcWithCapacity(startNodes[i], endNodes[i], capacities[i]);
      if (arc != i) {
        throw new Exception("Internal error");
      }
    }

    // Find the maximum flow between node 0 and node 4.
    MaxFlow.Status status = maxFlow.solve(0, 4);

    if (status == MaxFlow.Status.OPTIMAL) {
      System.out.println("Max. flow: " + maxFlow.getOptimalFlow());
      System.out.println();
      System.out.println("  Arc     Flow / Capacity");
      for (int i = 0; i < maxFlow.getNumArcs(); ++i) {
        System.out.println(maxFlow.getTail(i) + " -> " + maxFlow.getHead(i) + "    "
            + maxFlow.getFlow(i) + "  /  " + maxFlow.getCapacity(i));
      }
    } else {
      System.out.println("Solving the max flow problem failed. Solver status: " + status);
    }
  }

  private SimpleMaxFlowProgram() {}
}

C#

// From Taha 'Introduction to Operations Research', example 6.4-2.
using System;
using Google.OrTools.Graph;

public class SimpleMaxFlowProgram
{
    static void Main()
    {
        // Instantiate a SimpleMaxFlow solver.
        MaxFlow maxFlow = new MaxFlow();

        // Define three parallel arrays: start_nodes, end_nodes, and the capacities
        // between each pair. For instance, the arc from node 0 to node 1 has a
        // capacity of 20.
        // From Taha's 'Introduction to Operations Research',
        // example 6.4-2.
        int[] startNodes = { 0, 0, 0, 1, 1, 2, 2, 3, 3 };
        int[] endNodes = { 1, 2, 3, 2, 4, 3, 4, 2, 4 };
        int[] capacities = { 20, 30, 10, 40, 30, 10, 20, 5, 20 };

        // Add each arc.
        for (int i = 0; i < startNodes.Length; ++i)
        {
            int arc = maxFlow.AddArcWithCapacity(startNodes[i], endNodes[i], capacities[i]);
            if (arc != i)
                throw new Exception("Internal error");
        }

        // Find the maximum flow between node 0 and node 4.
        MaxFlow.Status status = maxFlow.Solve(0, 4);

        if (status == MaxFlow.Status.OPTIMAL)
        {
            Console.WriteLine("Max. flow: " + maxFlow.OptimalFlow());
            Console.WriteLine("");
            Console.WriteLine("  Arc     Flow / Capacity");
            for (int i = 0; i < maxFlow.NumArcs(); ++i)
            {
                Console.WriteLine(maxFlow.Tail(i) + " -> " + maxFlow.Head(i) + "    " +
                                  string.Format("{0,3}", maxFlow.Flow(i)) + "  /  " +
                                  string.Format("{0,3}", maxFlow.Capacity(i)));
            }
        }
        else
        {
            Console.WriteLine("Solving the max flow problem failed. Solver status: " + status);
        }
    }
}