Lecture 14 - Applications to Networks

Learning Objectives

What is a "Network"?

A network is a set of points called nodes connected by branches:

A network with four nodes and seven branches

The total flow into the network equals the total flow out of the network. The total flow into each node equals the total flow out of that node.

Networks can be used to model a variety of real world situations:

An Extended Example

To understand how our linear algebra ideas can help us understand flow in a network, consider this network, representing streets and traffic flow in a neighborhood:

A network with four nodes and eight branches, some of which are labeled with variables

The branches labeled with numbers represent streets where we know the traffic flow. The variable branches are where the flow is unknown. We would like to use the information we have to draw conclusions about the possible flow rates on the variable branches.

The principle that "flow in equals flow out" allows us to set up equations involving these variables. We will have one equation for each node, plus one more equation for the network as a whole.

The equations are:

Node A: \( x_1 + x_4 = 50 + 75 \)
Node B: \( 200 = x_1 + x_2 \)
Node C: \( x_2 + 75 = x_3 \)
Node D: \( 50 + x_3 = x_5 \)
Network: \( 200 + x_4 = x_5 \)

Now, we rewrite the equations in the standard format, moving the variables to the left-hand side and the numbers to the right-hand side: \[ \begin{eqnarray*} x_1 + x_4 & = & 125 \\ x_1 + x_2 & = & 200 \\ -x_2 + x_3 & = & 75 \\ -x_3 + x_5 & = & 50 \\ -x_4 + x_5 & = & 200 \end{eqnarray*} \]

Next, we construct and row-reduce the corresponding augmented matrix: \[ \begin{bmatrix} 1 & 0 & 0 & 1 & 0 & 125 \\ 1 & 1 & 0 & 0 & 0 & 200 \\ 0 & -1 & 1 & 0 & 0 & 75 \\ 0 & 0 & -1 & 0 & 1 & 50 \\ 0 & 0 & 0 & -1 & 1 & 200 \end{bmatrix} \longrightarrow \begin{bmatrix} 1 & 0 & 0 & 0 & 1 & 325 \\ 0 & 1 & 0 & 0 & -1 & -125 \\ 0 & 0 & 1 & 0 & -1 & -50 \\ 0 & 0 & 0 & 1 & -1 & -200 \\ 0 & 0 & 0 & 0 & 0 & 0 \end{bmatrix} \]

The general solution is:

Since we have a free variable, there are infinitely many solutions. However, there are still limitations on what the solutions could be, based on the fact that the branches represent one-way streets. This means that the variables cannot be negative.

For example, we cannot have \( x_5 = 400 \), since this would make \( x_1 \) negative. Similarly, we cannot have \( x_5 = 100 \), since then \( x_2 \) would be negative.

The restriction \( x_1 \ge 0 \) gives \( x_5 \le 325 \). The restriction \( x_4 \ge 0 \) gives \( x_5 \ge 200 \). As long as \( 200 \le x_5 \le 325 \), all five variables will be nonnegative.

Summary

When using linear algebra to analyze network flow, you will have on variable for each unknown branch. You will also have one equation for each node, plus one more equation for the entire network.

Keep in mind that the system of equations may not be homogeneous, which means that there may not be any solutions! If the system is consistent, make sure to find solution(s) where all of the variables are nonnegative.

« Lecture 13 Back to Top Lecture 15 »