Real-Life Application of Network Models—Saving Federal Travel Dollars

(Case Study Results)

 U.S. federal government offices are located in most cities in the

United States, and federal employees are required to attend
development conferences and training courses offered around
the country.
 The location of the city hosting conferences/training events can
impact travel costs.
 The goal of the Network Model study is to determine the optimal
location of host city for a scheduled conference/training event.
 For fiscal year 1997, the developed model was estimated to have
saved at least $400,000.
 SCOPE OF NETWORK MODELS
IN MANY COMPLEX PROBLEMS NETWORK MODELS CAN BE USED TO SOLVE
THEM. EXAMPLES OF SOME OF THESE SITUATIONS ARE:
1) Design of an offshore natural-gas pipeline network connecting wellheads in
the Gulf of Mexico to an inshore delivery point with the objective of
minimizing the cost of constructing the pipeline.
2) Determination of the shortest route between two cities in an existing network
of roads.
3) Determination of the maximum capacity (in tons per year) of a coal slurry
pipeline network joining coal mines in Wyoming with power plants in
Houston. (Slurry pipelines transport coal by pumping water through specially
designed pipes)
4) Determination of the time schedule (start and completion dates) for the
activities of a construction project.

The solution of these complex problems is possible through network

optimization algorithms. Four of these Network Algorithms are discussed.
1. Minimal spanning tree (Can be used in Situation 1)
2. Shortest-route algorithm (Can be used in Situation 2)
3. Maximal-flow algorithm (Can be used in Situation 3)
4. Critical Path Method (CPM) algorithm
i.e. CPM–PERT project-scheduling models
(Can be used in Situation 4)
 Network Terminology
 A network consists of a set of nodes linked by arcs (or branches). The notation for describing a network is
(N, A), where N is the set of nodes, and A is the set of arcs. As an illustration, the network in Figure 1 is
described as
N = {1, 2, 3, 4, 5}
A = {(1, 2), (1, 3), (2, 3), (2, 5), (3, 4), (3, 5), (4, 2), (4, 5)}
 Associated with each network is a flow (ex- oil products flow in a pipeline, automobile traffic flow in
highways). The maximum flow in a network can be finite or infinite, depending on the capacity of its arcs.
The maximum amount of flow is that the network would allow to flow from source to sink. If there's a
path from the source to the sink of infinite-capacity edges, then in that case, a maximum flow can be
found by pushing infinite flow across that path.
 An arc is said to be directed or oriented if it allows positive flow in one direction only. A directed network
has all directed arcs.
 A path is a set of arcs joining two distinct nodes, passing through other nodes in the network. For
example, in Figure 1, arcs (1, 2), (2, 3), (3, 4), and (4, 5) form a path between nodes 1 and 5. A path forms
a cycle or a loop if it connects a node back to itself through other nodes. In Figure 1, arcs (2, 3), (3, 4), and
(4, 2) form a cycle. A network is said to be connected if every two distinct nodes are linked by at least one
path. The network in Figure 1 demonstrates this type of network.
 A tree is a cycle-free (i.e. loop free) connected network comprised of a subset of all the nodes, and a
spanning tree links all the nodes of the network. Figure 2 provides examples of a tree and a spanning tree
from the network in Figure 1.
 Figure 1: Example of (N, A) Network

Figure 2: Examples of a tree and a spanning tree

 Minimum Spanning Tree Problems
Suppose that each arc (i, j) in a network has a length associated with it and that
arc (i, j) represents a way of connecting node i to node j. For example, if each node
in a network represents a computer at State University, then arc (i, j) might
represent an underground cable that connects computer i with computer j. In
many applications, we want to determine the set of arcs in a network that connect
all nodes such that the sum of the length of the arcs is minimized. Clearly, such a
group of arcs should contain no loop. (A loop is often called a closed path or
cycle.) For example, in Figure 3, the sequence of arcs (1, 2)–(2, 3)–(3, 1) is a loop.

Definition of Spanning Tree:

For a network with n nodes, a spanning tree is a group of n-1 arcs that
connects all nodes of the network and contains no loops.

Note: Theoretically, a Minimal Spanning Tree can be formulated and solved as a linear
program (LP). However, LP is not a practical option because numerous constraints
must be added to exclude all cycles, resulting in a huge LP, even for small networks.
 Figure 3: Description of Loop
and Minimum Spanning Tree

 In Figure 3, there are three spanning trees i.e.

i) Arcs (1, 2) and (2, 3)
ii) Arcs (1, 2) and (1, 3)
iii) Arcs (1, 3) and (2, 3)
 A Spanning Tree of minimum length in a network is a Minimum Spanning Tree
(MST). In Figure 3, the spanning tree consisting of arcs (1, 3) and (2, 3) is the
unique Minimum Spanning Tree.
 Determination of Minimum Spanning Tree Algorithm (Prim’s Algorithm):
Step 1: Begin at any node i, and join node i to the node in the network (call it node j)
that is closest (in terms of shortest distance units, etc.) to node i [The smallest weight
corresponding activity is selected first]. two nodes i and j now form a connected set of
nodes C = {i, j}, and arc (i, j) will be in the minimum spanning tree. The remaining
nodes in the network (call them C’) are referred to as the unconnected set of nodes.
Step 2: Now choose a member of C’ (call it n) that is closest to some node in C. Let m
represent the node in C that is closest to n. Then the arc (m, n) will be in the minimum
spanning tree. Now update C and C’. Because n is now connected to {i, j}, C now
equals {i, j, n} and we must eliminate node n from C’.
Step 3: Repeat this process until a minimum spanning tree is found that satisfies the
definition conditions of spanning tree [say ‘n’ nodes, all ‘n’ nodes are included and
there are ‘n-1’ arcs and no looping]. Ties for closest node and arc to be included in the
minimum spanning tree may be broken arbitrarily.
Note: At each step the algorithm chooses the shortest arc that can be used to expand
C, so the algorithm is often referred to as a “greedy” algorithm. It is remarkable that
the act of being “greedy” at each step of the algorithm can never force us later to
follow a “bad arc.”
Example: The State University campus has five
minicomputers. The distance between each pair of
computers (in city blocks) is given in Figure 4. The
computers must be interconnected by
underground cable. What is the minimum length of
cable required? Note that if no arc is drawn
connecting a pair of nodes, this means that
(because of underground rock formations) no
cable can be laid between these two computers.
Solution:
We find the MST for Figure 4 using MST algorithm.
1st Iteration:
We arbitrarily choose to begin at node 1. The
closest node to node 1 is node 2 (as distance is
minimum). Now C = {1, 2}, C’ = {3, 4, 5}.
Figure 4: Distances between
State University Computers
Iteration 2: Node 5 is closest (two units distance) to C. Since node 5 is two blocks from
node 1 and from node 2, we may include either arc (2, 5) or arc (1, 5) in the minimum
spanning tree. We arbitrarily choose to include arc (2, 5). Then C = {1, 2, 5} and C’ = {3, 4}
(see Figure 4b).
Iteration 3: Node 3 is two blocks from node 5, so we may include arc (5, 3) in the minimum
spanning tree. Now C = {1, 2, 3, 5} and C’ = {4} (see Figure 4c).
Iteration 4: Node 5 is the closest node to node 4, so we add arc (5, 4) to the minimum spanning
tree (see Figure 4d).

We have now obtained the Minimum Spanning Tree consisting of arcs (1, 2), (2, 5), (5, 3),
and (5, 4). The length of the minimum spanning tree is 1+2+2+4 = 9 units.
Note:
 If a tree has n vertices, then it will have n-1 arcs and no looping.
 There can be more than one minimum spanning tree (MST) in a
given network diagram.
 The algorithm stops when all the nodes are in the spanning tree.
 We build tree from the start. So, the tree property is maintained
throughout.