Chapter 06

CHAPTER 6
NETWORK OPTIMIZATION PROBLEMS

SOLUTION TO SOLVED PROBLEMS
6.S1 Distribution at Heart Beats

Heart Beats is a manufacturer of medical equipment. The company’s primary product is a
device used to monitor the heart during medical procedures. This device is produced in two
factories and shipped to two warehouses. The product is then shipped on demand to four
third-‐party wholesalers. All shipping is done by truck. The product distribution network is
shown below. The annual production capacity at Factories 1 and 2 is 400 and 250,
respectively. The annual demand at Wholesalers 1, 2, 3, and 4 is 200, 100, 150, and 200,
respectively. The cost of shipping one unit in each shipping lane is shown on the arcs. Due to
limited truck capacity, at most 250 units can be shipped from Factory 1 to Warehouse 1 each
year. Formulate and solve a network optimization model in a spreadsheet to determine how
to distribute the product at the lowest possible annual cost.

This is a minimum-‐cost flow problem. To set up a spreadsheet model, first list all of the arcs
as shown in B4:C11, along with their capacity (F4) and unit cost (G4:G11). Only the arc
from F1 to WH1 is capacitated. Then list all of the nodes as shown in I4:I11 along with each
node’s supply or demand (L4:L11).

1
B C D E F G H I J K L
3 From To Capacity Unit Cost Nodes Supply/Demand
4 F1 WH1 250 $40 F1 400
5 F1 WH2 $35 F2 250
6 F2 WH2 $25 WH1 0
7 WH1 WS1 $60 WH2 0
8 WH1 WS2 $35 WS1 -200
9 WH2 WS2 $55 WS2 -100
10 WH2 WS3 $50 WS3 -150
11 WH2 WS4 $65 WS4 -200

The changing cells are the amount of flow to send through each arc. These are shown in
Flow (D4:D11) below, with an arbitrary value of 10 entered for each. The flow through the
arc from F1 to WH1 must be less than the capacity of 250, as indicated by the constraint D4
<= F4.

B C D E F G
3 From To Flow Capacity Unit Cost
4 F1 WH1 10 <= 250 $40
5 F1 WH2 10 $35
6 F2 WH2 10 $25
7 WH1 WS1 10 $60
8 WH1 WS2 10 $35
9 WH2 WS2 10 $55
10 WH2 WS3 10 $50
11 WH2 WS4 10 $65
2
For each node, calculate the net flow as a function of the changing cells. This can be done
using the SUMIF function. In each case, the first SUMIF function calculates the flow leaving
the node and the second one calculates the flow entering the node. For example, consider
the F1 node (I4). SUMIF(From, Nodes, Flow) sums each individual entry in Flow (the
changing cells in D4:D11) if that entry is in a row where the entry in From (B4:B11) is the
same as in that row of Nodes (i.e., F1). Since I4 = F1 and the only rows that have F1 in From
(B4:B11) are rows 4 and 5, the sum in the ship column is only over these same rows, so this
sum is D4+D5.

3 From To Flow Capacity Unit Cost Nodes Net Flow Supply/Demand
4 F1 WH1 10 <= 250 $40 F1 20 = 400
5 F1 WH2 10 $35 F2 10 = 250
6 F2 WH2 10 $25 WH1 10 = 0
7 WH1 WS1 10 $60 WH2 10 = 0
8 WH1 WS2 10 $35 WS1 -10 = -200
9 WH2 WS2 10 $55 WS2 -20 = -100
10 WH2 WS3 10 $50 WS3 -10 = -150
11 WH2 WS4 10 $65 WS4 -10 = -200

J
3 Net Flow
4 =SUMIF(From,Nodes,Flow)-SUMIF(To,Nodes,Flow)
The goal is to minimize the total cost of shipping the product from the factories to the
wholesalers. The cost is the SUMPRODUCT of the Unit Costs with the Flow, or Total Cost =
SUMPRODUCT(UnitCost, Flow). This formula is entered into TotalCost (D13).

4 F1 WH1 10 <= 250 $40 F1 20 = 400
5 F1 WH2 10 $35 F2 10 = 250
6 F2 WH2 10 $25 WH1 10 = 0
7 WH1 WS1 10 $60 WH2 10 = 0
8 WH1 WS2 10 $35 WS1 -10 = -200
9 WH2 WS2 10 $55 WS2 -20 = -100
10 WH2 WS3 10 $50 WS3 -10 = -150
11 WH2 WS4 10 $65 WS4 -10 = -200
12
13 Total Cost $3,650

C D
13 Total Cost =SUMPRODUCT(UnitCost,Flow)
3
The Solver information and solved spreadsheet are shown below.

4 F1 WH1 250 <= 250 $40 F1 400 = 400
5 F1 WH2 150 $35 F2 250 = 250
6 F2 WH2 250 $25 WH1 0 = 0
7 WH1 WS1 200 $60 WH2 0 = 0
8 WH1 WS2 50 $35 WS1 -200 = -200
9 WH2 WS2 50 $55 WS2 -100 = -100
10 WH2 WS3 150 $50 WS3 -150 = -150
11 WH2 WS4 200 $65 WS4 -200 = -200
12

Solver Parameters
Set Objective Cell: TotalCost
Range Name Cells
To: Min
Capacity F4
By Changing Variable Cells:
Flow D4:D11
Flow
From B4:B11
Subject to the Constraints:
NetFlow J4:J11
D4 <= Capacity Nodes I4:I11
NetFlow = SupplyDemand SupplyDemand L4:L11
Solver Options: To C4:C11
Make Variables Nonnegative TotalCost D13
Solving Method: Simplex LP UnitCost G4:G11

J
3 Net Flow

C D

Thus, Flow (D4:D11) indicates how to distribute the product so as to achieve the minimum
Total Cost (D13) of $58,500.
4
6.S2 Assessing the Capacity of a Pipeline Network
Exxo 76 is an oil company that operates the pipeline network shown below, where each
pipeline is labeled with its maximum flow rate in million cubic feet (MMcf) per day. A new oil
well has been constructed near A. They would like to transport oil from the well near A to
their refinery at G. Formulate and solve a network optimization model to determine the
maximum flow rate from A to G.
15 D
14
B 6
20 12
4 G
A 2 F
20
8 16
15
C 23
E

This is a minimum-‐cost flow problem. Associated with each pipe in the network will be an
arc (or, for pipes which might flow in either direction, two arcs, one in each direction). To
set up a spreadsheet model, first list all of the arcs as shown in B5:C19, along with their
capacity (F5:F19). Then list all of the nodes as shown in H5:H11. All the transshipment
nodes (every node except the start node A and the end node G) will be constrained to have
net flow = 0 (Supply/Demand = 0). The start node (A) and end node (G) are left
unconstrained. We want to maximize the net flow out of node A.

B C D E F G H I J K
4 From To Capacity Nodes Supply/Demand
5 A B 20 A
6 A C 20 B = 0
7 B C 2 C = 0
8 B D 15 D = 0
9 B F 12 E = 0
10 C B 2 F = 0
11 C E 23 G
12 C F 8
13 D F 6
14 D G 14
15 E F 15
16 E G 16
17 F D 6
18 F E 15
19 F G 4

5
The changing cells are the amount of flow to send through each pipe (arc). These are shown
in Flow (D5:D19) below, with an arbitrary value of 0 entered for each. The flow through
each arc is capacitated as indicated by the <= in E5:E19.

B C D E F
3 thousands of gallons / hour
4 From To Flow Capacity
5 A B 0 <= 20
6 A C 0 <= 20
7 B C 0 <= 2
8 B D 0 <= 15
9 B F 0 <= 12
10 C B 0 <= 2
11 C E 0 <= 23
12 C F 0 <= 8
13 D F 0 <= 6
14 D G 0 <= 14
15 E F 0 <= 15
16 E G 0 <= 16
17 F D 0 <= 6
18 F E 0 <= 15
19 F G 0 <= 4
6
the A node (H5). SUMIF(From, Nodes, Flow) in I5 sums each individual entry in Flow (the
changing cells in D5:D19) if that entry is in a row where the entry in From (B5:B19) is the
same as in the entry in that row of Nodes (i.e., A). Since the only rows that have A in From
(B5:B19) are rows 5 and 6, the sum in the ship column is only over these same rows, so this
sum is D5+D6.

B C D E F G H I J K
3 thousands of gallons / hour thousands of gallons / hour
4 From To Flow Capacity Nodes Net Flow Supply/Demand
5 A B 0 <= 20 A 0
6 A C 0 <= 20 B 0 = 0
7 B C 0 <= 2 C 0 = 0
8 B D 0 <= 15 D 0 = 0
9 B F 0 <= 12 E 0 = 0
10 C B 0 <= 2 F 0 = 0
11 C E 0 <= 23 G 0
12 C F 0 <= 8
13 D F 0 <= 6
14 D G 0 <= 14
15 E F 0 <= 15
16 E G 0 <= 16
17 F D 0 <= 6
18 F E 0 <= 15
19 F G 0 <= 4

I
4 Net Flow
7
The goal is to maximize the amount shipped from A to G. Since nodes B through F are
transshipment nodes (net flow = 0), any amount that leaves A must enter G. Thus,
maximizing the flow out of A will achieve our goal. Thus, the formula entered into the
objective cell MaximumFlow (D21) is =I5.

B C D E F G H I J K
5 A B 0 <= 20 A 0
6 A C 0 <= 20 B 0 = 0
7 B C 0 <= 2 C 0 = 0
8 B D 0 <= 15 D 0 = 0
9 B F 0 <= 12 E 0 = 0
10 C B 0 <= 2 F 0 = 0
11 C E 0 <= 23 G 0
12 C F 0 <= 8
13 D F 0 <= 6
14 D G 0 <= 14
15 E F 0 <= 15
16 E G 0 <= 16
17 F D 0 <= 6
18 F E 0 <= 15
19 F G 0 <= 4
20
21 Maximum Flow 0

C D
21 Maximum Flow =I5
8

B C D E F G H I J K
5 A B 20 <= 20 A 34
6 A C 14 <= 20 B 0 = 0
7 B C 0 <= 2 C 0 = 0
8 B D 14 <= 15 D 0 = 0
9 B F 8 <= 12 E 0 = 0
10 C B 2 <= 2 F 0 = 0
11 C E 12 <= 23 G -34
12 C F 0 <= 8
13 D F 0 <= 6
14 D G 14 <= 14
15 E F 0 <= 15
16 E G 16 <= 16
17 F D 0 <= 6
18 F E 4 <= 15
19 F G 4 <= 4
20
21 Maximum Flow 34

Solver Parameters
Set Objective Cell: MaximumFlow
To: Max Range Name Cells
By Changing Variable Cells: Capacity F5:F19
Flow Flow D5:D19
Subject to the Constraints: From B5:B19
Flow <= Capacity MaximumFlow D21
NetFlow = SupplyDemand NetFlow I6:I10
Solver Options: Nodes H5:H11
Make Variables Nonnegative SupplyDemand K6:K10
Solving Method: Simplex LP To C5:C19

I
4 Net Flow
10 =SUMIF(From,Nodes,Flow)-SUMIF(To,Nodes,Flow) C D
11 =SUMIF(From,Nodes,Flow)-SUMIF(To,Nodes,Flow) 21 Maximum Flow =I5

Thus, Flow (D5:D19) indicates how to send oil through the network so as to achieve the
Maximum Flow (D21) of 34 thousand gallons/hour.
9
6.S3 Driving to the Mile-High City
Sarah and Jennifer have just graduated from college at the University of Washington in
Seattle and want to go on a road trip. They have always wanted to see the mile-‐high city of
Denver. Their road atlas shows the driving time (in hours) between various city pairs, as
shown below. Formulate and solve a network optimization model to find the quickest route
from Seattle to Denver?
Seattle
Butte
Billings
10
3 4
9 7
Portland 7 7
Boise 12 7
6 14
Cheyenne
7
Salt Lake 1
5
City
4
Grand Denver
Junction
This is a shortest path problem. To set up a spreadsheet model, first list all of the arcs as
shown in B4:C11, along with their capacity (F4). Only the arc from F1 to WH1 is
capacitated. Then list all of the nodes as shown in I4:I11 along with each node’s supply or
demand (L4:L11).

3 From To Capacity Unit Cost Nodes Supply/Demand
4 F1 WH1 250 $40 F1 400
5 F1 WH2 $35 F2 250
6 F2 WH2 $25 WH1 0
7 WH1 WS1 $60 WH2 0
8 WH1 WS2 $35 WS1 -200
9 WH2 WS2 $55 WS2 -100
10 WH2 WS3 $50 WS3 -150
11 WH2 WS4 $65 WS4 -200

The changing cells are the amount of flow to send through each arc. These are shown in
Flow (D4:D11) below, with an arbitrary value of 10 entered for each. The flow through the
arc from F1 to WH1 must be less than the capacity of 250, as indicated by the constraint D4
<= F4.
10

B C D E F G
3 From To Flow Capacity Unit Cost
4 F1 WH1 10 <= 250 $40
5 F1 WH2 10 $35
6 F2 WH2 10 $25
7 WH1 WS1 10 $60
8 WH1 WS2 10 $35
9 WH2 WS2 10 $55
10 WH2 WS3 10 $50
11 WH2 WS4 10 $65
the F1 node (I4). SUMIF(From, I4, Flow) sums each individual entry in Flow (the changing
cells in D4:D11) if that entry is in a row where the entry in From (B4:B11) is the same as in
I4 (i.e., F1). Since I4 = F1 and the only rows that have F1 in From (B4:B11) are rows 4 and
5, the sum in the ship column is only over these same rows, so this sum is D4+D5.

4 F1 WH1 10 <= 250 $40 F1 20 = 400
5 F1 WH2 10 $35 F2 10 = 250
6 F2 WH2 10 $25 WH1 10 = 0
7 WH1 WS1 10 $60 WH2 10 = 0
8 WH1 WS2 10 $35 WS1 -10 = -200
9 WH2 WS2 10 $55 WS2 -20 = -100
10 WH2 WS3 10 $50 WS3 -10 = -150
11 WH2 WS4 10 $65 WS4 -10 = -200

J
3 Net Flow
11
The goal is to minimize the total cost of shipping the product from the factories to the
wholesalers. The cost is the SUMPRODUCT of the Unit Costs with the Flow, or Total Cost =
SUMPRODUCT(UnitCost, Flow). This formula is entered into TotalCost (D13).

4 F1 WH1 10 <= 250 $40 F1 20 = 400
5 F1 WH2 10 $35 F2 10 = 250
6 F2 WH2 10 $25 WH1 10 = 0
7 WH1 WS1 10 $60 WH2 10 = 0
8 WH1 WS2 10 $35 WS1 -10 = -200
9 WH2 WS2 10 $55 WS2 -20 = -100
10 WH2 WS3 10 $50 WS3 -10 = -150
11 WH2 WS4 10 $65 WS4 -10 = -200
12

C D
12

4 F1 WH1 250 <= 250 $40 F1 400 = 400
5 F1 WH2 150 $35 F2 250 = 250
6 F2 WH2 250 $25 WH1 0 = 0
7 WH1 WS1 200 $60 WH2 0 = 0
8 WH1 WS2 50 $35 WS1 -200 = -200
9 WH2 WS2 50 $55 WS2 -100 = -100
10 WH2 WS3 150 $50 WS3 -150 = -150
11 WH2 WS4 200 $65 WS4 -200 = -200
12

Solver Parameters
Set Objective Cell: Total Cost Range Name Cells
To: Min Capacity F4
By Changing Variable Cells: Flow D4:D11
Flow From B4:B11
Subject to the Constraints: NetFlow J4:J11
D4 <= Capacity Nodes I4:I11
NetFlow = SupplyDemand SupplyDemand L4:L11
Solver Options: To C4:C11
Make Variables Nonnegative TotalCost D13
Solving Method: Simplex LP UnitCost G4:G11

J
3 Net Flow

C D

Thus, Flow (D4:D11) indicates how to distribute the product so as to achieve the minimum
Total Cost (D13) of $58,500.

13

Chapter 06

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 06

Uploaded by

Copyright:

Available Formats

CHAPTER 6

NETWORK OPTIMIZATION PROBLEMS

6.S1 Distribution at Heart Beats

You might also like