lOMoARcPSD|23416644

Chapter 6
Distribution and Network Models

Learning Objectives

1. Understand the usefulness of using optimization for supply chain problems.

2. Be able to identify the special features of the transportation problem.

3. Become familiar with the types of problems that can be solved by applying a transportation model.

4. Be able to develop network and linear programming models of the transportation problem.

5. Know how to handle the cases of (1) unequal supply and demand, (2) unacceptable routes, and (3)
maximization objective for a transportation problem.

6. Be able to identify the special features of the assignment problem.

7. Become familiar with the types of problems that can be solved by applying an assignment model.

8. Be able to develop network and linear programming models of the assignment problem.

9. Be familiar with the special features of the transshipment problem.

10. Become familiar with the types of problems that can be solved by applying a transshipment model.

11 Be able to develop network and linear programming models of the transshipment problem.

12. Know the basic characteristics of the shortest route problem.

13. Be able to develop a linear programming model and solve the shortest route problem.

14. Know the basic characteristics of the maximal flow problem.

15. Be able to develop a linear programming model and solve the maximal flow problem.

16. Know how to structure and solve a production and inventory problem as a transshipment problem.

17. Understand the following terms:

network flow problem capacitated transshipment problem

transportation problem shortest route
origin maximal flow
destination source node
capacitated transportation problem sink node
assignment problem arc flow capacities
transshipment problem supply chain
Solutions:

6-1

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

1. The network model is shown.

Atlanta 1400

5000 Phila. 6

6
Dallas 3200
2

1 Columbus 2000
2
New 5
3000 Orleans
7

Boston 1400

2. a.
Let x11 : Amount shipped from Jefferson City to Des Moines
x12 : Amount shipped from Jefferson City to Kansas City
•
•
•
Min 14x11 + 9x12 + 7x13 + 8x21 + 10x22 + 5x23
s.t.
x11 + x12 + x13  30
x21 + x22 + x23  20
x11 + x21 = 25
x12 + x22 = 15
x13 + x23 = 10

x11, x12, x13, x21, x22, x23,  0

b. Optimal Solution:
Amount Cost
Jefferson City - Des Moines 5 70
Jefferson City - Kansas City 15 135
Jefferson City - St. Louis 10 70
Omaha - Des Moines 20 160
Total 435
3. a.

6-2

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

Hamilton 400
1
10
Southern
500 20
1
15
Butler 200
12 2
15
Northwest
400 2 18

Clermont
300
3

b. Let xij = amount shipped from supply node i to demand node j.

Min 10x11 + 20x12 + 15x13 + 12x21 + 15x22 + 18x23

s.t.
x11 + x12 + x13  500
x21 + x22 + x23  400
x11 + x21 = 400
x12 + x22 = 200
x13 + x23 = 300

xij  0 for all i, j

c. Optimal Solution
Amount Cost
Southern - Hamilton 200 $ 2000
Southern - Clermont 300 4500
Northwest - Hamilton 200 2400
Northwest - Butler 200 3000
Total Cost $11,900

d. To answer this question the simplest approach is to increase the Butler County demand to 300 and
to increase the supply by 100 at both Southern Gas and Northwest Gas. The new optimal solution
is:

Amount Cost
Southern - Hamilton 300 $ 3000
Southern - Clermont 300 4500
Northwest - Hamilton 100 1200
Northwest - Butler 300 4500
Total Cost $13,200

From the new solution we see that Tri-County should contract with Southern Gas for the additional
100 units.

6-3

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

4.
a. The optimization model can be written as

x ij = Red GloFish shipped from i to j i = M for Michigan, T for Texas; j = 1, 2, 3.

y ij = Blue GloFish shipped from i to j , i = M for Michigan, T for Texas; j = 1, 2, 3.

Minimize x M1 + 2.50x M2 + 0.50x M3 + y M1 + 2.50y M2 + 0.50y M3 + 2.00y T1 + 1.50y T2 + 2.80y T3

subject to
x M1 + x M2 + x M3 ≤ 1,000,000
y M1 + y M2 + y M3 ≤ 1,000,000
y T1 +y T2 + y T3 ≤ 600,000
x M1 ≥ 320,000
x M2 ≥ 300,000
x M3 ≥ 160,000
y M1 + y T1 ≥ 380,000
y M2 + y T2 ≥ 450,000
y M3 + y T3 ≥ 290,000
x ij ≥ 0
Solving this linear program we find that we should produce 780,000 red GloFish in Michigan, 670,000
blue GloFish in Michigan, and 450,000 blue GloFish in Texas.
Using the notation in the model, the number of GloFish shipped from each farm to each retailer
can be expressed as:

xM1 = 320,000
xM2 = 300,000
xM3 = 160,000
yM1 = 380,000
yM2 = 0
yM3 = 290,000
yT1 = 0
yT2 = 450,000
yT3 = 0

b. The minimum transportation cost is $2.35 million.

c. We have to add variables xT1, xT2, and xT3 for Red GloFish shipped between Texas and Retailers 1,
2 and 3. The revised objective function is

Minimize x M1 + 2.50x M2 + 0.50x M3 + y M1 + 2.50y M2 + 0.50y M3 + 2.00y T1 + 1.50y T2 + 2.80y T3

+ x T1 + 2.50x T2 + 0.50x T3
We replace the third constraint above with

6-4

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

x T1 +x T2 + x T3 + y T1 +y T2 + y T3 ≤ 600,000

And we change the constraints

xM1 ≥ 320,000
xM2 ≥ 300,000
xM3 ≥ 160,000

to

xM1 + xT1 ≥ 320,000

xM2 + xT2 ≥ 300,000
xM3 + xT3 ≥ 160,000

Using this new objective function and constraint the optimal solution is $2.2 million, so the savings are
$150,000.

5. a.

A 180
1
100

Avery 125
160
1
115
100
B 75
120 2
135
Baker
160 115
2

120
C 100
3
155
150

140
Campbell
140
3
130
D
85
4

b. Let xij = number of hours from consultant i assigned to client j

6-5

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

Max 100x11 + 125x12 + 115x13 + 100x14 + 120x21 + 135x22 + 115x23

+ 120x24 + 155x31 + 150x32 + 140x33 + 130x34
s.t. x11 + x12 + x13 + x14  160
x21 + x22 + x23 + x24  160
x31 + x32 + x33 + x34  140
x11 + x21 + x31 = 180
x12 + x22 + x32 = 75
x13 + x23 + x33 = 100
x14 + x24 + x34 = 85
xij  0 for all i, j

Optimal Solution
Hours
Assigned Billing
Avery - Client B 40 $ 5,000
Avery - Client C 100 11,500
Baker - Client A 40 4,800
Baker - Client B 35 4,725
Baker - Client D 85 10,200
Campbell - Client A 140 21,700
Total Billing $57,925

c. New Optimal Solution

Hours
Assigned Billing
Avery - Client A 40 $ 4,000
Avery - Client C 100 11,500
Baker - Client B 75 10,125
Baker - Client D 85 10,200
Campbell - Client A 140 21,700
Total Billing $57,525

6. The network model, the linear programming formulation, and the optimal solution are shown.
Note that the third constraint corresponds to the dummy origin. The variables x31, x32, x33, and
x34 are the amounts shipped out of the dummy origin; they do not appear in the objective function
since they are given a coefficient of zero.

6-6

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

Demand

D1 2000
Supply 32

34
5000 C.S.

32
40
D2 5000
34
30

3000 D. 28

38
D3 3000
0
0

0
4000 Dum

D4 2000

Note: Dummy origin has supply of 4000.

Max 32x11 + 34x12 + 32x13 + 40x14 + 34x21 + 30x22 + 28x23 + 38x24

s.t.
x11 + x12 + x13 + x14  5000

x21 + x22 + x23 + x24  3000

x31 + x32 + x33 + x34  4000 Dummy
x11 + x21 + x31 = 2000
x12 + x22 + x32 = 5000
x13 + x23 + x33 = 3000
x14 + x24 + x34 = 2000
xij  0 for all i, j

6-7

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

Optimal Solution Units Cost

Clifton Springs - D2 4000 $136,000

Clifton Springs - D4 1000 40,000
Danville - D1 2000 68,000
Danville - D4 1000 38,000
Total Cost: $282,000

Customer 2 demand has a shortfall of 1000

Customer 3 demand of 3000 is not satisfied.

7.
a. Let xij = MW produced at plant i for city j; i = L for Los Angeles, T for Tulsa, S for Seattle, j = 1, …,
10.

Min 356.26xL1 + 356.25xL2 + 178.13xL3 + 356.25xL4 + 237.50xL5 + 415.63xL6 + 356.25xL7 + 356.25xL8 +

178.13xL9 + 356.25xL10 + 593.75xT1 + 593.75xT2 + 475.00xT3 + 475.00xT4 + 475.00xT5 + 415.63xT6 +
415.63xT7 + 356.25xT8 + 475.00xT9 + 296.88xT10 + 59.38xS1 + 178.13xS2 + 298.88xS3 + 298.88xS4 +
356.25xS5 + 296.88xS6 + 356.25xS7 + 475.00xS8 + 593.75xS9 + 593.75xS10

subject to
xL1 + xT1 + xS1 ≥ 950.00
xL2 + xT2 + xS2 ≥ 831.25
xL3 + xT3 + xS3 ≥ 2375.00
xL4 + xT4 + xS4 ≥ 593.75
xL5 + xT5 + xS5 ≥ 950.00
xL6 + xT6 + xS6 ≥ 593.75
xL7 + xT7 + xS7 ≥ 1187.50
xL8 + xT8 + xS8 ≥ 712.50
xL9 + xT9 + xS9 ≥ 1187.50
xL10 + xT10 + xS10 ≥ 1543.75
xij ≥ 0, i = L, T, S j = 1,…, 10

By solving this linear program, we find the optimal solution is to produce 6412.50 MWs in Los Angeles,
1543.75 MWs in Tulsa, and 2968.75 MWs in Seattle. The total distribution cost of this solution is
$2,552,382.81.

b. We must add the following constraints to the linear program shown above:

xL1 + xL2 + xL3 + xL4 + xL5 + xL6 + xL7 + xL8 + xL9 + xL10 ≤ 4000
xT1 + xT2 + xT3 + xT4 + xT5 + xT6 + xT7 + xT8 + xT9 + xT10 ≤ 4000
xS1 + xS2 + xS3 + xS4 + xS5 + xS6 + xS7 + xS8 + xS9 + xS10 ≤ 4000

Solving the LP with these added constraints yields an optimal production of 4000 MWs in Los Angeles,
2925 MWs in Tulsa, and 4000 MWs in Seattle at a total cost of $2,652,949.22. Therefore, the increase in
cost associated with the additional constraints is $2,652,949.22 - $2,552,382.81 = $100,566.41.

8. a.

6-8

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

1
Boston 50
7
1 11
100
Denver
8
13 2
70
20 Dallas
17
2
100 Atlanta 12

10
3
Los 60
8 Angeles
18
3 13
150
Chicago
16
4
St. Paul 80

b. There are alternative optimal solutions.

Solution #1 Solution # 2

Denver to St. Paul: 10 Denver to St. Paul: 10

Atlanta to Boston: 50 Atlanta to Boston: 50
Atlanta to Dallas: 50 Atlanta to Los Angeles: 50
Chicago to Dallas: 20 Chicago to Dallas: 70
Chicago to Los Angeles: 60 Chicago to Los Angeles: 10
Chicago to St. Paul: 70 Chicago to St. Paul: 70

Total Profit: $4240

If solution #1 is used, Forbelt should produce 10 motors at Denver, 100 motors at Atlanta, and 150
motors at Chicago. There will be idle capacity for 90 motors at Denver.

If solution #2 is used, Forbelt should adopt the same production schedule but a modified shipping
schedule.

6-9

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

9. The linear programming formulation and optimal solution are shown.

Let x1A = Units of product A on machine 1

x1B = Units of product B on machine 1



x3C = Units of product C on machine 3

Min x1A + 1.2x1B + 0.9x1C + 1.3x2A + 1.4x2B + 1.2x2C + 1.1x3A + x3B + 1.2x3C
s.t.
x1A + x1B + x1C  1500

x2A + x2B + x2C  1500

x3A + x3B + x3C  1000
x1A + x2A + x3A = 2000
x1B + x2B + x3B = 500
x1C + x2C + x3C = 1200
xij  0 for all i, j

Optimal Solution Units Cost

1-A 300 $ 300

1-C 1200 1080
2-A 1200 1560
3-A 500 550
3-B 500 500
Total: $3990

Note: There is an unused capacity of 300 units on machine 2.

10. a. The total cost is the sum of the purchase cost and the transportation cost. We show the calculation
for Division 1 - Supplier 1 and present the result for the other Division-Supplier combinations.

Division 1 - Supplier 1
Purchase cost (40,000 x $12.60) $504,000
Transportation Cost (40,000 x $2.75) 110,000
Total Cost: $614,000

6 - 10

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

Cost Matrix ($1,000 s)

Supplier

1 2 3 4 5 6

1 614 660 534 680 590 630

2 603 639 702 693 693 630

Division
3 865 830 775 850 900 930

4 532 553 511 581 595 553

5 720 648 684 693 657 747

b. Optimal Solution:

Supplier 1 - Division 2 $ 603

Supplier 2 - Division 5 648
Supplier 3 - Division 3 775
Supplier 5 - Division 1 590
Supplier 6 - Division 4 553
Total $3,169

11. a. Network Model

Demand

6
300
C1
Supply
1 4
450
P1 6
7 4
4 7
8 300
W1 C2
8 4
2
600
P2 5 3
6
5 7 8
300
W2 C3
5
7
3 6
380
P3

9
400
C4

6 - 11

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

b. & c.
The linear programming formulation and solution is shown below.

LINEAR PROGRAMMING PROBLEM

MIN 4X14 + 7X15 + 8X24 + 5X25 + 5X34 + 6X35 + 6X46 + 4X47 + 8X48 + 4X49 + 3X56 + 6X57 +
7X58 + 7X59

S.T.

1) X14 + X15 < 450

2) X24 + X25 < 600
3) X34 + X35 < 380
4) X46 + X47 + X48 + X49 - X14 - X24 - X34 = 0
5) X56 + X57 + X58 + X59 - X15 - X25 - X35 = 0
6) X46 + X56 = 300
7) X47 + X57 = 300
8) X48 + X58 = 300
9) X49 + X59 = 400

OPTIMAL SOLUTION

Objective Function Value = 11850.000

Variable Value Reduced Costs

-------------- --------------- ------------------
X14 450.000 0.000
X15 0.000 3.000
X24 0.000 3.000
X25 600.000 0.000
X34 250.000 0.000
X35 0.000 1.000
X46 0.000 3.000
X47 300.000 0.000
X48 0.000 1.000
X49 400.000 0.000
X56 300.000 0.000
X57 0.000 2.000
X58 300.000 0.000
X59 0.000 3.000

There is an excess capacity of 130 units at plant 3.

6 - 12

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

12. a. Three arcs must be added to the network model in problem 23a. The new network is shown.

Demand

6
300
C1
Supply
1 4
450
P1 6
7 4
4 7
8 300
W1 C2
8 4
2
600 2 2 3
P2 5
6
5 7 8
300
W2 C3
5
7
3 6
380
P3
7 9
400
C4

b.&c.
The linear programming formulation and optimal solution is shown below.

LINEAR PROGRAMMING PROBLEM

MIN 4X14 + 7X15 + 8X24 + 5X25 + 5X34 + 6X35 + 6X46 + 4X47 + 8X48 + 4X49 + 3X56 + 6X57 +
7X58 + 7X59 + 7X39 + 2X45 + 2X54

S.T.

1) X14 + X15 < 450

2) X24 + X25 < 600
3) X34 + X35 + X39 < 380
4) X45 + X46 + X47 + X48 + X49 - X14 - X24 - X34 - X54 = 0
5) X54 + X56 + X57 + X58 + X59 - X15 - X25 - X35 - X45 = 0
6) X46 + X56 = 300
7) X47 + X57 = 300
8) X48 + X58 = 300
9) X39 + X49 + X59 = 400

6 - 13

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

OPTIMAL SOLUTION

Objective Function Value = 11220.000

Variable Value Reduced Costs

-------------- --------------- ------------------
X14 320.000 0.000
X15 0.000 2.000
X24 0.000 4.000
X25 600.000 0.000
X34 0.000 2.000
X35 0.000 2.000
X46 0.000 2.000
X47 300.000 0.000
X48 0.000 0.000
X49 20.000 0.000
X56 300.000 0.000
X57 0.000 3.000
X58 300.000 0.000
X59 0.000 4.000
X39 380.000 0.000
X45 0.000 1.000
X54 0.000 3.000

The value of the solution here is $630 less than the value of the solution for problem 23. The new
shipping route from plant 3 to customer 4 has helped (x39 = 380). There is now excess capacity
of 130 units at plant 1.

6 - 14

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

13. a. Network Model

Supply DC Capacity Demand

500

25 4 30 8
1
Iowa Just 200
Detroit 25 27.5
350 Sports
35 500 30
40 20
5
32.5
35 45 Maryland
2 6 9
350 40 500
LA Sports
35 500 Stuff
35
42.5 6 40
40 Idaho
32.5
40
500 25
3 42.5 27.5 10
700 Austin Sports 650
7 Dude
32.5 Ark 42.5

b. Let = units shipped from node i to node j

Min 25x1,4 + 25x1,5 + 35x1,6 + 40x1,7 + 35x2,4 + 45x2,5 + 35x2,6 + 42.5x2,7 + 40x3,4 + 40x3,5 + 42.5x3,6 + 32.5x3,7
+ 30x4,8 + 27.5x4,9 + 30x4,10 + 20x5,8 + 32.5x5,9 + 40x5,10 + 35x6,8 + 40x6,9 + 32.5x6,10 + 27.5x7,8 + 25x7,9 +
42.5x 7,10

subject to
x1,4 + x1,5 + x1,6 + x1,7 ≤ 350
x2,4 + x2,5 + x2,6 + x2,7 ≤ 350
x3,4 + x3,5 + x3,6 + x3,7 ≤ 700
x1,4 + x2,4 + x3,4 = x4,8 + x4,9 + x4,10
x1,5 + x2,5 + x3,5 = x5,8 + x5,9 + x5,10
x1,6 + x2,6 + x3,6 = x6,8 + x6,9 + x6,10
x1,7 + x2,7 + x3,7 = x7,8 + x7,9 + x7,10
x1,4 + x2,4 + x3,4 ≤ 500
x1,5 + x2,5 + x3,5 ≤ 500
x1,6 + x2,6 + x3,6 ≤ 500
x1,7 + x2,7 + x3,7 ≤ 500
x4,8 + x5,8 + x6,8 + x7,8 ≥ 200
x4,9 + x5,9 + x6,9 + x7,9 ≥ 500
x4,10 + x5,10 + x6,10 + x7,10 ≥ 650
xij ≥ 0 for all i and j

Solving this formulation we get an optimal cost of $79,625 per week.

6 - 15

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

c. Replace the seventh constraint above with x1,4 + x2,4 + x3,4 ≤ 800. The optimal cost associated with
this increased capacity at the Iowa distribution center is $79,250 per week. Therefore, the savings
over 50 weeks is ($79,625 - $79,250) * 50 = $18,750 per year. The $40,000 cost of expansion
exceeds the potential savings, so Sports of All Sorts should not expand.

14.
6
Macon 2

1
3 Muncie 8
44
6 34
4 7
Louisville Greenwood 4
34
3 32
2
6 Brazil
8 57
35
5 8
28
Cincinnati Concord 3
9 24
3 3
5 Xenia

9
Chatham 3

6 - 16

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

A linear programming model is

Min
8x14 + 6x15 + 3x24 + 8x25 + 9x34 + 3x35 + 44x46 + 34x47 + 34x48 + 32x49 + 57x56 + 35x57 + 28x58 + 24x59
s.t.
x14 + x15 3

x24 + x25 6

x34 + x35 5

-x 14 - x24 - x34 + x46 + x47 + x48 + x49 = 0

- x15 - x25 - x35 + x56 + x57 + x58 + x59 = 0
x46 + x56 = 2
x47 + x57 = 4
x48 + x58 = 3
x49 + x59 = 3
xij  0 for all i, j

Optimal Solution Units Shipped Cost

Muncie to Cincinnati 1 6
Cincinnati to Concord 3 84
Brazil to Louisville 6 18
Louisville to Macon 2 88
Louisville to Greenwood 4 136
Xenia to Cincinnati 5 15
Cincinnati to Chatham 3 72
419

Two rail cars must be held at Muncie until a buyer is found.

15.
+3

2
10 12

-6

11
+5 1 3 5 -4

11 8
12 9

+2

6 - 17

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

The positive numbers by nodes indicate the amount of supply at that node. The negative numbers
by nodes indicate the amount of demand at the node.

16. a.
Min 20x12 + 25x15 + 30x25 + 45x27 + 20x31 + 35x36
+ 30x42 + 25x53 + 15x54 + 28x56 + 12x67 + 27x74
s.t.
x31 - x12 - x15 = 8
x25 + x27 - x12 - x42 = 5
x31 + x36 - x53 = 3
x54 + x74 - x42 = 3
x53 + x54 + x56 - x15 - x25 = 2
x36 + x56 - x67 = 5
x74 - x27 - x67 = 6
xij  0 for all i, j

b. x12 = 0 x53 = 5
x15 = 0 x54 = 0
x25 = 8 x56 = 5
x27 = 0 x67 = 0
x31 = 8 x74 = 6
x36 = 0 x56 = 5
x42 = 3

Total cost of redistributing cars = $917

6 - 18

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

17. a.

10
1 1
1 16 Client 1 1
Jackson

32

14

2 22 2
1 1
Ellis Client 2
40

22
24
3 3
1 34 1
Smith Client 3

b.
Min 10x11 + 16x12 + 32x13 + 14x21 + 22x22 + 40x23 + 22x31 + 24x32 + 34x33
s.t.
x11 + x12 + x13 1
x21 + x22 + x23 1
x31 + x32 + x33  1
x11 x
+ 21 + x31 = 1
x12 + x22 + x32 = 1
x13 + x23 + x33 = 1

xij  0 for all i, j

Solution x12 = 1, x21 = 1, x33 = 1 Total completion time = 64

6 - 19

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

18. a.

Crews Jobs
30
1 44
1 25 1 1
Red 38
47
31

2
1 2 1
White

3
1 3 1
Blue

4
1 4 1
Green
34
26 44
43
5
1 5 1
Brown 28

6 - 20

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

b.
Min 30x11 + 44x12 + 38x13 + 47x14 + 31x15 + 25x21 +  + 28x55
s.t.
x11 + x12 + x13 + x14 + x15  1
x21 + x22 + x23 + x24 + x25  1
x31 + x32 + x33 + x34 + x35  1
x41 + x42 + x43 + x44 + x45  1
x51 + x52 + x53 + x54 + x55  1
x11 + x21 + x31 + x41 + x51 = 1
x12 + x22 + x32 + x42 + x52 = 1
x13 + x23 + x33 + x43 + x53 = 1
x14 + x24 + x34 + x44 + x54 = 1
x15 + x25 + x35 + x45 + x55 = 1

xij  0, i = 1, 2,.., 5; j = 1, 2,.., 5

Optimal Solution:

Green to Job 1 $26

Brown to Job 34
2
Red to Job 3 38
Blue to Job 4 39
White to Job 5 25
$162

Since the data is in hundreds of dollars, the total installation cost for the 5 contracts is $16,200.

19. This can be formulated as a linear program with a maximization objective function. There are 24
variables, one for each program/time slot combination. There are 10 constraints, 6 for the potential
programs and 4 for the time slots.

Optimal Solution:

NASCAR Live 5:00 – 5:30 p.m.

Hollywood Briefings 5:30 – 6:00 p.m.
World News 7:00 – 7:30 p.m.
Ramundo & Son 8:00 – 8:30 p.m.

Total expected advertising revenue = $30,500

20. a. This is the variation of the assignment problem in which multiple assignments are possible. Each
distribution center may be assigned up to 3 customer zones.

The linear programming model of this problem has 40 variables (one for each combination of
distribution center and customer zone). It has 13 constraints. There are 5 supply ( 3) constraints
and 8 demand (= 1) constraints.

The optimal solution follows.

6 - 21

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

Assignments Cost
($1000s)
Plano: Kansas City, Dallas 34
Flagstaff: Los Angeles 15
Springfield: Chicago, Columbus, 70
Atlanta
Boulder: Newark, Denver 97
Total Cost - $216

b. The Nashville distribution center is not used.

c. All the distribution centers are used. Columbus is switched from Springfield to Nashville. Total
cost increases by $11,000 to $227,000.

21. A linear programming formulation and the optimal solution are given. For the decision variables,
xij, we let the first subscript correspond to the supplier and the second subscript correspond to the
distribution hub. Thus, xij = 1 if supplier i is awarded the bid to supply hub j and xij = 0 if supplier i
is not awarded the bid to supply hub j.

Min 190x11 +175x12 + 125x13 + 230x14 + 150x21 + 235x22 + 155x23 + 220x24 + 210x31 + 225x32
+ 135x33 +260x34 + 170x41 + 185x42 + 190x43 + 280x44 + 220x51 + 190x52 + 140x53
+ 240x54 + 270x61 + 200x62 + 130x63 + 260x64
s.t.

x11 + x12 + x13 + x14 ≤ 1

+ x21 + x22 + x23 + x24 ≤ 1
x31 + x32 + x33 + x34 ≤ 1
x41 + x42 + x43 + x44 ≤ 1
x51 + x52 + x53 + x54 ≤ 1
x61 + x62 + x63 + x64 ≤ 1
x11 + x21 + x31 + x41 + x51 + x61 = 1
x12 + x22 + x32 + x42 + x52 + x62 = 1
x13 + x23 + x33 + x43 + x53 + x63 = 1
x14 + x24 + x34 + x44 + x45 + x46 = 1
xij  0 for all i, j

Optimal Solution Bid

Martin – Hub 2 175

Schmidt Materials – Hub 4 220
D&J Burns – Hub 1 170
Lawler Depot – Hub 3 130
695

6 - 22

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

22. A linear programming formulation of this problem can be developed as follows. Let the first letter
of each variable name represent the professor and the second two the course. Note that a DPH
variable is not created because the assignment is unacceptable.

Max 2.8AUG + 2.2AMB + 3.3AMS + 3.0APH + 3.2BUG + + 2.5DMS

···
s.t.
AUG + AMB + AMS + APH  1
BUG + BMB + BMS + BPH  1
CUG + CMB + CMS + CPH  1
DUG + DMB + DMS  1
AUG + BUG + CUG + DUG = 1
AMB + BMB + CMB + DMB = 1
AMS + BMS + CMS + DMS = 1
APH + BPH + CPH = 1
All Variables  0

Optimal Solution: Rating

A to MS course 3.3
B to Ph.D. course 3.6
C to MBA course 3.2
D to Undergraduate course 3.2
Max Total Rating 13.3

23. Origin – Node 1

Transshipment Nodes 2 to 5

Destination – Node 7

The linear program will have 14 variables for the arcs and 7 constraints for the nodes.

1 if the arc from node i to node j is on the shortest route


Let xij 
0 otherwise


Min 7 x12  9 x13  18 x14  3x23  5 x25  3x32  4 x35  3x46  5 x52  4 x53  2 x56
6 x57  2 x65  3x67
s.t.
Flow Out Flow In
Node 1 x12  x13  x14 = 1
Node 2 x23  x25  x12  x32  x52 = 0
Node 3 x32  x35  x13  x23  x53 = 0
Node 4 x46  x14 = 0
Node 5 x52  x53  x56  x57  x25  x35  x65 = 0
Node 6 x65  x67  x46  x56 = 0
Node 7  x57  x67 = 1

6 - 23

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

xij > 0 for all i and j

Optimal Solution: x12 1 , x25 1 , x56 1 , and x67 1

Shortest Route 1-2-5-6-7

Length = 17

24. The linear program has 13 variables for the arcs and 6 constraints for the nodes. Use same six
constraints for the Gorman shortest route problem as shown in the text. The objective function
changes to travel time as follows.

Min 40 x12 + 36 x13 + 6 x23 + 6 x32 + 12 x24 + 12 x42 + 25 x26 + 15 x35 + 15 x53

+ 8 x45 + 8 x54 + 11 x46 + 23 x56

Optimal Solution: x12 1 , x24 1 , and x46 1

Shortest Route 1-2-4-6

Total Time = 63 minutes

25. a. Origin – Node 1

Transshipment Nodes 2 to 5

Destination – Node 6

The linear program will have 13 variables for the arcs and 6 constraints for the nodes.

1 if the arc from node i to node j is on the shortest route


Let xij 
0 otherwise


Min 35x12  30 x13  12 x23  18 x24  39 x26  12 x32  15 x35 18 x42 12 x45
16x46 15x53 12x54 30x56

s.t.
Flow Out Flow In
Node 1 x12  x13 =1
Node 2 x23  x24  x26  x12  x32  x42 =0
Node 3 x32  x35  x13  x23  x53 =0
Node 4 x42  x45  x46  x24  x54 =0
Node 5  x53  x54  x56  x35  x45 =0
Node 6  x26  x46  x56 =1

xij > 0 for all I and j

b. Optimal Solution: x12 1 , x24 1 , and x46 1

Shortest Route 1-2-4-6

6 - 24

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

Total time = 69 minutes

c. Allowing 8 minutes to get to node 1 and 69 minutes to go from node 1 to node 6, we expect to
make the delivery in 77 minutes. With a 20% safety margin, we can guarantee a delivery in
1.2(77) = 92 minutes. It is 1:00 p.m. now. Guarantee delivery by 2:32 p.m.

26. Origin – Node 1

Transshipment Nodes 2 to 5 and node 7

Destination – Node 6

The linear program will have 18 variables for the arcs and 7 constraints for the nodes.

1 if the arc from node i to node j is on the shortest route



Let ij 
x
0 otherwise


Min 35 x12  30 x13  20 x14  8 x23  12 x25  8 x32  9 x34  10 x35  20 x36

9 x43  15 x47  12 x52  10 x53  5 x56  20 x57  15x74  20 x75  5 x76

s.t.
Flow Out Flow In
Node 1 x12  x13  x14 =1
Node 2 x23  x25  x12  x32  x52 =0
Node 3 x32  x34  x35  x36  x13  x23  x43  x53 =0
Node 4 x43  x47  x14  x34  x74 =0
Node 5 x52  x53  x56  x57  x25  x35  x75 =0
Node 6  x36  x56  x76 =1
Node 7 x74  x75  x76  x47  x57 =0

xij > 0 for all i and j

Optimal Solution: x14 1 , x47 1 , and x76 1

Shortest Route 1-4-7-6

Total Distance = 40 miles

27. Origin – Node 1

Transshipment Nodes 2 to 9

Destination – Node 10 (Identified by the subscript 0)

The linear program will have 29 variables for the arcs and 10 constraints for the nodes.

6 - 25

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

1 if the arc from node i to node j is on the shortest route


Let xij 
0 otherwise


Min 8 x12  13x13  15 x14  10 x15  5 x23  15 x27  5 x32  5 x36  2 x43  4 x45
3x46  4 x54  12 x59  5 x63  3 x64  4 x67  2 x68  5 x69  15 x72  4 x76
2 x78  4 x70  2 x86  5 x89  7 x80  12 x95  5 x96  5 x98  5 x90

s.t.
Flow Out Flow In
Node 1 x12  x13  x14  x15 =1
Node 2 x23  x27  x12  x32  x72 =0
Node 3 x32  x36  x13  x23  x43  x63 =0
Node 4 x43  x45  x46  x14  x54  x64 =0
Node 5 x54  x59  x15  x45  x95 =0
Node 6 x63  x64  x67  x68  x69  x36  x46  x76  x86  x96 =0
Node 7 x72  x76  x78  x70  x27  x67 =0
Node 8 x86  x89  x80  x68  x78  x98 =0
Node 9 x95  x96  x98  x90  x59  x69  x89 =0
Node 10  x70  x80  x90 =1
xij > 0 for all i and j

Optimal Solution: x15 1 , x54 1 , x46 1 , x67 1 , and x70 1

Shortest Route 1-5-4-6-7-10

Total Time = 25 minutes

28. Origin – Node 0

Transshipment Nodes 1 to 3

Destination – Node 4

The linear program will have 10 variables for the arcs and 5 constraints for the nodes.

1 if the arc from node i to node j is on the minimum cost route


Let xij 
0 otherwise


Min 600 x01  1000 x02  2000 x03  2800 x04  500 x12  1400 x13  2100 x14

800 x23  1600 x24  700 x34

s.t.
Flow Out Flow In
Node 0 x01  x02  x03  x04 =1

6 - 26

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

Node 1 x12  x13  x14  x01 =0

Node 2 x23  x24  x02  x12 =0
Node 3 x34  x03  x13  x23 =0
Node 4  x04  x14  x24  x34 =1
xij > 0 for all i and j

Optimal Solution: x02 1 , x23 1 , and x34 1

Shortest Route 0-2-3-4

Total Cost = $2500

29. The capacitated transshipment problem to solve is given:

Max x61
s.t.
x12 + x13 + x14 - x61 =0
x24 + x25 - x12 - x42 =0
x34 + x36 - x13 - x43 =0
x42 + x43 + x45 + x46 - x14 - x24 - x34 - x54 = 0
x54 + x56 - x25 - x45 =0
x61 - x36 + x46 - x56 =0
x12  2 x13  6 x14  3
x24  1 x25  4
x34  3 x36  2
x42  1 x43  3 x45  1 x46  3
x54  1 x56  6
xij  0 for all i, j

3
2 5

2 1 1 4

Maximum Flow
3 3
1 4 6 9,000 Vehicles
Per Hour

4 2 2

3
The system cannot accommodate a flow of 10,000 vehicles per hour.

30.

6 - 27

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

4
2 5

3 1 2 6

3 3
1 4 6 11,000

5 3 2

31. The maximum number of messages that may be sent is 10,000.

32. a. 10,000 gallons per hour or 10 hours

b. Flow reduced to 9,000 gallons per hour; 11.1 hours.

33. Current Max Flow = 6,000 vehicles/hour.

With arc 3-4 at a 3,000 unit/hour flow capacity, total system flow is increased to 8,000
vehicles/hour. Increasing arc 3-4 to 2,000 units/hour will also increase system to 8,000
vehicles/hour. Thus a 2,000 unit/hour capacity is recommended for this arc.

34. Maximal Flow = 23 gallons / minute. Five gallons will flow from node 3 to node 5.

35. a. Modify the problem by adding two nodes and two arcs. Let node 0 be a beginning inventory node
with a supply of 50 and an arc connecting it to node 5 (period 1 demand). Let node 9 be an ending
inventory node with a demand of 100 and an arc connecting node 8 (period 4 demand to it).
b.
Min 2x15 + 5x26 + 3x37 + 3x48 + 0.25x56 + 0.25x67 + 0.25x78 + 0.25x89
s.t.
x05 = 50
x15  600
x26  300
x37  500
x48  400
x05 + x15 - x56 = 400
x26 + x56 - x67 = 500
x37 + x67 - x78 = 400
x48 + x78 - x89 = 400
x89 = 100
xij  0 for all i and j

Optimal Solution:

6 - 28

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Distribution and Network Models

x05 = 50 x56 = 250

x15 = 600 x67 = 0
x26 = 250 x78 = 100
x37 = 500 x89 = 100
x48 = 400

Total Cost = $5262.50

36. a. Let R1, R2, R3 represent regular time production in months 1, 2, 3

O1, O2, O3 represent overtime production in months 1, 2, 3
D1, D2, D3 represent demand in months 1, 2, 3

Using these 9 nodes, a network model is shown.

275 R1

D1 150

100 O1

200 R2

D2 250

50 O2

100 R3

D3 300

50 O3

b. Use the following notation to define the variables: first two letters designates the "from node" and
the second two letters designates the "to node" of the arc. For instance, R1D1 is amount of regular
time production available to satisfy demand in month 1, O1D1 is amount of overtime production in

6 - 29

Downloaded by Amr Hemdan ([email protected])

 lOMoARcPSD|23416644

Chapter 6

month 1 available to satisfy demand in month 1, D1D2 is the amount of inventory carried over
from month 1 to month 2, and so on.

MIN 50R1D1 + 80O1D1 + 20D1D2 + 50R2D2 + 80O2D2 + 20D2D3 + 60R3D3 + 100O3D3

S.T.
1) R1D1  275
2) O1D1  100
3) R2D2  200
4) O2D2  50
5) R3D3  100
6) O3D3  50
7) R1D1 + O1D1 - D1D2 = 150
8) R2D2 + O2D2 + D1D2 - D2D3 = 250
9) R3D3 + O3D3 + D2D3 = 300

c. Optimal Solution:

Variable Value
-------------- ---------------
R1D1 275.000
O1D1 25.000
D1D2 150.000
R2D2 200.000
O2D2 50.000
D2D3 150.000
R3D3 100.000
O3D3 50.000

Value = $46,750

Note: Slack variable for constraint 2 = 75.

d. The values of the slack variables for constraints 1 through 6 represent unused capacity. The only
nonzero slack variable is for constraint 2; its value is 75. Thus, there are 75 units of unused
overtime capacity in month 1.

6 - 30

Downloaded by Amr Hemdan ([email protected])