Chapter 14:

Applications of Linear Optimization

Business Analytics: Methods, Models,

and Desicions, 1st edition
James R. Evans

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-1
Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-2
Chapter 14 Topics
 Applications of Linear Optimization
 Types of Constraints in Optimization Models
 Process Selection Models
 Blending Models
 Portfolio Investment Models
 Transportation Models
 Multiperiod Production Planning Models
 Multiperiod Financial Planning Models
 Models with Bounded Variables
 A Production/Marketing Allocation Model

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-3
Applications of Linear Optimization
 Linear optimization models are the most
ubiquitous of optimization models used in
organizations today.
 Applications abound in operations, finance,

marketing, engineering, etc.

 Building models is more of an art than a science.
 Learning to build models requires logical thought

facilitated by studying examples and observing

their characteristics.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-4
Applications of Linear Optimization

Table 14.1

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-5
Types of Constraints in Optimization
Models
Simple Bounds
 minimum or maximum values for variables

Limitations
 allocation of scarce resources

Requirements
 minimum levels of performance

Proportional Relationships
 requirements for mixing or blending

Balance Constraints
 input equals output

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-6
Process Selection Models
Example 14.1 Camm Textiles
 A mill that operates on a 24/7 basis produces

three types of fabric on a make-to-order basis.

 The key decision is which type of loom to use for

each fabric type over the next 13 weeks.

 The mill has 3 dobbie looms and 15 regular

looms.
 If demand cannot be met, the fabric is outsourced.

Table 14.2
Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-7
Process Selection Models

Example 14.1 (continued) Camm Textiles

 D = yards fabric i to produce on dobbie loom
i
 Ri = yards fabric i to produce on regular loom
 Pi = yards fabric i to outsource
Objective
 Minimize total cost of milling and outsourcing

Constraints
 Limitations - loom capacities
 Requirements - meet demand
 Nonnegativity

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-8
Process Selection Models
Example 14.1 (continued) Camm Textiles
Loom capacity limitation constraints
 Convert yards/hour into hours/yard.
 For fabric 1 on the dobbie loom

hours/yard = 1/(4.7 yards/hour) = 0.213 hours/yard

 For the RHS capacity of the dobbie loom

(24 hours/day)(7 days/week)(13 weeks)(3 looms)

= 6,552 hours/loom

Table 14.2

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-9
Process Selection Models

Example 14.1 (continued) Camm Textiles

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-10
Process Selection Models
Spreadsheet Design and Solver Reports
Camm Textiles LP model

Figure 14.1b
Pink cells are model parameters.
Beige cells are solution values of the decision variables.
Green cell is the optimal value of the objective.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-11
Process Selection Models
Spreadsheet Design and Solver Reports
Camm Textiles LP model

Figure 14.1a

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-12
Process Selection Models
Spreadsheet Design and Solver Reports
Risk Solver Parameter
dialog for Camm Textiles

Set a bound of C14 = 0

since we included C14 in
the B14:D16 variables but
the regular loom cannot
produce fabric 1.

Figure 14.2

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-13
Process Selection Models
Spreadsheet Design and Solver Reports
Regular Solver Parameter
dialog for Camm Textiles

Specify a constraint
C14 = 0 since the
regular loom cannot
produce fabric 1.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-14
Process Selection Models
Example 14.2 Interpreting Solver Reports for the
Camm Textiles Problem
Solver Answer Report
Profit = $87,756.12
D1 = 30794.4 yards
P1 = 14205.6 yards
R2 = 76500.0 yards
R3 = 10000.0 yards

The regular loom

has excess capacity.

Figure 14.3

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-15
Process Selection Models
Example 14.2 (continued) Interpreting Solver
Reports for the Camm Textiles Problem
Solver Sensitivity Report
The outsourcing cost
of fabric 2 needs to
decrease by $0.14
to become
economical.

Producing an extra
yard of fabric 1
costs $0.85 (its
outsourcing cost).
Figure 14.4

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-16
Blending Models

Example 14.3 BG Seed Company

 BG Seed Company mixes food products for a

variety of birdseed mixes.

 A new birdseed product is being developed and it

can use 8 ingredients.

 Nutritional requirements specify that the mixture

contain at least 13% protein, at least 15% fat, and

no more than 14% fiber.
 BG’s objective is to determine the minimum cost

mixture that meets nutritional requirements.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-17
Blending Models
Example 14.3 (continued) BG Seed Company
 Define the 8 decision variables as:

Xi = pounds of ingredient i in 1 pound of mix

Table 14.3

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-18
Blending Models
Example 14.3 (continued) BG Seed Company

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-19
Blending Models

Dealing With Infeasibiltiy

BG Seed LP formulation in Excel

Figure 14.5

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-20
Blending Models
Dealing With Infeasibility
Risk Solver Parameter dialog for BG Seed Company

Figure 14.6

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-21
Blending Models

Dealing With Infeasibility

 Solver Feasibility Report for BG Seed Company

Figure 14.7

A conflict exists in trying to meet both fat and fiber constraints.

Only sunflower seeds and safflower contain enough fat but they
also have a lot of fiber.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-22
Blending Models
Dealing With Infeasibility
BG Seed Company

Table 14.3

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-23
Blending Models

Dealing With Infeasibility

BG Seed Company – Two What-If Scenarios
1st Scenario:
Fat requirement is lowered from
15% to 14.5%.
2nd Scenario:
Fiber limitation is raised from
14% to 14.5%.

Optimal Cost per pound:

$0.148 if fat requirement lowered
Figure 14.8
$0.152 if fiber limitation raised

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-24
Portfolio Investment Models
Example 14.4 Innis Investments
 Innis Investments manages 6 mutual funds.
 A client wants to invest a $500,000 inheritance.
 The objective is to minimize risk.

Constraints:
 Invest no more than $200,000 in any one fund.
 Invest at least $50,000 each in the multinational and

balanced funds.
 Invest at least 40% combined in the income equity and

balanced funds.
 Achieve an average return of at least 5%.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-25
Portfolio Investment Models
Example 14.4 (continued) Innis Investments
 Define the 6 decision variables as:

Xi = dollar amount invested in fund i

Table 14.4

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-26
Portfolio Investment Models
Example 14.4 (continued) Innis Investments

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-27
Portfolio Investment Models
Innis Investments – Excel worksheet formulas

Figure 14.9b

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-28
Portfolio Investment Models

Risk Solver Parameter dialog for Innis Investments

Figure 14.10

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-29
Portfolio Investment Models
Innis Investments – Excel worksheet solution

Risk = 6.3073
X1 = 0
X2 = $50,000
X3 = 0
X4 = $200,000
X5 = $66,371
X6 = $183,628

Figure 14.9a

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-30
Portfolio Investment Models
Example 14.5 Risk Versus Reward
Innis Investment Solutions for varying target returns

As target
return
increases,
so does
risk.

Figure 14.11

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-31
Transportation Models
Example 14.6 General Appliance Corporation
 GAC produces refrigerants at 2 plants.
 GAC ships refrigerants to 5 distribution centers.
 The objective is to minimize the total cost of

shipping between plants and distribution centers.

 Define the 10 decision variables as: X = amount
ij
shipped from plant i to distribution center j

Table 14.5

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-32
Transportation Models
Example 14.6 (continued)
General Appliance Corporation

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-33
Transportation Models
Example 14.6 (continued)
General Appliance Corporation
 Excel worksheet formulas

From Figure 14.12

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-34
Transportation Models

Example 14.6 (continued) General Appliance

Corporation
Risk Solver
Parameter dialog

Figure 14.13

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-35
Transportation Models
Example 14.6 (continued)
General Appliance Corporation
 Excel worksheet solution

Cost = $28,171
X11 = 0
X12 = 350
X13 = 0
X14 = 850
X21 = 150
X22 = 0
X23 = 500
X24 = 150
Figure 14.12

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-36
Transportation Models
Example 14.7 Interpreting Sensitivity Information
for the GAC Model
Original version of Sensitivity Report
Reduced costs and
shadow prices are
formatted as whole
numbers.

Reformat those cells

to see decimal
values.

Figure 14.14

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-37
Transportation Models
Example 14.7 (continued) Interpreting Sensitivity
Information for the GAC Model
Accurate Sensitivity Report Reduced cost are zero for
variables that are positive.

Allowable increase in all the

demand constraints is zero.

Shadow prices reflect the

cost savings for a unit
decrease in demand.

The solution is degenerate

because the number of
positive variables is less
than the number of
Figure 14.15
constraints (5 < 6).
Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-38
Multiperiod Production Planning Models
Example 14.8 K&L Designs
 K&L Designs makes hand-painted jewelry boxes.

 Forecasted sales are 150 in autumn, 400 in winter, and

50 in spring.
 Unpainted boxes cost $20 and each box takes 2 hours

to complete.
 The cost of capital is 6% per quarter.

 Holding cost per item = 0.06(20) = $1.20/quarter

 $/hour labor rates are 5.5, 7, and 6.25 during autumn,

winter, and spring, respectively.

 Objective: Minimize the combined cost of production

and inventory holding costs.

Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-39
Multiperiod Production Planning Models

Example 14.8 (continued) K&L Designs

 P = amount to produce in season i
i
 Ij = inventory at the end of season j

Figure 14.16

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-40
Multiperiod Production Planning Models
Example 14.8 (continued) K&L Designs

Objective function coefficient for PS variable:

($6.25/hour)(2 hours/box) = $12.50
Box costs need not be subtracted because they are constant.
Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-41
Multiperiod Production Planning Models

Example 14.8 (continued) K&L Designs

 Solving the LP using Risk Solver

Figure 14.17b

Figure 14.18

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-42
Multiperiod Production Planning Models

Example 14.8 (continued) Produce:

K&L Designs 550 units in autumn
0 units in winter
50 units in spring

Figure 14.17

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-43
Multiperiod Production Planning Models

Example 14.9 An Alternative Optimization Model for

K&L Designs
 The constraints simply ensure demand is met.

So we can eliminate the Ij decision variables.

Figure 14.19
Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-44
Multiperiod Production Planning Models

Example 14.9 (continued) An Alternative

Optimization Model for K&L Designs
 Solving the LP using Risk Solver

Figure 14.19b
Figure 14.20

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-45
Multiperiod Production Planning Models
Example 14.9 (continued)
An Alternative Optimization Again, produce:
550 units in autumn
Model for K&L Designs 0 units in winter
50 units in spring

Figure 14.19

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-46
Multiperiod Production Planning Models
Example 14.9 (continued) An Alternative
Optimization Model for K&L Designs
We obtain the same results using the “streamlined” alternate model.

Figure 14.21

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-47
Multiperiod Financial Planning Models

Example 14.10 D.A. Branch & Sons

 The company’s financial manager needs to insure

that funds are available to pay expenses yet

needs to maximize investment income.
Three short-term investments are being considered:
 1-month CD paying 0.25%
 3-month CD paying 1.00%
 6-month CD paying 2.30%

A cash balance of $10,000 must be maintained

 Currently the cash balance is $200,000.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-48
Multiperiod Financial Planning Models
Example 14.10 (continued) D.A. Branch & Sons
 Net expenditures for the next 6 months

Figure 14.22

Ai = amount ($) to invest in a 1-month CD at the start of month i

Bi = amount ($) to invest in a 3-month CD at the start of month i
Ci = amount ($) to invest in a 6-month CD at the start of month i

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-49
Multiperiod Financial Planning Models
Example 14.10 (continued) D.A. Branch & Sons

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-50
Multiperiod Financial Planning Models
Example 14.10 (continued) D.A. Branch & Sons

Figure 14.23b

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-51
Multiperiod Financial Planning Models

Example 14.10 (continued) D.A. Branch & Sons

 Solving the LP using Risk Solver

Figure 14.23b Figure 14.24

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-52
Multiperiod Financial Planning Models
Example 14.10 (continued) D.A. Branch & Sons

Total return = $125,634

Invest in a 1-month CD
at the end of months 1,
2, and 6.
Invest in a 3-month CD
at the end of months 1,
2, and 4.

Figure 14.23a

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-53
Multiperiod Financial Planning Models

Analytics in Practice:
Linear Optimization in
Bank Financial Planning
 Carolina Central Bank uses linear optimization to
coordinate activities of its banking business.
 A 1-year single-period model containing 66 asset

and 32 liability and equity categories is used.

 Even though the model ignores many time-related

linkages, it is a model that is of appropriate detail

for senior management use.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-54
Models with Bounded Variables
Example 14.11 J&M Manufacturing
 J&M Manufacturing makes 4 models of gas grills
 Production managers want to determine how

many grills to produce in order to maximize profit.

Table 14.6

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-55
Models with Bounded Variables
Example 14.11 (continued) J&M Manufacturing

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-56
Models with Bounded Variables

Example 14.11 (continued) J&M Manufacturing

 Solving the LP using Risk Solver

Figure 14.26
Figure 14.25b

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-57
Models with Bounded Variables

Example 14.11 (continued) J&M Manufacturing

Total profit = $407,857

Optimal number of gas
grills to produce:
A = 3857
B=0
C =1236
D=0

Figure 14.25a

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-58
Models with Bounded Variables
Example 14.11 (continued) J&M Manufacturing
Solver Answer Report

Binding constraints
Upper bound on sales
of grill D
Lower bound on sales
of grill B

Figure 14.27

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-59
Models with Bounded Variables

Example 14.11 (continued) J&M Manufacturing

Solver Sensitivity Report
The 8 boundary
constraints do not
appear in this report.

The reduced cost of

a bounded variable is
equal to the shadow
price of its boundary
constraint.

Figure 14.28

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-60
Models with Bounded Variables

Example 14.12 Using Auxiliary Variable Cells

(for the J&M Manufacturing model)
 Auxiliary variables allow easier interpretation of

shadow prices for bounded variables.

 If we add auxiliary variables to the Excel model

formulation, then Solver’s sensitivity report will:

- show the boundary constraints
- give us information about allowable increases
and decreases in the RHS values.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-61
Models with Bounded Variables
Example 14.12 (continued)
Using Auxiliary Variable Cells
 The 4 auxiliary variables in row 29 are simply set

equal to the 4 decision variables in row 25.

Figure 14.29

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-62
Models with Bounded Variables
Example 14.12 (continued)
Using Auxiliary Variable Cells
 Solving the LP using Risk Solver

Add the 8 boundary

constraints to the model.

Figure 14.30

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-63
Models with Bounded Variables
Example 14.12 (continued) Using Auxiliary Variables
Solver Sensitivity Report
The 8 boundary
constraints now
appear in the report.

Auxiliary variable
constraints
correspond to the
boundary
constraints.

Figure 14.31

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-64
A Production/Marketing Model

Example 14.13 Walker Wines

 Walker Wines buys grapes from local growers and

blends the pressings to make two types of wine

(Shiraz and Merlot).
 Grape costs are $1.60 per bottle Shiraz and $1.40 per

bottle Merlot.
 Their contract requires between 40% and 70% of their

wine to be Shiraz.
 Predicted demand for Shiraz is 1000 bottles but

increases by 5 bottles for each dollar spent on

advertising.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-65
A Production/Marketing Model

Example 14.13 (continued) Walker Wines

 Demand for Merlot is 2000 bottles but increases

by 8 for each dollar spent on advertising.

 Selling prices per bottle are $6.25 for Shiraz and

$5.25 for Merlot.

 Walker Wines has $50,000 available to purchase

grapes and advertising.

 Wine production should not exceed demand.
 The objective is to maximize profit.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-66
A Production/Marketing Model
Example 14.13 (continued) Walker Wines

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-67
A Production/Marketing Model

Example 14.13 (continued) Walker Wines

 Solving the LP using Risk Solver

Figure 14.32b
Figure 14.33

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-68
A Production/Marketing Model
Example 14.13 (continued) Walker Wines

Total profit = $124,776

Optimal number of
bottles to produce:
Shiraz = 20,562
Merlot = 8,812

Optimal advertising:
Shiraz = $3912
Merlot = $851

Figure 14.32a

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-69
A Production/Marketing Model
Example 14.14
Evaluating a Cost Increase for Walker Wines
Suppose the cost of Shiraz grapes increase $0.05 per
bottle. Solver Sensitivity Report
We need to re-solve
the LP to evaluate
the cost increase
because this cost
appears in multiple
places in the model
(in the objective
function and in the
binding budget
constraint).
Figure 14.34

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-70
A Production/Marketing Model
Example 14.14 (continued)
Evaluating a Cost Increase for Walker Wines
Solution with a $0.05 increase in
the cost of Shiraz grapes.

Change in total profit = - $2545

Change in the optimal number of
bottles to produce:
Shiraz = - 8,369
Merlot = + 9478

Change in optimal advertising:

Shiraz = - $1673
Merlot = + $1185
Figure 14.35

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-71
Chapter 14 - Key Terms
 Auxiliary variables
 Balance constraints
 Degenerate solution
 Feasibility report
 Limitations
 Proportional relationships
 Requirements
 Simple bounds
 Transportation problem

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-72
Case Study
Performance Lawn Equipment (14)
 Recall that PLE produces lawnmowers and a
medium size diesel power lawn tractor.
 PLE need to develop a model to more effectively
plan production for next year.
 The operations manager would like you to develop
an optimization model that would minimize
inventory costs while meeting anticipated demand.
 Provide detailed information on monthly costs,
production, and inventory, and write a formal report
summarizing your results.

Copyright © 2013 Pearson Education, Inc.

publishing as Prentice Hall 14-73
Copyright © 2013 Pearson Education, Inc.
publishing as Prentice Hall 14-74