Linear Optimization Models

Quantitative Business Analysis

 Introduction

• Optimization Problems:
– Can be used to support and improve managerial decision making.
– Maximize or minimize some function, called the objective function, and
have a set of restrictions known as constraints.
– Can be linear or nonlinear.
 Introduction

• Typical Applications:
– A manufacturer wants to develop a production schedule and an inventory policy that
will satisfy demand in future periods and at the same time minimize the total
production and inventory costs.
– A financial analyst would like to establish an investment portfolio from a variety of
stock and bond investment alternatives that maximizes the return on investment.
– A marketing manager wants to determine how best to allocate a fixed advertising
budget among alternative advertising media such as web, radio, television, newspaper,
and magazine that maximizes advertising effectiveness.
– A company had warehouses in a number of locations. Given specific customer
demands, the company would like to determine how much each warehouse should
ship to each customer so that total transportation costs are minimized.
 Introduction

• Linear optimization models are also known as linear programs.

• Linear programming:
– A problem-solving approach developed to help managers make better
decisions.
– Numerous applications in today’s competitive business environment.
– For instance, GE Capital uses linear programming to help determine
optimal lease structuring.
– Marathon Oil Company uses linear programming for gasoline blending
and to evaluate the economics of a new terminal or pipeline.
Problem Formulation
Mathematical Model for the Par, Inc. Problem

A SIMPLE MAXIMIZATION PROBLEM

 A Simple Maximization Problem

Illustration: Par, Inc.:

– Par, Inc.—A small manufacturer of golf equipment and supplies.
– Management has decided to move into the market for medium- and high-
priced golf bags.
– Par’s distributor has agreed to buy all of the golf bags produced over the
next three months.
 A Simple Maximization Problem
Illustration: Par, Inc. (cont.):
• Operations involved in manufacturing a golf bag:
– Cutting and dyeing the material.
– Sewing.
– Finishing (inserting umbrella holder, club separators, etc.).
– Inspection and packaging.

Table 12.1: Production

Requirements Per Golf
Bag
 A Simple Maximization Problem

Illustration: Par, Inc. (cont.):

• Estimated total time available for the next three months to perform different operations:

Department Number of hours

Cutting and Dyeing 630
Sewing 600
Finishing 708
Inspection and Packaging 135

• Required profit contribution:

– Standard bag: $10/unit.
– Deluxe bag: $9/unit.
 A Simple Maximization Problem

Illustration: Par, Inc. (cont.):

• Develop a mathematical model of the Par, Inc. problem to determine the
number of standard bags and the number of deluxe bags to produce to
maximize total profit contribution.
Problem Formulation:
• Problem formulation or modeling: Process of translating the verbal
statement of a problem into a mathematical statement.
 A Simple Maximization Problem

Problem Formulation (cont.):

• General guidelines for problem formulation:
• Understand the problem thoroughly.
• Describe the objective.
• Describe each constraint.
• Define the decision variables.
• Write the objective in terms of the decision variables.
• Write the constraints in terms of the decision variables.
 A Simple Maximization Problem

Problem Formulation (cont.):

• Describe each constraint.
Constraint Description
1 Number of hours of cutting and dyeing time used must be less than
or equal to the number of hours of cutting and dyeing time
available.
2 Number of hours of sewing time used must be less than or equal to
the number of hours of sewing time available.
3 Number of hours of finishing time used must be less than or equal
to the number of hours of finishing time available.
4 Number of hours of inspection and packaging time used must be
less than or equal to the number of hours of inspection and
packaging time available.
 A Simple Maximization Problem

Problem Formulation (cont.):

• Define the decision variables:
– S = number of standard bags.
– D = number of deluxe bags.
• Write the objective in terms of the decision variables.
– If Par makes $10 for every standard and $9 for every deluxe bag,

Total profit contribution 10 S  9D objective function.

Objective: Max 10 S  9D.
 A Simple Maximization Problem
Problem Formulation (cont.):
• Write the constraints in terms of the decision variables:
– Constraint 1:

 Hours of sewing   Hours of sewing   Hours of cutting and   Hours of cutting and 
       
 time used   time available   dyeing time used   dyeing time available 

7
S  1D 630
1 5
S  D 600
10
2 6

• Constraint 2:
 A Simple Maximization Problem

 Hours of finishing   Hours of finishing  1S  2 D 708

Problem Formulation (cont.):
    3
 time used   time available 
– Constraint 3:
 Hours of inspection and   Hours of inspection and  1 1
– Constraint 4:  packaging time used    S  D 135
   packaging time available  10 4

– Nonnegativity constraints: Based on the fact that the number of

standard or deluxe bags produced cannot be negative.
S 0 and D 0 or S , D 0
 A Simple Maximization Problem

Mathematical Model for the Par, Inc. Problem:

• Mathematical model: A set of mathematical relationships Max 10 S  9D.
subject to (s.t.)
7
S  1D 630 Cutting and dyeing
10
1 5
S  D 600 Sewing
2 6
2
1S  D 708 Finishing
3
1 1
S  D 135 Inspection and packaging
10 4
S , D 0
 A Simple Maximization Problem
Mathematical Model for the Par, Inc. Problem (cont.):
• This is a linear programming model (or linear program) because the
objective function and all constraint functions are linear functions of the
decision variables.
• Linear function: Mathematical function in which each variable appears in
a separate term and is raised to the first power.
The Geometry of the Par, Inc. Problem
Solving Linear Programs with Excel Solver

SOLVING THE PAR, INC. PROBLEM

 Solving the Par, Inc. Problem

• To find the optimal solution to the problem modeled as a

linear program:
– The optimal solution must have the highest objective function value.
– The optimal solution must be a feasible solution—a setting of the
decision variables that satisfies all of the constraints of the problem.
– Search over the feasible region—a set of all possible solutions.
– Find the solution that gives the best objective function value.
 Solving the Par, Inc. Problem

The Geometry of the Par, Inc. Problem:

• When there are only two decision variables and the functions of these
variables are linear, they form lines in two-dimensional space.
• If constraints are inequalities, the constraint cuts the space in two:
– The line and the area on one side of the line is the space the
satisfies that constraint.
– These subregions are called half spaces.
– The intersection of the half spaces make up the feasible region.
 Solving the Par, Inc. Problem

Figure 12.1: Feasible Region for

the Par, Inc. Problem
 Solving the Par, Inc. Problem

Figure 12.2: The Optimal Solution to

the Par, Inc. Problem
 Solving the Par, Inc. Problem

The Geometry of the Par, Inc. Problem (cont.):

• Based on the geometry of Figure 12.2, to solve a linear optimization problem
we only have to search the extreme points of the feasible region to find the
optimal solution.
• Extreme points are found where constraints intersect on the boundary of the
feasible region.
 Solving the Par, Inc. Problem

The Geometry of the Par, Inc. Problem (cont.):

• The values of S and D and the objective function value are:
 Solving the Par, Inc. Problem

The Geometry of the Par, Inc. Problem (cont.):

• The approach of investigating only extreme points works well and
generalizes for larger problems.
• The simplex algorithm, developed by George Dantzig, is quite effective at
investigating extreme points in an intelligent way to find the optimal solution
to even very large linear programs.
• Excel Solver is software that utilizes Dantzig’s simplex algorithm to solve
linear programs by systematically finding which set of constraints form the
optimal extreme point of the feasible region.
• Once it finds an optimal solution, Solver then reports the optimal values of
the decision variables and the optimal objective function value.
 Solving the Par, Inc. Problem

Solving Linear Programs with Excel Solver:

• The first step is to construct the relevant what-if model.
• A what-if model for optimization allows the user to try different values of the
decision variables and see:
– Whether that trial solution is feasible.
– The value of the objective function for that trial solution.
 Solving the Par, Inc. Problem

Figure 12.3: What-If Spreadsheet Model for Par, Inc.

 Solving the Par, Inc. Problem

Figure 12.4: Solver Dialog Box and

Solution to the Par, Inc. Problem
 Solving the Par, Inc. Problem

Solving Linear Programs with Excel Solver (cont.):

• The optimal solution:
– To make 540 Standard bags and 252 Deluxe bags for a profit of
$7,668.
– Using all the cutting and dyeing time as well as all finishing time,
from cells B19:B22 compared to C19:C22.
• The results are consistent with the results obtained in Figures 12.1 and 12.2.
 Solving the Par, Inc. Problem

Figure 12.5: The Solver Answer

Report for the Par, Inc. Problem
 Solving the Par, Inc. Problem

Solving Linear Programs with Excel Solver (cont.):

• A binding constraint is one that holds as an equality at the optimal solution.
• The slack value for each less-than-or-equal-to constraint indicates the
difference between the left-hand and right-hand values for a constraint.
• By adding a nonnegative slack variable, we can make the constraint equality.
Problem Formulation
Solution for the M&D Chemicals Problem

A SIMPLE MINIMIZATION PROBLEM

 A Simple Minimization Problem

Illustration: M&D Chemicals:

• Production requirements for M&D Chemicals:
– The combined production for products A and B must total at least 350
gallons.
– Separately, a major customer’s order for 125 gallons of product A must
also be satisfied.
• Processing time:
– Product A: 2 hours/gallon.
– Product B: 1 hour/gallon.
– For the coming month, 600 hours of processing time are available.
 A Simple Minimization Problem

Illustration: M&D Chemicals (cont.):

• Production cost: Product A: $2/gallon; Product B: $3/gallon.
• Objective: Minimizing the total production cost.

Problem Formulation:
• To find the minimum-cost production schedule:
– Define the decision variables and the objective function:
Let A = number of gallons of product A to produce.
B = number of gallons of product B to produce.
– Objective function

Min 2 A  3B.
 A Simple Minimization Problem

Problem Formulation (cont.):

• Linear program for the M&D Chemicals problem:
Min 2A  3B
s.t.
1A 125 Demand for product A
1A  1B 350 Total production
2 A  1B 600 Processing time
A, B 0
 A Simple Minimization Problem

Solution for the M&D Chemicals Problem:

• A spreadsheet model for the M&D Chemicals problem along with the Solver
dialog box are shown in Figure 12.6.
• The optimal solution (Figure 12.7) is to make 250 gallons of product A and
100 gallons of product B, for a total cost of $800.
• Both the total production constraint and the processing time constraints are
binding; the requirement that at least 125 gallons of Product A be produced
is not binding.
• A surplus variable tells how much over the right-hand side the left-hand side
of a greater-than-or-equal-to constraint is for a solution.
 A Simple Minimization Problem

Figure 12.6: Solver Dialog Box and Solution to the M&D Chemical Problem
 A Simple Minimization Problem

Figure 12.7: The Solver

Answer Report for the
M&D Chemicals Problem
Alternative Optimal Solutions
Infeasibility
Unbounded

SPECIAL CASES OF LINEAR PROGRAM

OUTCOMES
 Special Cases of Linear Program Outcomes

Alternative Optimal Solutions:

• An alternative optimal solution is one in which the optimal objective
function contour line coincides with one of the binding constraint lines
on the boundary of the feasible region.
• In these situations, more than one solution provides the optimal value for
the objective function.
 Special Cases of Linear Program Outcomes

Illustration: Par, Inc.:

• Original objective function: 10 S  9D.
• Assume the profit for the standard golf bag decreased to $6.30.
• Revised objective function: 6.3S  9D.
• The optimal solution occurs at two extreme points:
– Extreme point 4 (S = 300, D = 420).
– Extreme point 3 (S = 540, D = 252).
 Special Cases of Linear Program Outcomes

Figure 12.8: Par, Inc. Problem

with an Objective Function of

6.3S  9D (Alternative
Optimal Solutions)
 Special Cases of Linear Program Outcomes

Infeasibility:
• Infeasibility means no solution to the linear programming problem
satisfies all the constraints, including nonnegativity
conditions.
• Graphically, a feasible region does not exist.
• Infeasibility occurs because:
– Management’s expectations are too high.
– Too many restrictions have been placed on the problem.
 Special Cases of Linear Program Outcomes

Figure 12.9: No Feasible Region

for the Par, Inc. Problem with
Minimum Production
Requirements of 500 Standard
and 360 Deluxe Bags
 Special Cases of Linear Program Outcomes

Infeasibility (cont.):
• Interpretation of Infeasibility for the Par, Inc. problem.
• Let management know that the resources available are not sufficient to
make 500 standard bags and 360 deluxe bags.
• Provide details to management on:
– Minimum amounts of resources that must be available.
– The amounts currently available.
– Additional amounts that would be required to accomplish this level
of production.
 Special Cases of Linear Program Outcomes

Table 12.2: Resources Needed to Manufacture 500 Standard Bags and

360 Deluxe Bags
 Special Cases of Linear Program Outcomes

Infeasibility (cont.):
• An infeasible problem when solved in Excel Solver:
– Will return a message indicating that no feasible solutions exists—indicating no
solution to the linear programming problem will satisfy all constraints.
– Careful inspection of your formulation is necessary to identify why the problem is
infeasible.
– One of the approaches is to drop one or more constraints and re-solve the
problem.
– If we find an optimal solution for this revised problem, then the constraint(s) that
were omitted, in conjunction with the others, are causing the problem to be
infeasible.
 Special Cases of Linear Program Outcomes

Unbounded:
• Unbounded: The situation in which the value of the solution:
– May be made infinitely large—for a maximization linear
programming.
– May be made infinitely small—for a minimization linear
programming.
– Without violating any of the constraints.
 Special Cases of Linear Program Outcomes

Illustration:
Consider the following linear program with two decision variables, X and Y:
Max 20 X  10Y
s.t.
1 X 2
1Y 5
X , Y 0
 Special Cases of Linear Program Outcomes

Figure 12.10: Example of an

Unbounded Problem
 Special Cases of Linear Program Outcomes

Unbounded (cont.):
• Solving an unbounded problem using Excel Solver returns a message
“Objective Cell values do not converge.”
• In linear programming models of real problems, the occurrence of an
unbounded solution means that the problem has been improperly
formulated.
• If a profit maximization problem results in an unbounded solution, the
mathematical model does not represent the real-world problem sufficiently.
• In many cases, this error is the result of inadvertently omitting a constraint
during problem formulation.
Interpreting Excel Solver Sensitivity Report

SENSITIVITY ANALYSIS
 Sensitivity Analysis

• Sensitivity analysis: The study of how the changes in the input parameters
of an optimization model affect the optimal solution.
• It helps in answering the questions:
– How will a change in a coefficient of the objective function affect the
optimal solution?
– How will a change in the right-hand-side value for a constraint affect
the optimal solution?
• Because sensitivity analysis (often referred to as postoptimality analysis) is
concerned with how these changes affect the optimal solution, the analysis
does not begin until the optimal solution to the original linear programming
problem has been obtained.
 Sensitivity Analysis

Interpreting Excel Solver Sensitivity Report

• Consider the M&D Chemicals problem:
A = number of gallons of product A.
B = number of gallons of product B.
Min 2 A  3B
s.t.
1A 125 Demand for product A
1A  1B 350 Total production
2 A  1B 600 Processing time
A, B 0
 Sensitivity Analysis

Interpreting Excel Solver Sensitivity Report (cont.):

• The shadow price for a constraint is the change in the optimal objective
function value if the right-hand side of that constraint is increased by one.
– Nonbinding constraints will always have a shadow price of zero.
– Making a binding constraint more restrictive degrades or leaves
unchanged the optimal objective function value.
– Making a binding constraint less restrictive improves or leaves
unchanged the optimal objective function.
• Shadow prices are symmetric; the negative of the shadow price is the
change in the objective function for a decrease of 1 in the right-hand side.
 Sensitivity Analysis

Figure 12.11: Solver

Sensitivity Report for
the M&D Chemicals
Problem
 Sensitivity Analysis

Interpreting Excel Solver Sensitivity Report (cont.):

• The reduced cost for a decision variable is the shadow price of the
nonnegativity constraint for that variable.
• The Allowable Increase and Allowable Decrease indicate the change in the
objective function coefficient for which the current optimal solution will
remain optimal.
• The Excel Solver Sensitivity Report can provide useful information about the
sensitivity of the optimal solution to changes in the model input data.
 Sensitivity Analysis

Interpreting Excel Solver Sensitivity Report (cont.):

• Classical sensitivity analysis:
– Based on the assumption that only one piece of input data has
changed.
– It is assumed that all other parameters remain as stated in the
original problem.
• When interested in what would happen if two or more pieces of input data
are changed simultaneously:
– The easiest way to examine the effect of simultaneous changes is to
make the changes and rerun the model.
Investment Portfolio Selection
Transportation Planning
Maximizing Banner Ad Revenue

GENERAL LINEAR PROGRAMMING NOTATION

AND MORE EXAMPLES
 General Linear Programming Notation and More Examples

• The general notation for linear programs uses the letter x with a subscript.
• In the Par, Inc. problem the decision variables could be denoted as:

x1 number of standard bags.

x2 number of deluxe bags.

• Advantage: Formulating a mathematical model for a problem that

involves a large number of decision variables is much easier.
• Disadvantage: Not being able to easily identify what the decision
variables actually represent in the mathematical model.
 General Linear Programming Notation and More Examples

Illustration: Par, Inc. model using the general notation:

Max 10 x1  9 x2 7
x1  1x2 630 Cutting and dyeing
s.t. 10
1 5
x1  x2 600 Sewing
2 6
2
1x1  x2 708 Finishing
3
1 1
x1  x2 135 Inspection and packaging
10 4
x1 , x2 0
 General Linear Programming Notation and More Examples

Investment Portfolio Selection:

• Portfolio selection problems involve situations in which a financial
manager must select specific investments—for example, stocks and
bonds—from a variety of investment alternatives.
• Objective: Maximization of expected return or minimization of risk.
• Constraints: Restrictions on the type of permissible investments, state
laws, company policy, and so on.
 General Linear Programming Notation and More Examples

Illustration:
• Welte Mutual Funds, Inc., located in New York City, is looking for
investment opportunities for $100,000.
• The firm’s top financial analyst identified five investment opportunities
and projected their annual rates of return.
 General Linear Programming Notation and More Examples

Table 12.3: Investment Opportunities for Welte Mutual Funds

Projected
Rate of
Return
Investment (%)

Atlantic Oil 7.3

Pacific Oil 10.3

Midwest Steel 6.4

Huber Steel 7.5

Government
 General Linear Programming Notation and More Examples

Illustration: Welte Mutual Funds (cont.):

• Investment guidelines:
– Neither industry (oil or steel) should receive more than $50,000.
– Amount invested in government bonds should be at least 25% of the
steel industry investments.
– The investment in Pacific Oil, the high-return but high-risk investment,
cannot be more than 60% of the total oil industry investment.
 General Linear Programming Notation and More Examples

Illustration: Welte Mutual Funds (cont.):

• Define the following decision variables:
X1  dollars invested in Atlantic Oil
X2  dollars invested in Pacific Oil
X 3  dollars invested in Midwest Steel
X 4  dollars invested in Huber Steel
X 5  dollars invested in government bonds
• Specify the objective: Maximizing return.
Max 0.073 X1  0.103 X2  0.064 X 3  0.075 X 4  0.045 X 5
 General Linear Programming Notation and More Examples

Illustration: Welte Mutual Funds (cont.):

• Define the constraints:
Constraint 1: X1  X2  X 3  X 4  X 5 100,000
Constraint 2: X1  X2 50,000
X 3  X 4 50,000
Constraint 3: X 5 0.25  X 3  X 4 
Constraint 4: X2 0.60  X1  X2 
Nonnegativity constraints: X1 , X2 , X 3 , X 4 , X 5 0
 General Linear Programming Notation and More Examples

Illustration: Welte Mutual Funds (cont.):

• Linear programming model for the Welte Mutual Funds investment
problem:
 General Linear Programming Notation and More Examples

Figure 12.12: The Solution for the

Welte Mutual Funds Problem
 General Linear Programming Notation and More Examples

Transportation Planning:
• The transportation problem arises in planning for the distribution of
goods and services from several supply locations to several demand
locations.
• Quantity of goods available at each supply location (origin) is limited.
• Quantity of goods needed at each of several demand locations
(destinations) is known.
• Objective: Minimize the cost of shipping goods from the origins to the
destinations.
 General Linear Programming Notation and More Examples

Illustration: Foster Generators:

• Involves the transportation of a product from three plants to four
distribution centers.
• To determine how much of its production should be shipped from each
plant to each distribution center.
 General Linear Programming Notation and More Examples

Illustration: Foster Generators (cont.):

• Production capacities over the next three-month planning period for one type of generator:

Origin Plant Three-Month Production Capacity (Units)

1 Cleveland 5,000
2 Bedford 6,000
3 York 2,500
Total 13,500
 General Linear Programming Notation and More Examples

Illustration: Foster Generators (cont.):

• The three-month forecast of demand for the distribution centers:
Origin Plant Three-Month Production Capacity (Units)
1 Boston 6,000
2 Chicago 4,000
3 St. Louis 2,000
4 Lexington 1,500
Total 13,500
 General Linear Programming Notation and More Examples

Figure 12.13: The Network

Representation of the Foster
Generators Transportation
Problem
 General Linear Programming Notation and More Examples

Illustration: Foster Generators (cont.):

• Objective is to determine:
– Routes to be used.
– Quantity to be shipped via each route.
– Minimum total transportation cost.

Let xij = number of units shipped from origin i to destination j

where i 1, 2,  , m and j 1, 2,  , n.
 General Linear Programming Notation and More Examples

Table 12.4: Transportation Cost Per Unit for the Foster

Generators Transportation Problem ($)
Origin Boston Chicago St. Louis Lexington
Cleveland 3 2 7 6
Bedford 6 5 2 3
York 2 5 4 5
 General Linear Programming Notation and More Examples

Illustration: Foster Generators (cont.):

x11  x12  x13  x14 5,000 Cleveland supply
• Supply constraints:
x21  x22  x23  x24 6,000 Bedford supply
x31  x32  x33  x34 2,500 York supply

x11  x21  x31 6,000 Boston demand

x12  x22  x32 4,000 Chicago demand
x13  x23  x33 2,000 St. Louis demand

• Demand constraints: x14  x24  x34 1,500 Lexington demand

 General Linear Programming Notation and More Examples

Illustration: Foster Generators (cont.):

• A 12-variable, 7-constraint linear programming formulation of the Foster Generators
transportation problem:
 General Linear Programming Notation and More Examples

Figure 12.14: Spreadsheet Model and Solution for the Foster Generator Problem
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• Applications of linear programming to marketing include: advertising
campaign mix, marketing mix, and marketing research.
• Consider the Business, Science, and Sports sections of a large content
publisher, MHT.
• MHT has contracts for banner ads with five companies: Nile, Zstart,
Cheetah, Stride, and Stove.
• Expected click-through rates for each advertiser by section of their web
content is shown in Table 12.5.
 General Linear Programming Notation and More Examples

Table 12.5: Data for the MHT Banner Ad Allocation Problem

Click-through rates
Section Nile Zstar Cheetah Stride Stove
Business 0.0155 0.0265 0.0100 0.0170 0.0105
Science 0.0165 0.0110 0.0125 0.0265 0.0125
Sports 0.0145 0.0235 0.0190 0.0225 0.0160
Ads Upper Limit 2,000,000 2,000,000 1,800,000 2,000,000 2,000,000

Contact Impression Limits

Nile Zstar Cheetah Stride Stove
Lower Limit 1,500,000 1,000,000 1,100,000 1,000,000 1,500,000
Ads Upper Limit 2,000,000 2,000,000 1,800,000 2,000,000 2,000,000
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• For next Friday, MHT expects to have enough visitors to its website to
guarantee 2,550,000 visitors to its Business section; 2,150,000 to its
Science section; and 2,500,000 visitors to its Sports section.
• MHT receives $0.30 for each click-through achieved.
• Advertising planners need to determine, for next Friday, how to allocate
impressions for each of the five advertisers across the three sections of
the MHT web site.
• A linear programming model can be used to maximize revenue.
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• The decision variable is as follows:

where: i = 1 is Business j =1 is Nile

i = 2 is Science j = 2 is Zstar
i = 3 is Sports j = 3 is Cheetah
j = 4 is Stride
j = 5 is Stove
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• The objective function is to maximize expected revenue.
• Expected revenue from the three sections:
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• The objective function is the sum of the expected revenues generated from the three
sections:
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• Since each section is limited to the number of impressions that MHT can guarantee, we
have the following impression availability constraints:
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• Combining the previous
information with a lower and
upper limit on impressions
received, we have the following
linear program shown to the right.
• The spreadsheet model and an
optimal solution are shown in
Figure 12.15.
 General Linear Programming Notation and More Examples

Figure 12.15: A spreadsheet

model and the Solution to the
MHT Banner Ad Problem
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• The solution indicates that all available impressions should be used, and
the resulting revenue is $45,270.
• Nile impressions are split between the Business and Science sections.
• Stove impressions are split over the Science and Sports sections.
• Zstart, Cheetah, and Stride are allocated exclusively to the Business,
Sports, and Science sections, respectively.
• The sensitivity report for the MHT solution is shown in Figure 12.16.
 General Linear Programming Notation and More Examples

Figure 12.16: The Excel Sensitivity

Report for the MHT Banner Ad
Problem
 General Linear Programming Notation and More Examples

Maximizing Banner Ad Revenue

• The Excel Solver Sensitivity Report can provide useful information about
the sensitivity of the optimal solution to changes in the model input.
• This type of classical sensitivity analysis is somewhat limited:
– Assumes that only one piece of input data has changed
– Assumes that all other parameters remain as stated in the original problem.
• The easiest way to examine the effect of simultaneous changes is to
change the parameters in the model and rerun the model.
GENERATING AN ALTERNATIVE OPTIMAL
SOLUTION FOR A LINEAR PROGRAM
 Generating an Alternative Optimal Solution for a Linear Program

Illustration: Consider the Foster Generators transportation problem:

• From Figure 12.14, the optimal solution:
x11 1000, x12 4000, x13 0, x14 0
x21 2500, x22 0, x23 2000, x24 1500
x31 2500, x32 0, x33 0, x34 0

• Optimal cost: $39,500.

• For the revised model to be optimal, the solution must give a total cost of
$39,500.
 Generating an Alternative Optimal Solution for a Linear Program

Illustration: Foster Generators (cont.):

• From Figure 12.14:
x13 x14 x22 x32 x33 x34 0

• If the sum of these variables is maximized and if the optimal objective function
value of the revised problem is positive.
• A different feasible solution that is also optimal is found.
 Generating an Alternative Optimal Solution for a Linear Program

Illustration: Foster Generators

(cont.):
• Revised
model:
 Generating an Alternative Optimal Solution for a Linear Program

Table 12.6: An Alternative Optimal Solution to the Foster

Generators Transportation Problem
Total Cost = $39,500 Amount Shipped
To: Boston To: Chicago To: St. Louis To: Lexington Total
From: Cleveland 3,500 1,500 0 0 5,000
From: Bedford 0 2,500 2,000 1,500 6,000
From: York 2,500 0 0 0 2,500
Total 6,000 4,000 2,000 1,500
 Generating an Alternative Optimal Solution for a Linear Program

Illustration: Foster Generators (cont.):

• In the original solution (Figure 12.14):
– Boston distribution center is sourced from all three plants, whereas each of the other
distribution centers is sourced by one plant.
– The manager in Boston has to deal with three different plant managers, whereas each of
the other distribution center managers has only one plant manager.
• The alternative solution (Table 12.6) provides a more balanced solution:
– Managers in Boston and Chicago each deal with two plants, and those in St. Louis and
Lexington, which have lower total volumes, deal with only one plant.
– Because the alternative solution seems to be more equitable, it might be preferred.
• Both the solutions give a total cost of $39,500.
 Generating an Alternative Optimal Solution for a Linear Program

• General approach to find an alternative optimal solution to a linear program:

– Step 1: Solve the linear program.
– Step 2: Make a new objective function to be maximized; It is the
sum of those variables that were equal to zero in the solution from
Step 1.
– Step 3: Keep all the constraints from the original problem; add a
constraint that forces the original objective function to be equal to
the optimal objective function value from Step 1.
– Step 4: Solve the problem created in Steps 2 and 3; if the objective
function value is positive, an alternative optimal solution is found.