D AT E | M O N T H | Y E A R BU275: Business Decision

Presented by
First Name Last Name Models
Instructor – Dr Jason Hurley
WINTER 2023
Introduction to Linear Programming
D AT E | M O N T H | Y E A R

Presented by
First Name Last Name
Deterministic Models and Algorithms

• The course this far: probabilities played a big role in our model building

• Last part of the course: everything is deterministic

• With the computational power now available in the modern world, there are
D AT E | M O N T H | Y E A R

algorithms available for solving many different business/personal problems

Presented by
First Name Last Name
• Linear Programming Problems – use an algorithm that helps solve a certain
class of problems
• Excel Solver has an algorithm to solve linear programming problems
Linear Programming: The Big Picture

Real World Formulate Solution to the Solution to the

Mathematical Modeling

Solution Algorithm

Interpret and Validate

Problem Problem Mathematical Real World
Mathematically Problem Problem
D AT E | M O N T H | Y E A R

Presented by
First Name Last Name
Linear Programming

• Make simultaneous decisions to reach an optimal output by considering

constraining factors (e.g., limited/scarce resources)

• Planning of activities represented by linear mathematical model

D AT E | M O N T H | Y E A R

Presented by
• Categories of LP
First Name Last problems
Name include:
• Resource-allocation
• Cost-benefit tradeoff
• Product mixing
• Transportation
• Assignment
Linear Programming Formulation

Step 1: Determine Decision Variables

D AT E | M O N T H | Y E A R

Presented by
Step 2: Determine Objective Function
First Name Last Name

Step 3: Determine Constraints

Basic Example of Linear Programming

Lego Company

• Build and sell Chairs (Profit: $7) and Tables (Profit $8)
• Chair: requires 3 large blocks, 2 medium blocks, 2 small blocks
• Table:
D AT Erequires
| M O N T H | Y2
E Alarge
R blocks, 2 medium blocks, 4 small blocks
Presented by
First Name Last Name
Determine the product mix (i.e., how many chairs and tables) to
maximize profit if there are only 12 large blocks, 8 medium blocks, and
10 small blocks available.
Step 1: Decision Variables

• The best way to determine the decision variables is to put

yourself in the role of the decision maker and ask the question:
“What factors (decisions) can I change (control)?”
D AT E | M O N T H | Y E A R

Presented by
• Define Decision
First Name Last Name Variables:

Let C = # of chairs
T = # of tables
Step 2: Objective Function

Specify the problem objective: “What is the goal of changing the

decision variables?”
D AT E | M O N T H | Y E A R

• Define
Presented by
First NameObjective
Last Name Function:

Maximize Profit (Z) = 7C + 8T

Step 3: Constraints
Specify the limitations: “What are the constraining factors that I
cannot change?”
• Define Constraints:
• There are only 12 large blocks
D AT 3C +O2T
E | M N T H≤| 12
YEAR

Presented
• There areby
only 8 medium blocks
First Name Last Name
2C + 2T ≤ 8
• There are only 10 small blocks
2C + 4T ≤ 10
• Implicit Constraints (e.g., nonnegativity – cannot produce negative chairs, tables)
C, T ≥ 0
LP Formulation

• Decision Variables: Let C = # of chairs, T = # of tables

• Objective Function: Max Z = 7C + 8T

D AT E | M O N T H | Y E A R
• Subject to:
Presented by
3CName
First + 2TLast<= 12
Name (large block constraint)
2C + 2T <= 8 (medium block constraint)
2C + 4T <= 10 (small block constraint)
C, T > = 0 (nonnegativity constraint)
Characteristics of LPs

• Objective function and constraints are linear functions

• Constraint types can be inequalities ( or ) and equalities (=)

D AT E | M O N T H | Y E A R

• Maximize
Presented by
or Minimize a single objective
First Name Last Name

• Decision variables can assume any fractional value

• Decision variables are all non-negative

Verifying LPs

• Decision Variables:
• Make sure you have specified units of the decision variables

• Objective Function:
• Make sure that the only variables in the objective function are the decision variables
• If you have variables in the objective that are not defined as decision variables, they should be defined
D AT E | M O N T H | Y E A R
as decision variables
• MakePresented
sure youby have stated whether it is a maximization or minimization problem
First Name Last Name
• Constraints:
• Make sure that the only variables in the constraints are the decision variables
• If you have variables in the constraints that are not defined as decision variables, they should be
defined as decision variables
• Make sure that the units on the LHS and RHS of constraints match
• Make sure that both explicit and implicit constraints are written
Solution Methods

Linear Programming Solution Methods:

1. Graphical
2. Simplex
D AT E | M O N T H | Y E A R

Presented by
First Name Last Name
We will use Excel Solver (add-in) to solve problems that have
more than two decision variables

• Excel Solver uses the Simplex Method

Method 1: Graphical
Step 1: Determine the feasible region
• Plot a constraint line for each constraint
• For each constraint line, determine the feasible side
• Identify the set of solutions that satisfy all the constraints

Step 2:
• Determine
D AT E | the
M O Ncorner
T H | Y Epoints
AR of the feasible region
• Evaluate the objective function at each corner point and choose the corner point that maximizes
Presented by
(minimizes)
First Name theLast
objective
Name function

Step 2 (alternate approach)

• Identify an arbitrary objective function
• For a maximization (minimization) problem, move the arbitrary objective function away from
(towards) (0,0) until further movement would take the line completely out of the feasible region
Graphical Method - Step 1: Plot Constraint 1

First point: C=0, T=6

3C + 2T ≤ 12 3C + 2T = 12
Second point: T=0, C=4

T
D AT E | M O N T H | Y E A R
3C + 2T < 12
Presented by
First Name Last Name

C
Graphical Method - Step 1: Plot Constraint 2

T 3C + 2T < 12
2C + 2T < 8
3C + 2T = 12
D AT E | M O N T H | Y E A R

Presented by
First Name Last Name

2C + 2T =8

C
Graphical Method - Step 1: Plot Constraint 3

T 3C + 2T < 12
2C + 2T < 8
3C + 2T = 12 2C + 4T < 10

D AT E | M O N T H | Y E A R
2C + 2T = 8
Presented by
First Name Last Name

2C + 4T = 10

C
Graphical Method - Step 1: Determine Feasible
Region
Feasible (shaded) region satisfies all the constraints
T simultaneously (i.e., collection of all feasible solutions)

Constraints
D AT E | M O N T H | Y E A R 3C + 2T < 12
2C + 2T < 8
Presented by 2C + 4T < 10
First Name Last Name

C
Graphical Method - Step 2: Corner Points

• The Corner Point Method uses the following property of a linear

programming problem which has been proven mathematically:

• TheD AToptimal solution of a linear programming will always occur at a

E | MONTH | YEAR

Presented by
corner point
First Name Lastof the feasible region
Name

• There can be more than one corner point which provides an

optimal solution
Graphical Method - Step 2: Corner Points

Step 1: Determine the feasible region

Step 2-1: Identify all feasible corner points

D AT E | M O N T H | Y E A R

Step 2-2: Substitute

Presented by the corner points into the objective function formula
First Name Last Name

The optimal solution(s) is the corner point(s) with the best value of objective
function
Graphical Method - Step 2: Corner Points

T Max Z = 7C + 8T

Subject to:
3C + 2T < 12
2C + 2T < 8
OFV at (0,2.5) = 7(0) + 8(2.5) = 20 2C + 4T < 10

.
D AT E | M O N T H | Y E A R
C, T > 0
Presented by

..
First Name Last Name OFV at (3,1)= 7(3) + 8(1) = 29

.
(3,1) Optimal solution
C
(0,0) (4,0)
OFV at (0,0) = 7(0) + 8(0) = 0 OFV at (4,0) = 7(4) + 8(0) = 28
Graphical Method - Step 2: Arbitrary Objective
Function
T Max Z = 7C + 8T

Subject to:
3C + 2T < 12
7C + 8T = ? (an arbitrary value) 2C + 2T < 8
D AT E | M O N T H | Y E A R 2C + 4T < 10
Presented by C, T > 0
First Name Last Name

C
Graphical Method - Step 2: Arbitrary Objective
Function
Optimal solution is at (3,1)
Max Z = 7C + 8T
Hence make 3 chairs and 1 table
Subject to:
3C + 2T < 12
2C + 2T < 8
D AT E | M O N T H | Y E A R Objective function value =
2C + 4T < 10
Presented by 7(3) + 8(1) = 29 C, T > 0
First Name Last Name

.
(3,1)
Profit & Gambit Co. Example: Minimization
Problem
Management at The Profit & Gambit Co. has decided to
undertake a major advertising campaign that will focus on the
following three key products:
• A spray prewash stain remover
• A liquid laundry detergent
D AT E | M O N T H | Y E A R

Presented by
• A powder
First Name Last laundry
Name detergent

The campaign will use both television and print media.

The general goal is to increase sales of these products.

Profit & Gambit Co. Example: Minimization
Problem
Increase in sales for each unit of advertising:

Increase in Sales per Unit of Advertising

Product Television Print Media

Stain Remover 0% 1%
Liquid Detergent 3% 2%
D AT E | M O N T HPowder
| Y E ADetergent
R -1% 4%
Unit Cost ($millions) 1 2
Presented by
First Name Last Name
Management has set the following goals for the campaign:
• Sales of the stain remover should increase by at least 3%
• Sales of the liquid detergent should increase by at least 18%
• Sales of the powder detergent should increase by at least 4%

How much should they advertise in each medium to meet the sales goals at a
minimum total cost?
Profit & Gambit Co. Example: Minimization
Problem
Decision Variables:
Let TV = the number of units of advertising on Television
PM = the number of units of advertising in Print Media

Minimize Z = TV + 2PM (in millions of dollars)

D AT E | M O N T H | Y E A R

Presented by
First Name Last Name

Subject to:
Stain remover increased sales: PM ≥ 3
Liquid detergent increased sales: 3TV + 2PM ≥ 18
Powder detergent increased sales: –TV + 4PM ≥ 4
Non-negativity constraints: TV ≥ 0, PM ≥ 0
Profit & Gambit Co. Example: Minimization
Problem
Amount of print media advertising
PM

Feasible
10
region

8
D AT E | M O N T H | Y E A R

Presented by 6
First Name Last Name

PM = 3
2
-TV + 4 PM = 4
3 TV + 2 PM = 18

-4 -2 0 2 4 6 8 10 TV
Amount of TV advertising
Profit & Gambit Co. Example: Minimization
Problem

D AT E | M O N T H | Y E A R

Presented by
First Name Last Name
 Thanks for your attention!

D AT E | M O N T H | Y E A R

Presented by
Next class:
First Name Last Name
• Linear Programming (part 2)