Introduction to Operations

Research
Chapter 3 Introduction to
Linear Programming

What is linear programming

p g
g
Linear
Linear programming provides methods
for allocating limited resources among
competing activities in an optimal way.
way
Linear: all mathematical functions are linear
Programming: involves the planning of
activities

Any problem
A
bl
whose
h
model
d l fit
fits th
the fformatt
for the linear programming model is a
linear programming problem.

Agenda
Chapter 3 Introduction to Linear Programming
3.1 Prototype Example
3 2 The Linear Programming Model
3.2
3.3 Assumptions of Linear Programming
3.4 Additional Examples
3.5 Some Case Studies
3.6 Displaying
p y g and Solvingg Linear Programming
g
g Models on a
Spreadsheet
3.7 Formulating Very Large Linear Programming Models
3.8 Conclusions

LP Example
p
Leo
needsresources?
to make someCompeting
money to payactivities?
his MBA
(a)
Limited
tuition in two years. Leo can participate in either or
(b)both
data into ai ltable
bOrganize
h off two entrepreneurial
ventures. H
He can choose
h

any number of shares (fractional or integer) in each

venture
venture.
Each share of Venture 1 needs $1,000 investment
and requires
q
Leo to work 3 hours p
per day.
y Each share
of Venture 1 can produce $20,000 profit.
Each share of Venture 2 requires Leo to work 1 hour
per day and provides an advance of $1
$1,000
000 (cant
(can t be
used as profit) per share in return for the work
commitment that accompanies shares of this venture.
E h share
Each
h
off V
Venture
t
2 can produce
d
$10
$10,000
000 profit.
fit
Leo has $1000 for investment and hes
he s willing to work
a maximum 7 hours per day.

LP Example
p
Gathering data

LP Example:
p formulate the model
Leo needs to decide how many shares of each
venture to take (decision variables)
x1 = number of shares of Venture 1
x2 = number of shares of Venture 2
Leo wants to maximize his total income
Z = total profit (in thousands of dollars) from
investing these two ventures

Subject to

money
time

Graphical solutionfeasible region

Graphical solutionfeasible region

g

Feasible
region

Graphical solution

Graphical solution
Drawing
D
i th
the liline:
20x1 + 10 x2 = 40
(1,2)

Let x1=0, get x2=4;

Let x2=0, get x1=2

Begin by trial and error to find

the maximum value of Z. For
example, try (1, 2) in the
feasible region to get:
20x1 + 10 x2 =40

Graphical solution
Now try a larger arbitrary value
of Z, say, Z=60=20x1+10x2
A segment of the line lies
within the feasible region,
g
so
that the maximum permissible
value of Z must be at least 60.
Notice the two lines are
parallel since the slope is fixed
fixed.
Z = 20x1 + 10 x2
x2 = -2x1 + 0.1Z

Graphical solution
Move the ruler with fixed slope
p
through the feasible region in the
direction of improving
p
g Z until it
passes through a point in the
feasible region.
g
This p
point is the
optimal solution.
In the example,
example Z reaches the
maximum value at (0, 7), Z=70
Leo should invest 7 shares of
venture 2 and 0 share of venture
1, with a total profit of $70,000

Optimal
solution

Graphical method for linear programming

Step 1determine the feasible region
Step 2
If the objective is maximize,
maximize move the ruler with fixed

slope through the feasible region in the direction of

improving Z until it passes through a point in the feasible
region. This point is the optimal solution.
If the objective is minimize, move the ruler with fixed
slope through the feasible region in the direction of
decreasing Z until it passes through a point in the
feasible region. This point is the optimal solution.

The g
graphic
p
method for linear p
programming
g
g can
be used to solve any linear programming
problem with 2 decision variables.

Exercise
(a) Formulate a linear programming model
for this problem.
((b)) Use the graphical
g p
method to solve this
model.

Exercise
The Apex Television Company has to decide on
the number of 27- and 20-inch sets to be
produced at one of its factories. Market research
indicates that at most 40 of the 27
27-inch
inch sets and
10 of the 20-inch sets can be sold per month.
The maximum number of work-hours
work hours available is
500 per month. A 27-inch set requires 20 workhours and a 20-inch set requires 10 work-hours.
Each 27-inch
27 inch set produces a profit of $120 and
each 20-inch set produces a profit of $80. A
wholesaler
o esa e has
as ag
agreed
eed to
o pu
purchase
c ase a
all the
e
television sets produced if the numbers do not
exceed the maxima indicated by the market
research.
research

Exercise
(a) Formulate a linear programming
model
d l for
f this
thi problem.
bl
Step 1: determine the decision variables
The decisions that need to be made are the number of
27-inch and 20-inch TV sets to be p
produced p
per month
by the Apex Television Company. Therefore, the
decision variables for the model are
x1 = number of 27-inch TV sets to be produced per
month,
x2 = number of 20-inch TV sets to be produced per
month.

Exercise
Step 2: determine the objective function
Z = total profit per month
Max Z = 120x1 + 80x2
Step 3: determine the resource constraints
Number of 27-inch sets sold per month: x1 40
Number of 20-inch sets sold per month: x2 10

Exercise
Step 4: construct the LP model
Maximize Z = 120x1 + 80x2
Subject to x1
40
x2
10
20 x1 + 10 x2 500
and x1 0 x2 0

Work-hours availability: 20 x1 + 10 x2 500.

Nonnegativity constraints on TV sets produced:

x1 0, x2 0

Exercise
(b) Use the graphical method to solve
this model.
Z=2400

The Linear Programming Model

A standard form of the modelP
modelP. 32
Max Z = CTX
S T AX B,
S.T.
B
X 0

Other forms of LP problems

Common terminology for the LP model:
The objective function: Z=c1x1+..+cnxn
Constraints
Parameters: cj, bi, aij

The Linear Programming

g
g Model

Linear programming assumptions

S
Solution
l ti Any
A specification
ifi ti off values
l
ffor th
the d
decision
i i
variables (xj)
Feasible solution A solution for which all constraints
are satisfied
Infeasible solution A solution for which at least one
constraint is violated
Feasible region
g
The collection of all feasible solutions
Optimal solution A feasible solution that has the most
favorable value of the objective function

Linear programming assumptions

Increasing
marginal
return
(economies
Case
of3scale)

Startup cost
Profit from door

x1(door)

Proportionality
satisfied

12

Decreasi
ng
marginal
3
return
t
5 (marketi
g cost))
6 ng

12

11

18

Case 1

Case 2

When the proportionality assumption does not

hold, nonlinear programming should be used.

Proportionality-The
Proportionality The contribution of each activity to Z
or a constraint is proportional to the level of activity xj