Linear Programming
Linear Programming
Section 3
Linear Programming
g g in Two Dimensions: A In this section,
section we will explore applications which utilize
Geometric Approach the graph of a system of linear inequalities.
We have seen this p problem before. An extra condition will be added to This is an example
p of a linear p programming
g gpproblem. Every y
make the example more interesting. linear programming problem has two components:
Suppose a manufacturer makes two types of skis: a trick ski and a 1. A linear objective function is to be maximized or minimized.
slalom ski. Each trick ski requires 8 hours of design work and 4 hours
of finishing. Each slalom ski requires 8 hours of design and 12 hours of In our case the objective function is Profit = 5x + 10y (5 dollars
finishing. Furthermore, the total number of hours allocated for design profit for each trick ski manufactured and $10 for every slalom
work is 160 and the total available hours for finishing work is 180. ski produced).
Finally, the number of trick skis produced must be less than or equal to
2. A collection of linear inequalities that must be satisfied
15 H
15. How many tricki k skis
ki andd how
h many slalom
l l skis
ki can be
b made d under
d
these conditions? simultaneously. These are called the constraints of the problem
Now, here is the twist: Suppose the profit on each trick ski is $5 and the because these inequalities give limitations on the values of x
profit for each slalom ski is $10. How many each of each type of ski and y. In our case, the linear inequalities are the constraints.
should the manufacturer produce to earn the greatest profit?
3 4
Linear Programming Problem
Constraints
(continued)
3. The feasible set is the set of all points that are possible
x and
d y have
h tto solutions.
solutions In this case,
case we want to determine the value(s) of x, x the
be positive number of trick skis and y, the number of slalom skis that will
The number of trick skis yield the maximum profit. Only certain points are eligible. Those
must be less than or are the points within the common region of intersection of the
equal to 15 Design constraint: 8 graphs of the constraining inequalities. Let’s return to the graph
of the system of linear inequalities. Notice that the feasible set is
hours to design
the yellow shaded region.
each trick ski and 8
O task
Our k iis to maximize
i i theh profit
fi function
f i P = 5x 5 + 10y
10 byb
hours to design
producing x trick skis and y slalom skis, but use only values of x
Finishing constraint: 4 hours each slalom ski. and y that are within the yellow region graphed in the next slide.
for each trick ski and 12 hours
for each slalom ski.
5 6
7 8
Maximizing the Profit
Constant-profit Lines (continued)
9 10
11 12
Geometric Method for Solving
Example 1
Linear Programming Problems
13 14
Example 1 Example 1
(no solution) (continued)
Maximize the quantity z = x + 2y subject to the constraints The pink lines are the graphs of z = x + 2y for z = 2, 3 and 4.
x + y ≥ 1,
1 x ≥ 0,0 y ≥ 0.
0 We can see from the graph there is no feasible point that makes
1. The objective function is z = x + 2y, which is to be maximized. z largest. The region is unbounded.
2. Graph the constraints: (see next slide) We conclude that this
3. Determine the feasible set (see next slide) linear programming
4. Determine the corner points of the feasible set. There are two problem has no solution.
corner points from our graph: (1,0) and (0,1)
5 Determine
5. D t i the
th value
l off the
th objective
bj ti function
f ti att eachh vertex.
t
At (1, 0), z = (1) + 2(0) = 1; at (0, 1), z = 0 + 2(1) = 2.
15 16
Example 2
Example 2
(unique solution)
A manufacturing plant makes two types of boats, a
1. Write the objective function.
p
two-person boat and a four-person
p boat. Each two-
person boat requires 0.9 labor-hours from the cutting 2.
2 Write
W it the
th problem
bl constraints
t i t andd the
th nonnegative
ti
department and 0.8 labor-hours from the assembly constraints.
department. Each four-person boat requires 1.8 3. Graph the feasible region. Find the corner points.
labor-hours from the cutting department and 1.2 4. Test the corner points in the objective function to find the
labor-hours from the assembly department. The
maximum profit.
maximum labor-hours available per month in the
cutting department and the assembly department are Answers: 1. Objective Function
864 and d 672
672, respectively.
ti l TheTh company makesk a
profit of $25 on each two-person boat and $40 on If x is the number of two-person boats and y is the number of
each four-person boat How many boats of each kind four-person boats, and the company makes a profit of $25 on
should the company produce in order to maximize
each two-person boat and $40 on each four-person boat, the
profit?
objective function is P = 25x + 40y.
17 18
Example 2 Example 2
(continued) (continued)
.8x + 1.2y = 672
2. Constraints 500
3. Feasible region
g
Since each 2-person boat requires 0.9 labor hours from the
The yellow shaded area
cutting department and each 4-person boat requires 1.8 hour
is the feasible region.
of cutting, and the maximum hours available in the cutting .9x + 1.8y = 864
department is 864, we have 0.9x + 1.8y < 864 Corner points are
(0, 0), (0, 480),
Since each 2-person boat requires 0.8 labor-hours from
(480, 240), and (840, 0)
assembly and each 44-person
person requires 1.2
1 2 hour from assembly,
assembly
and the maximum hours available in assembly is 672, we
have 0.8x + 1.2y < 672.
Also, x > 0 and y > 0.
19 20
Example 2
Example 3
(continued)
4. Maximum Profit
Maximize z = 4x + 2y subject to
2x + y < 20
Corner point Value of P = 25x + 40y
10x + y > 36
(0,0) 0 2x + 5y > 36
(0,480) 40(480) = $19,200 x, y > 0
Example 3 Example 3
(multiple solution) (continued)
Maximize z = 4x + 2y subject to
The feasible The solution to example
p 3 is a
2x + y < 20 36
10x+y=36 solution is shaded multiple optimal solution. In
green. The corner general, if two corner points are
10x + y > 36
2x+y=20 points are (2,16), both optimal solutions to a linear
2x + 5y > 36 (8,4) and (3,6). programming problem, then any
point on the line segment joining
x, y > 0 2x+5y=36 them is also an optimal solution.
10 Thus any point on the line
Test corner points: 2x + y = 20, where 2 < x < 8, such
(2,16) z = 4(2) + 2(16) = 40 These are both
(8,4) z = 4(8) + 2(4) = 40 } optimal.
as (3, 14), would be an optimal
solution.
(3,6) z = 4(3) + 2(6) = 2
23 24