Modelling with Linear

Programming
Chapter (2)
Part 02

1
 1- Finding the Feasible Region
• Use the axis in a 2-dimensional graph to represent the values that
the decision variables can take.
• For all constraints, replace the inequalities with equations and
graph the resulting straight lines on the 2-dimensional graph.
• For the inequality constraints, find the side (half-space) of the
graph meeting the original conditions.
• Find the intersections of all feasible regions defined by all the
constraints. 2
 Graphical Solution
of the Reddy Mikks Problem
Constraint 2 (C2): 2XE + XI  8 A solution is any specification of
values for the decision variables.
Constraint 3 (C3): -XE + XI  1
I A feasible solution is a solution
for which all of the constraints
are satisfied.
The feasible region is the set of
4

all feasible solutions.

Constraint 4 (C4): XI  2
Notice that the feasible region is
3

convex
2

Feasible Constraint 1 (C1): XE + 2XI  6

1

Region
1 2 3 4 5 6 7 E
Point (0, 0)
3
 2- Finding the Optimal Solution
• Determine the slope of the objective function:
 Select a convenient point in the feasible region.
 Draw the corresponding straight line.
• Determine the direction of improvement of the objective
function:
 Select a second point in the feasible region and evaluate
the objective function at that point.
• Follow the direction of improvement until reaching the
(corner) point beyond which any improvement of the objective
function would take you outside of the feasible region.
4
 The Graphical Solution
Take point (XE , XI ) = (2, 0):
Z=6
Take any point, such as:
(XE , XI ) = (3, 0):
Maximize z = 3XE + 2XI
Z=9 Z= 9
 I Z = 12
Z = 12.66
C2
C3
C1 Optimal point (Optimal Solution):
4 XE = 3.33
XI = 1.33
3

C4 How can we get the coordinates of this point?

2
1

1 2 3 4 5 6 7  E

Improving (Increasing) direction 5

Optimum Solution …
The optimum solution is attained at the intersection of
the two lines:
L1 : 2XE + XI = 8 Maximize z = 3XE + 2XI
L2 : XE + 2XI = 6

X* = (XE, XI) = (10/3, 4/3)

and Z = ?
* 12 2/3
6
Solution Procedure Summary for a Max
Problem
• Prepare a graph of the feasible solutions for each of the
constraints.
• Determine the feasible region that satisfies all the constraints
simultaneously.
• Draw an objective function line.
• Move parallel objective function lines toward larger objective
function values without entirely leaving the feasible region.
• Any feasible solution on the objective function line with the
largest value is an optimal solution.
How about a minimization problem?
7
 Team Exercise (5 Minutes)
 For the Reddy Mikks problem, find all the corner-point
feasible solutions.

 Suppose that another constraint is added to the problem :

XE + XI  1,
and the problem is changed from maximization to minimization.
For this new problem, find the new optimal solution.
8
 The Graphical Solution
Take point (XE , XI ) = (2, 0):
Z=6
Take any point, such as:
(XE , XI ) = (3, 0):
Minimize z = 3XE + 2XI
Z=9
 I
Newly added constraint:
C2 XE +XI
C3
New Constraint C1

4
3
Optimal point (Optimal Solution):
XE = 0 2 C4
XI = 1
1

Z=2
1 2 3 4 5 6 7  E
0+ 0 1
Improving (decreasing) direction 9
 Diet Problem
• A Farm uses at least 800 lb of special feed
daily. lb per lb of
feedstuff
• The special feed is a mixture of corn and
soybean meal with the following compositions: Cost
•At least 30% of the special feed are
Feedstuff Protein Fiber
$/lb
protein.
Corn 0.09 0.02 0.30
•At most 5% of the special feed are fiber.
Soybean 0.60 0.06 0.90
• Given the information in the table.
• Determine the daily minimum-cost feed mix.

10
 Diet Problem

• The decision variables.

• Objective function.

• Constraints.

11
 Diet Problem
Define:
x1 = number of lb of corn in the daily mix.
x2 = number of lb of soybean meal in the
daily mix.
Minimize z = 0.30x1 + 0.90x2
Subject to:
x1 + x2  800 (daily amount of special feed)
(protein in special feed)
0.21x1 – 0.30x2  0
(fiber in special feed)
0.03x1 – 0.01x2  0
(Non-negativity constraints)
x1 , x2  0
12
 Using Software to Solve LP Model
• In the textbook:
• Excel Solver.
• AMPL
• Many other tools: LINGO, GAMS, TORA, … etc.

13
Quiz 01 on Monday 5th February 2024.

• The questions will be similar to the problems in Chapter 2 up

to problem 34 on page 82 if you have Edition 10 (up to page 61
if you have Edition 9).

• Bring a ruler, dark pencils, eraser, sharpener and triangle.

14
 Assumptions of the LP Model
• Proportionality
• The contribution of any decision variable to the objective function
is proportional to its value.
• For example, in the Reddy Mikks Problem, the contribution of one
ton of interior paint is $2000, of two tons is $4000, and so on.
• In many cases, a volume discount is available whereby the price per
ton goes down as bulk quantity is purchased.
• Such discounts are often nonlinear which means that a linear
programming model is either inappropriate or is really just an
approximation of the real world.
15
 Assumptions of the LP Model
• Additivity
• The contribution to the objective function for any variable is
independent of the other decision variables.
• For example, in the Reddy Mikks problem, the production of one
ton of interior paint will always contribute to $2000 regardless of
the amount of exterior paint produced.
• Same goes for the constraints.
• 2 tons of raw material A are consumed to produce 1 ton of interior
paint, regardless of how many tons of either paints are produced.
• Proportionality and Additivity are also implied by the linear
constraints. 16
 Assumptions of the LP Model
• Divisibility
• The LP model assumes that the decision variables can take fractional values.
• For example, in bus scheduling problem, it allows for a solution to run 0.8
buses at a certain time of day.
• In many situations, the LP is being used on a large-enough scale that one can
round the optimal decision variables up or down to the nearest integer and get
an answer that is feasible and reasonably close to the optimal integer solution.
• Divisibility also implies that the decision variables can take on the full range of
real values.
• For large problems, most likely, rounding or truncating the optimal LP
decision variables will not greatly affect the solution.
17
 Assumptions of the LP Model
• Certainty
• The LP model assumes that all the constant terms, objective function
and constraint coefficients as well as the right-hand side (RHS)
parameters, are known with absolute certainty and will not change
(deterministic).
• If the values of some of these quantities are not known with certainty
(uncertain or stochastic), for example, if the demand forecast data or
price data given in the Reddy Mikks problem are not 100% accurate,
then this assumption is violated.

18
More Applications of Linear Programming

19
 Reading Assignment

Read Chapter 02 in the textbook.

20
 References
• Chapter 2 from H. Taha, “Operations Research: An Introduction”, 10th
Edition, 2018.
• Chapter 3 from F. Hillier and G. Liberman, “Introduction to Operations
Research”, 10th Edition, 2014.
• Chapter 3 from W. L. Winston, “Operations Research: Applications and
Algorithms”, 4th edition, 2003.

21