Chapte 3
Chapte 3
Course
Operations Research
Gebreyohannes G.
11/14/22
Chapter Two
Linear Programming
Definition:
Linear programming is a technique applied for
the optimizing a linear objective function, subject to linear
equality and linear inequality constraints.
Is a particular technique used for economic allocation of
‘scarce’ or ‘limited’ resources.
11/14/22
Cont…
e.g. labour, material, machine, time,
warehouse space, capital, energy, etc.:
several competing activities e.g. products,
services, jobs, new equipment, projects, etc. on
the basis of a given criterion of optimality.
11/14/22
Components of LPP
The linear programming model consists of the
following components:
Decision variables: mathematical symbols representing levels
of activity of an operation.
Objective function: a linear relationship reflecting the
objective of an operation.
Constraint: a linear relationship representing a restriction on
decision making.
Non-negativity: values of the decision variables are zero or
non-zero positives.
11/14/22
General form of LP
Z = c1x1 + c2x2 +…………. +cnxn
Subject to restrictions
a11x1 + a12x2 + …..........+a1nxn (≤ or ≥) b1
a21x1 + a22x2 + …………..+a2nxn (≤ or ≥) b2
.
Contribution to Z per
11/14/22
Linear Programming Model Construction
General steps taken are as under:
Define the decision variables .
Identify the coefficient of each decision variables.
Formulate the objective function.
Identify the coefficients of each decision variables in the obj fun.
Identify the constrained resources or RHS values.
Formulate the suitable mathematical constraints related to each
respective resource.
Mention the non-negativity constraints associated with the
decision variables.
11/14/22
Assumptions of Linear Programming
Proportionality
Additive
Divisibility
Deterministic
Finiteness
Optimality
One objective
11/14/22
Profit Maximization case
Example 1: Problem Definition
1. A company manufactures two products, tables and chairs by
using three machines A, B, and C so as to maximize a profit.
Machine A has 10 hrs of capacity available during the coming
week. Similarly, the available capacity of machine B and C
during the coming week is 36 hrs and 30 hrs, respectively.
One unit of a table requires 2 hrs of machine A, 3 hrs of
machine B and 3 hrs of machine C. Similarly, one unit of a
chair requires 1 hr, 12 hrs and 6 hrs of machine A, B, and C
respectively. When one unit of table is produced and sold, it
yields a profit of Birr 5 and that of a chair is Birr 7. Given the
above information, formulate the linear programming
model.
11/14/22
Solution
I. Decision variable
how tables and chairs can the managers produce?
the quantities to be produced can be denoted as:
X1: number of tables to be produced
X2: number of chairs to be produced
II. The Objective function
It is to max profit (sum of each)
Then total profit = $5x1+ 7x2 i.e.
11/14/22
Cont…
III. Model Constraints
three resources are used for production are hours of
machine A, B, and C respectively.
hours of machine A constraint is 2x1+ 1x2 ≤ 10
hours of machine B constraint is 3x1 + 12x2 ≤36
hours of machine C constraint is 3x1 + 6x2 ≤ 30
IV. Non-negativity constraint:
x1 ≥ 0, x2 ≥ 0.
11/14/22
Cont. . .
The complete linear programming model for
this problem can now be summarized bellow as
Max Z = $5x1+ 7x2
ST: 2x1 + 1x2 ≤ 10
5x1 + 2x2 ≤ 36
3x1 + 6x2 ≤ 30
x1, x2 ≥ 0
11/14/22
Example 2: Problem Definition
• Example1: Assume ‘XY’ company is a small craft operation run by a
Native American tribal council. The company employs skilled artisans
to produce clay bowls and mugs with authentic Native American
designs and colours. The company also employed inspectors who will
check the products after finished. The three primary resources used
by the company are special pottery clay, skilled labour and inspectors.
Given these limited resources, the company desires to know how
many bowls and mugs to produce each day in order to max profit.
The two products have the following resource requirements for
production and profit per item produced (i.e. the model parameters).
11/14/22
Example 1: Problem Definition . . .
Resource requirement
Product Labor Inspector Clay Profit
Hr/unit Hr/Unit Ib/unit $/unit
Bowl 2 1 5 50
Mug 2 4 2 35
There are 40 hours of labour 50 hours of inspector and 60 pounds of
clay available each day for production.
Required: Formulate the LP model of this problem.
11/14/22
Example 1: Problem Definition . . .
The final complete LP model
Max Z = $50x1 + 35x2
ST: 2x1 + 2x2 ≤ 40
5x1 + 2x2 ≤ 50
x1 + 4x2 ≤ 60
x1, x2 ≥ 0
11/14/22
Exercise 1
A firm manufactures two types of products A and B and sells them at a profit of $2 on
type A and $3 on type B. Each product is processed on two machines G and H. Type A
requires 1 minute of processing time on G and 2 minutes on H; type B requires 1
minute on G and 1 minute on H. The machine G is available for not more than 6 hours
40 minutes while machine H is available for 10 hours during any working day.
Required: a) Formulate the problem as a linear programming problem
b) Solve the problem using graphical method
Answer for que a)
Maximize Z = 2x1 + 3x2
ST X1 + x2 ≤ 400
2x1 + 3x2 ≤ 600
X1 ≥ 0, x2 ≥ 0
Cost Minimization case – Example 1 (sec 1 & 2)
A company owns 2 oil mills A and B which have
different production capacities for low, high and
medium grade oil. The company enters into a
contract to supply oil to a firm every week with 12, 8,
24 barrels of each grade respectively. It costs the
company $1000 and $800 per day to run the mills A
and B. On a day A produces 6, 2, 4 barrels of each
grade and B produces 2, 2, 12 barrels of each grade.
Required: Develop the LP model.
11/14/22
Solution
Let x1 be the number of days a week the mill A has to work
x2 be the number of days per week the mill B has to work
Low 6 2 12
High 2 2 8
Medium 4 12 24
11/14/22
Graphical Method
A graphical approach is limited to linear
programming problems with only two decision
variables.
The complexity of its application increases as
the number of constraint functions increases.
It is a relatively straightforward method for
determining the optimal solution to certain
linear programming problems.
11/14/22
The following are the steps to be followed in solving LP
problems graphically:
11/14/22
Cont…
Plot the objective function then move the line out from the
origin till it touches the extreme maximum points of the
feasible solution area.
Plot the objective function then move the line down till it
touches the extreme minimum point of the feasible solution
area.
Substitute these values into the objective function to find the
set of values that result in the maximum profit or the
minimum cost.
Interpreting the results
11/14/22
Cont. . .
Case 1: Graphical Solution Method for Maximization Problem
Example 1: Max Z = $50x1+ 35x2
ST: 2x1 + 2x2 ≤ 40
5x1 + 2x2 ≤ 50
x1 + 4x2 ≤ 60
x1, x2 ≥ 0
Required: Compute the optimal solution using graphical
method. Apply all possible steps.
11/14/22
Solution:
First develop the problem into mathematical linear
programing if not given.
The structured model is given so we need to proceed
into change the inequalities into equations for
computing the coordinate points.
2x1 + 2x2 = 40
5x1 + 2x2 = 50
x1 + 4x2 = 60
11/14/22
Cont. . .
2x1 + 2x2 = 40 5x1 +2x2 = 50 x1 + 4x2 = 60
x1 0 10 x1 0 60
x1 0 20
x2 25 0 x2 15 0
x2 20 0
11/14/22
Cont…
In this graph, the FSA is represented by the shaded
region.
The feasible region has four corner points i.e. O, A, B,
and C.
From these corner points at least one will give us
maximum profit
The one that generate the highest profit would be
regarded as optimal corner point/s.
11/14/22
Cont…
Corner Coordinate Profits
points Points
O (0,0) Z = $50(0)+ 35(0) = $0
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
Interpretation
Form this result; we can conclude that the
manager has different alternatives to make
decisions but since the objective of the
problem is to maximize profit, she has to
select the best alternative i.e. alternative
corner point B at x1 = 40/9 and x2 = 125/9
11/14/22
Case 2: Graphical Solution Methods for Minimization
Problem
Example 1:
Minimize Z = 1000x1 + 800x2
Subject to
6x1 + 2x2 ≥ 12
2x1 + 2x2 ≥ 8
4x1 +12x2 ≥ 24
X1 ≥ 0, x2 ≥ 0
11/14/22
Cont…
Required:
A.Sketch the graph and identify the feasible solution
area
B.How many numbers of products type x1 and type x2
the company has to produce to minimize cost?
C. What is the z value of the problem?
D.Does the problem have unique solution? Why?
11/14/22
Ans for A.
To sketch the graph, determine the CPs of each constraint by
changing the ≥ into =
6x1 + 2x2 = 12
Let x1 = 0, x2 = 6 coordinate points = (0, 6)
Let x2 = 0, x1 = 2 coordinate points = (2, 0)
2x1 + 2x2 = 8
Let x1 = 0, x2 = 4 coordinate points = (0, 4)
Let x2 = 0, x1 = 4 coordinate points = (4, 0)
4x1 +12x2 = 24
•Let x1 = 0, x2 = 2 coordinate points = (0, 2)
•
11/14/22
Cont…
11/14/22
Ans for B.
CP Coordinates Profits
11/14/22
Mill B has to work 3 days a week
Ans for C.
11/14/22
Graphical Solution of Irregular Types of Linear
Programming Problems:
11/14/22
A. Multiple optimal solutions
If the objective function is optimized at two adjacent
corner points of the feasible area, it is optimized at
every point on the line segment joining these corner
points, in which case there are infinitely many solutions
to the problem.
Multiple optimal solutions can benefit the decision
maker since the number of decision options is enlarged.
Multiple optimal solutions provide greater flexibility
to the decision maker.
11/14/22
Con…
Example 1:
Max Z = 4x1 + 3x2
Subject to
4x1+ 3x2 ≤ 24
x1 ≤ 4.5
x2 ≤ 6
x1 ≥ 0, x2 ≥ 0
11/14/22
Solution
The first constraint 4x1+ 3x2 ≤ 24, written in a form of
equation
4x1+ 3x2 = 24
Put x1 =0, then x2 = 8
Put x2 =0, then x1 = 6
The coordinates are (0, 8) and (6, 0)
11/14/22
Cont…
The second constraint x1 ≤ 4.5, written in a form of
equation
x1 = 4.5
The third constraint x2 ≤ 6, written in a form of
equation
x2 = 6
11/14/22
Cont…
11/14/22
Cont…
The corner points of feasible region are A, B, C and D.
So the coordinates for the corner points are
A (0, 6)
B (1.5, 6) (Solve the two equations 4x1+ 3x2 = 24 and x2
= 6 to get the coordinates)
C (4.5, 2) (Solve the two equations 4x1+ 3x2 = 24 and x1
= 4.5 to get the coordinates)
D (4.5, 0)
11/14/22
Cont…
We know that Max Z = 4x1 + 3x2
At A (0, 6)
Z = 4(0) + 3(6) = 18
At B (1.5, 6)
Z = 4(1.5) + 3(6) = 24
At C (4.5, 2)
Z = 4(4.5) + 3(2) = 24
At D (4.5, 0)
Z = 4(4.5) + 3(0) = 18
11/14/22
Cont…
11/14/22
B. No Optimal Solution/infeasible solution
An infeasible solution has no feasible solution
area; every possible solution point violates one
or more constraints. In other words, if a
solution violates at least one constraint, the
solution is said to be infeasible.
11/14/22
Cont . . .
Example1: Solve graphically
Max Z = 3x1 + 2x2
Subject to
x1+ x2 ≤ 1
x1+ x2 ≥ 3
x1 ≥ 0 , x 2 ≥ 0
11/14/22
Cont…
Solution
The first constraint x1+ x2 ≤ 1 is written in the
form of x1+ x2 = 1
Put x1 =0, then x2 = 1
Put x2 =0, then x1 = 1
The coordinates are (0, 1) and (1, 0)
11/14/22
Cont…
The second constraint x1+ x2 ≥ 3, written in a form of
equation
x1+ x2 = 3
Put x1 =0, then x2 = 3
Put x2 =0, then x1 = 3
The coordinates are (0, 3) and (3, 0)
11/14/22
Cont…
11/14/22
Cont…
11/14/22
C. Unbounded Solution
In an unbounded problem, the feasible solution area
formed by the model constraints is not closed.
In unbounded problem the objective function can
increase indefinitely without reaching a maximum
value.
Unlimited profits are not possible in the real world;
and an unbounded solution, like an infeasible
solution, typically reflects an error in defining the
problem or in formulating the model.
11/14/22
Cont…
Example1: Solve by graphical method
Max Z = 3x1 + 5x2
Subject to
2x1+ x2 ≥ 7
x1+ x 2 ≥ 6
x1+ 3x2 ≥ 9
x1 ≥ 0 , x 2 ≥ 0
11/14/22
Cont…
Solution
The first constraint 2x1+ x2 ≥ 7, written in a form of
equation
2x1+ x2 = 7
Put x1 =0, then x2 = 7
Put x2 =0, then x1 = 3.5
The coordinates are (0, 7) and (3.5, 0)
11/14/22
Cont…
The second constraint x1+ x2 ≥ 6, written in a
form of equation
x1+ x2 = 6
Put x1 =0, then x2 = 6
Put x2 =0, then x1 = 6
The coordinates are (0, 6) and (6, 0)
11/14/22
Cont…
The third constraint x1+ 3x2 ≥ 9, written in a
form of equation
x1+ 3x2 = 9
Put x1 =0, then x2 = 3
Put x2 =0, then x1 = 9
The coordinates are (0, 3) and (9, 0)
11/14/22
Cont…
11/14/22
Cont…
The corner points of feasible region are A, B, C and D. So the
coordinates for the corner points are:
A (0, 7)
B (1, 5) (Solve the two equations 2x1+ x2 = 7 and x1+ x2 = 6 to
get the coordinates)
C (4.5, 1.5) (Solve the two equations x1+ x2 = 6 and x1+ 3x2 =
9 to get the coordinates)
D (9, 0)
11/14/22
D. Redundancy
A redundant constraint is simply one that does not
affect the feasible solution region.
11/14/22
Con…
Example 1:
Maximize Z = X + 2Y
Subject to: X + Y ≤ 12
2X + Y ≤ 30
X ≤ 25
X, Y ≥ 0
11/14/22
2. Simplex Solution Methods
This technique is used to alleviate the limitation of
the graphical solution method.
In the simplex method, the search usually starts at
the origin and moves to that adjacent corner that
increases (for maximization) or decreases (for
minimization) the value of the objective function.
The process continues until no further improvement
is possible. In each of the iteration, the objective
function improves.
11/14/22
Basic Terms Involved in Simplex Procedure
1. Standard Form: when all constraints are written as
equalities it is called standard form
2. Utilization of Resources
I. Slack Variable: A Variable added to the left-hand side of a
‘less-than or equal to’ constraint to convert the constraint into
an equality is called a slack variable.
II. Surplus variable: A variable subtracted from the left hand
side of the “greater than or equal to” constraints to convert the
constraints into equality is called a surplus variable.
11/14/22
Cont…
3. Basic variable: for m simultaneous linear
equations in n variables(n>m), a solution
obtained by setting (n-m) variables equal to zero
and solving for the remaining m variables is called
a basic variables
4. Basic feasible solution: a basic feasible solution
is a basic solution for which the variables have a
value of greater than or equal zero.
11/14/22
Cont…
11/14/22
Cont…
8. Cj – Zj Row: The row containing the net profit that will
result from introducing one unit of the variable indicated in
that column in the solution numbers in index rows are also
known as shadow prices.
11/14/22
Cont…
10. Pivot row: the row corresponding to the variable that will leave the
basis in order to make room for the entering variable.
Outgoing Variable
To determine the outgoing variable, compute the ratio of the
Quantity to the coefficient of the incoming variable for each basis row.
For both the maximization and minimization problems, the outgoing
variable is the basic variable with the smallest ratio.
The coefficient of the incoming variable in the outgoing row is called
the pivot element.
11/14/22
Cont…
11. Pivot Number: The element at the
intersection of the pivot row and pivot column
12. Iteration: An iteration of the simplex method
consists of the sequence of steps performed in
moving from one basic feasible solution to
another.
11/14/22
Cont…
The procedures of Simplex Method
I.Standardize the problem.
II.Generate an initial Solution.
III.Test for optimality if in case the solution is optimal
Otherwise, go to Step 4.
IV.Identify the incoming and outgoing variables.
V.Generate an improved solution. Go to Step 3.
VI.Check for other optimal solutions.
11/14/22
Case 1: Simplex Method for the Maximization Problem
Example1: Max z= 40x1 +30x2
S.T x1 + 2x2 ≤ 40 hour of labour
4x1 + 3x2 ≤ 120 Ib of clay
X1, x2 ≥ 0
Required:
Standardize the model
Generate an initial solution and the initial simplex tableau
Determine the value of x1 and x2 at the optimal simplex tableau
What is the z-value at the optimal solution?
Does the problem have a multiple solution? Why?
11/14/22
Solution:
A.
Convert the model into standard form by adding slack variables to
each constraint as follows.
N.B: slack variables are added to ≤ constraints and represent unused
resources.
Max Z = Max z= 40x1 +30x2+0s1 + 0s2
S.T x1 + 2x2 +s1 +0s2 = 40
4x1 + 3x2 + 0s1+ s2 =120
X1, x2,s1, s2 ≥ 0
11/14/22
Cont…
B.
At the origin where nothing is being produced, the values of all the
decision variables are zero which implies the value of the slack variable
equals to the right hand side value.
1(0) + 2(0) +s1+ 0S2 = 40
S1= 40 -------------------------------------------------- 1
4(0) + 3(0) + 0S1 + s2 = 120
S2 = 120------------------------------------------------ 2
N.B: at the origin all decision variables are non-basic whereas all slack are
basic.
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
The optimal solution has been reached because all values of
Cj-zj row are less than and
11/14/22
Cont…
E.
Multiple optimal solutions on a simplex tableau can be determined
from Cj- Zj or Zi-Cj row value. An alternative optimal solutions have
the same z value but different variable values.
The answer for the question is yes because optimal simplex tableau
is said to be multiple if at least one non-basic variable has zero value
in the Cj-Zi or Zj-Cj row.
Now you can see the above table, x2 is non-basic variable but its
value corresponds to Cj-zj row is zero.
Alternative optimal solution is determined by selecting the non-
basic variable with Cj-zj = 0 as the entering or incoming variable.
11/14/22
Cont…
11/14/22
Case 2: Simplex Tableau for minimization problem
Example 1: Min Z = 45x1 + 12x2
St: x1 +x2 ≥ 300
x1 ≥ 250
x1, x2 ≥ 0
Required:
•Standardize the model
•Generate an initial solution and the initial simplex tableau.
•Determine the value of x1 and x2 at the optimal simplex tableau
•What is the z-value at the optimal solution?
•Does the problem have a multiple solution? Why?
11/14/22
Solution
A
Surplus variable:
A variable subtracted from the left hand side of the greater
than or equal to constraints to convert the constraints into
equality is called a surplus variable.
Min Z = 45x1 + 12x2
St: x1 +x2 – s1 = 300
x1- s2 = 250
x 1, x 2 , s1, s 2 ≥ 0
11/14/22
Cont…
Surplus variable represents the extra amount of that exceeds the
minimum requirement specified in the constraints. However, the
simplex method requires that the initial basic feasible solution be at
the origin, where x1= 0 and x2 = 0.
Testing these solution values, we have
x1 +x2 – s1= 300
s1 = -300------------------------------------------------------------------- 1
x1- s2 = 250
s2 = -250------------------------------------------------------------------- 2
11/14/22
Cont…
11/14/22
Cont…
The artificial variables do not have a meaning as slack or
surplus variable does. It is inserted into the equation simply
to give a positive solution at the origin; we artificially creating
a solution.
x1 +x2 – s1 +A1 = 300
A1 = 300
x1- s2 + A2 = 250
A2 = 250
11/14/22
Cont…
The purpose of the artificial variables is to get us off ground; but
once we get started, it has no real use and thus is discarded.
The artificial solution helps get the simplex process started, but we
do not want it to end up in the optimal solution, since it has no real
meaning.
Like a slack variable, a surplus variable has no effect on the
objective function in terms of increasing or decreasing cost. Thus, a
coefficient of 0 is assigned to each surplus variable in the objective
function.
11/14/22
Cont…
If the artificial variable appeared in the solution, it would
render the final solution meaningless. Therefore, we must
ensure that an artificial variable is not in the final solution.
Artificial variable are assigned a large cost in the objective
function to eliminate them from the final solution.
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
C.
Apply same procedure as you did in the case of
maximization problem.
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
11/14/22
Cont…
The optimal solution has been reached because all zj-cj row
values are less than or equal to zero and
All artificial variables are also eliminated from the simplex
tableau but this is not necessarily a condition for a problem
to be reached at optimal.
11/14/22
Cont…
D.
11/14/22