Republic of the Philippines

Central Luzon State University

COLLEGE OF ENGINEERING
Science City of Muñoz, Nueva Ecija

DEPARTMENT OF AGRICULTURAL AND BIOSYSTEMS ENGINEERING

I. INTRODUCTION

Optimization is defined as “the process of finding the most effective or favorable value or
condition”, which has the purpose of achieving the “best” design relative to a set of prioritized
criteria or constraints. These include maximizing factors such as productivity, strength, reliability,
longevity, efficiency, and utilization.
Engineers are often assigned design projects that require them to seek a solution that
efficiently locates a design that meets the identified criteria within the given constraints. Engineers
are often forced to identify a few appropriate design solutions and then decide which one best
meets the need of the client. This decision-making process is known as optimization.
Linear programming (LP) is one of the simplest ways to perform optimization. It helps you
solve some very complex optimization problems by making a few simplifying assumptions.
Technically Linear programming is an optimization technique for a system of linear constraints
and a linear objective function. An objective function defines the quantity to be optimized, and the
goal of linear programming is to find the values of the variables that maximize or minimize the
objective function.
This linear programs can be solve manually using simplex method. Simplex method is an
approach to solving linear programming models by hand using slack variables, tableaus, and pivot
variables as a means to finding the optimal solution of an optimization problem.
An alternate optimal solution is also called as an alternate optima, which is when a linear /
integer programming problem has more than one optimal solution. Typically, an optimal solution
is a solution to a problem which satisfies the set of constraints of the problem and the objective
function which is to maximize or minimize.
 In simplex method, if the z-row value for one or more non-basic variable is zero in the
optimal tableau, an alternate optimal solution exists. Graphically, alternate optima happens when
the objective function is parallel to a non-redundant binding constraint, which is a constraint that
is satisfied as an equation at the optimal solution. The zero coefficient of non-basic variable xi
indicates that it can be made basic, altering the value of basic variable without changing the value
of z.

Alternative Optima

When the objective function is parallel to a non-redundant binding constraint (i.e.,

a constraint that is satisfied as an equation at the optimal solution), the objective function can
assume the same optimal value at more than one solution point, thus giving rise to alternative
optima. The next example shows that there is an infinite number of such solutions. It also
demonstrates the practical significance of encountering such solutions.

Alternative optima occurs when:

 If the z-row value for one or more non-basic variables is 0 in the optimal tubule,
alternate optimal solution exists.
 When the objective function is parallel to a binding constraint objective
function will assume same optimal value.
 We have infinite number of such points
 II. SAMPLE PROBLEMS

EXAMPLE NUMBER 1.

Maximize : z = 2X1 + 4X2 ----------------------------------- z - 2X1 - 4X2 + 0S1 + 0S2 = 0

Subject to : X1 + 2X2 <5 ------------------------------------- X1 + 2X2 + S1 = 5

X1 + X2 <4 --------------------------------------- X1 + X2 + S2 = 4

X1, X2 >0

ENTERING
VARIABLE

BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
DEPARTING
S1 1 2 1 0 5 5/2 VARIABLE

S2 1 1 0 1 4 4/1

Z -2 -4 0 0 0

BASIC X1 X2 S1 S2 Solution
VARIABLE
S1 1 2 1 0 5 R1 / 2
S2 1 1 0 1 4
Z -2 -4 0 0 0

BASIC X1 X2 S1 S2 Solution
VARIABLE
X2 1/2 1 1/2 0 5/2
S2 1 1 0 1 4 R2-R1
Z -2 -4 0 0 0 R3+4R1
 BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 1/2 1 1/2 0 5/2
S2 1/2 0 -1/2 1 3/2
Z 0 0 2 0 10

So the optimal solution is: Z= 10; X2= 5/2; X1= 0

How do we know from the matrix above that alternative optima exist?
By simply looking at z-row coefficient of the non-basic variable.
BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 1/2 1 1/2 0 5/2
S2 1/2 0 -1/2 1 3/2
Z 0 0 2 0 10

The coefficient for X1 is 0, which indicates that X1 can enter the basic solution without
changing the value of z.
After knowing that there is alternative optima exist, we do iteration again to see other
possible solutions.
ENTERING
VARIABLE

BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
S1 1/2 1 1/2 0 5/2 1
1/2 0 -1/2 1 3/2 1 DEPARTING
S2 VARIABLE

Z 0 0 2 0 10

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 1/2 1 1/2 0 5/2

X1 1/2 0 -1/2 1 3/2 R2*2

Z 0 0 2 0 10
 BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 1/2 1 1/2 0 5/2 R1 – 1/2R2
X1 1 0 -1 2 3
Z 0 0 2 0 10

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 0 1 1 -1 1
X1 1 0 -1 2 3
Z 0 0 2 0 10

So the second optimum solution is: Z = 10; X1 = 3; X2 = 1

EXAMPLE NUMBER 2.

Maximize : z = 4X1 + 14X2 ----------------------------------- z - 4X1 - 14X2 + 0S1+ 0S2 = 0

Subject to : 2X1 + 7X2 <21 ------------------------------------- 2X1 + 7X2 + S1 = 21
7X1 + 2X2 <21 --------------------------------------- 7X1 + 2X2 + S2 = 21
X1, X2 >0

ENTERING
VARIABLE
BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE

S1 2 7 1 0 21 21/7 DEPARTING
VARIABLE

S2 7 2 0 1 21 21/2

Z -2 -4 0 0 0
 BASIC
X1 X2 S1 S2 Solution
VARIABLE
S1 2 7 1 0 21 R1 / 7

S2 7 2 0 1 21
Z -4 -14 0 0 0

BASIC X1 X2 S1 S2 Solution
VARIABLE
X2 2/7 1 1/7 0 3
S2 7 2 0 1 21 R2-2R1
Z -4 -14 0 0 0 R3+14R1

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 2/7 1 1/7 0 3
S2 45/7 0 -2/7 1 15
Z 0 0 2 0 42

So the optimal solution is: Z= 42; X2= 3; X1= 0

How do we know from the matrix above that alternative optima exist?
By simply looking at z-row coefficient of the non-basic variable.

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 2/7 1 1/7 0 3
S2 45/7 0 -2/7 1 15
Z 0 0 2 0 42
 The coefficient for X1 is 0, which indicates that X1 can enter the basic solution
without changing the value of z.
After knowing that there is alternative optima exist, we do iteration again to see
other possible solutions.
ENTERING
VARIABLE
BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
S1 2/7 1 1/7 0 3 3/(2/7)
S2 45/7 0 -2/7 1 15 15/(45/7) DEPARTING
VARIABLE
Z 0 0 2 0 42

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 2/7 1 1/7 0 3
X1 45/7 0 -2/7 1 15 R2/(45/7)
Z 0 0 2 0 42

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 2/7 1 1/7 0 3 R1-2/7R2
X1 1 0 -2/45 7/45 7/3
Z 0 0 2 0 42

BASIC
X1 X2 S1 S2 Solution
VARIABLE
X2 0 1 7/45 -2/45 7/3
X1 1 0 -2/45 7/45 7/3
Z 0 0 2 0 42

So the second optimum solution is: Z = 42; X1 = 7/3; X2 = 7/3

EXAMPLE NUMBER 3.

Maximize : z = 4X1 + 2X2 ----------------------------------- z - 4X1 - 2X2 + 0S1 + 0S2 = 0

Subject to : 2X1 + X2 <9 ------------------------------------- 2X1 + X2 + S1 = 9

X1 + 2X2 <9 --------------------------------------- X1 + 2X2 + S2 = 9

X1, X2 >0

ENTERING
VARIABLE

BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
DEPARTING
S1 2 1 1 0 9 9/2 VARIABLE

S2 1 2 0 1 9 9/1

Z -4 -2 0 0 0

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X1 2 1 1 0 9 R1 + 2

S2 1 2 0 1 9

Z -4 -2 0 0 0

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X1 1 1/2 1/2 0 9/2

S2 1 2 0 1 9 R2 + R1

Z -4 -2 0 0 0 R3 + 4R1
 BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X1 1 1/2 1/2 0 9/2

S2 0 3/2 -1/2 1 9/2

Z -4 -2 0 0 18

So the optimal solution is: Z= 18; X1= 9/2; X2= 0

BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
9 1
X1 1 1/2 1/2 0 9/2 ÷
2 2
9 3
S2 0 3/2 -1/2 1 9/2 ÷
2 2
Z -4 -2 0 0 18

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X1 1 1/2 1/2 0 9/2

X2 0 3/2 -1/2 1 9/2 (R2*2/3)

Z -4 -2 0 0 18

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X1 1 1/2 1/2 0 9/2 (R1 - 1/2R2)

X2 0 1 -1/3 2/3 3

Z -4 -2 0 0 18
 BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X1 1 0 2/3 -1/3 3

X2 0 1 -1/3 2/3 3

Z 0 0 2 0 18

So the second optimum solution is: Z = 18; X1 = 3; X2 = 3

EXAMPLE NUMBER 4.

Maximize : z = 30X1 + 20X2 ---------------------------------- z - 30X1 - 20X2 + 0S1 + 0S2 = 0

Subject to : X1 + 2X2 <80 ------------------------------------- X1 + 2X2 + S1 = 80

3X1 + 2X2 <120 --------------------------------------- 3X1 + 2X2 + S2 = 120

X1, X2 >0

ENTERING
VARIABLE

BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
S1 1 2 1 0 80 580/1
DEPARTING
S2 3 2 0 1 120 120/3 VARIABLE

Z -30 -20 0 0 0

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
S1 1 2 1 0 80

X1 3 2 0 1 120 R2/3

Z -30 -20 0 0 0
 BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
S1 1 2 1 0 80 R1 – R2

X1 1 2/3 0 1/3 40

Z -30 -20 0 0 0 R3 + 30R2

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
S1 0 4/3 1 -1/3 40

X1 1 2/3 0 1/3 40

Z 0 0 0 120 120

So the optimal solution is: Z= 120; X2= 0; X1= 40

BASIC
X1 X2 S1 S2 SOLUTION RATIO
VARIABLE
40
S1 0 4/3 1 -1/3 40
4/3
40
X1 1 2/3 0 1/3 40
2/3
Z 0 0 0 0 120

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X2 0 4/3 1 -1/3 40 R1*3/4

X1 1 2/3 0 1/3 40

Z 0 0 0 0 120
 BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X2 0 1 3/4 -1/4 30

X1 1 2/3 0 1/3 40 R2 – 2/3R1

Z 0 0 0 0 120

BASIC
X1 X2 S1 S2 SOLUTION
VARIABLE
X2 0 1 3/4 -1/4 30

X1 1 0 -1/2 1/2 20

Z 0 0 0 10 120
So the second optimum solution is: Z = 120; X1 = 20; X2 = 30

EXAMPLE NUMBER 5.

Maximize : z = 3x + 2y ----------------------------------- z – 3x -2y + 0S1+ 0S2 + 0S3 = 0

Subject to : x ≤ 40 ----------------------------------------- x + S1 = 40
y ≤ 60 ----------------------------------------- y + S2 = 60
3x + 2y ≤ 180 ------------------------------- 3x + 2y + S3 = 180
x,y≥0
ENTERING
VARIABLE
BASIC
X Y S1 S2 S3 SOLUTION RATIO
VARIABLE
S1 1 0 1 0 0 40 40 DEPARTING
VARIABLE

S2 0 1 0 1 0 60 0

S3 3 2 0 0 1 180 60

Z -3 -2 0 0 0 0
 BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40

S2 0 1 0 1 0 60

S3 3 2 0 0 1 180 R3-3R1

Z -3 -2 0 0 0 0 R4+3R1

ENTERING
VARIABLE

BASIC
X Y S1 S2 S3 SOLUTION RATIO
VARIABLE
X 1 0 1 0 0 40 0
60
S2 0 1 0 1 0 60
DEPARTING
S3 0 2 -3 0 1 60 30 VARIABLE

Z 0 -2 3 0 0 120

BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40

S2 0 1 0 1 0 60

Y 0 2 -3 0 1 60 R3/2

Z 0 -2 3 0 0 120
 BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40

S2 0 1 0 1 0 60 R2-R3

Y 0 1 -3/2 0 1/2 30

Z 0 -2 3 0 0 120 R4+2R3

BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40

S2 0 1 3/2 1 -1/2 30

Y 0 1 -3/2 0 1/2 30

Z 0 0 0 0 0 180

So the optimal solution is: Z= 180; X= 40; Y= 30

How do we know from the matrix above that alternative optima exist?
By simply looking at z-row coefficient of the non-basic variable.

BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40

S2 0 1 3/2 1 -1/2 30

Y 0 1 -3/2 0 1/2 30

Z 0 0 0 0 0 180
 The coefficient for X1 is 0, which indicates that X1 can enter the basic solution
without changing the value of z.
After knowing that there is alternative optima exist, we do iteration again to see
other possible solutions.

ENTERING
VARIABLE

BASIC
X Y S1 S2 S3 SOLUTION RATIO
VARIABLE
X 1 0 1 0 0 40 40
DEPARTING
S2 0 0 3/2 1 -1/2 30 20 VARIABLE

Y 0 1 -3/2 0 1/2 30 -20

Z 0 0 0 0 0 180

BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40

S2 0 0 3/2 1 -1/2 30 R2*2/3

Y 0 1 -3/2 0 1/2 30

Z 0 0 0 0 0 180

BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 1 0 0 40 R1-R2

S1 0 0 1 2/3 -1/3 20

Y 0 1 -3/2 0 1/2 30 R3+3/2R2

Z 0 0 0 0 0 180
 BASIC
X Y S1 S2 S3 SOLUTION
VARIABLE
X 1 0 0 -2/3 1/3 20

S1 0 0 1 2/3 -1/3 20

Y 0 1 0 1 0 60

Z 0 0 0 0 1 180

So the second optimum solution is: Z = 180; X = 20; Y = 60

III. REFERENCES

TAHA, Hamdy A. (2013). Operation Research- An introduction. 9th edition. Retrieved at

slideshare.net. Date Accessed: November 19, 2019

Verma, D. (2015). Special Cases in Simplex Method. Retrieved November 18, 2019 from
https://www.slideshare.net/itsmedv91/special-cases-in-simplex

Universal Teacher Publications. (n.d.). Multiple Optimal Solutions: Simplex Method. Retrieved
on November 18, 2019 from
http://www.universalteacherpublications.com/univ/ebooks/or/Ch3/splcase2.htm

Trick, M.A. (1998). Alternate Optimal Solutions, Degeneracy, Unboudedness, Infeasibility.

Retrieved on November 18, 2019 from
https://mat.gsia.cmu.edu/classes/QUANT/NOTES/chap7/node5.html

Zeepedia. (n.d.) Linear Programming:Multiple or Alternative optimal Solutions. Retrieved

November 18, 2019 from https://www.zeepedia.com/read.php?l
inear_programming_multiple_or_alternative_optimal_solutions_operations_research&b=
66&c=30&fbclid=IwAR3kmNIafiUwtcOIzrHMaEZL73yWHB0DOeZnldirTWciSbSbm
pKQr7oGC8o

University of Texas. (n.d.). Special Situations in the Simplex Algorithm. Retrieved on November
18, 2019 from https://personal.utdallas.edu/~scniu/OPRE-6201/documents/LP10-Special-
Situations.pdf?fbclid=IwAR02Rg0hVmSpY4
HPGskwj2uTks8KyjL2mt9LqX8y5p3LvPAwodz2y5jNxe