Dr. B.

Venkateswarlu

vit

VIT UNIVERSITY VELLORE

UNIT - I
LINEAR PROGRAMMING PROBLEMS:
Linear programming problems deal with determining optimal allocations of limited
resources to meet given objectives. The resources may be in the form of men, raw
materials, market demand, money and machines and so on .The objective is usually
maximizing profit or minimizing total cost.
DEFINITION OF L.P.P :
L.P.P. deals with the optimization (i.e. maximization or minimization) of a function
of decision variables.(The variables whose values determine the solution of a problem
are called decision variables of the problem) known as objective function subject to a
set of simultaneous linear equations (inequalities) known as constraints. The term linear
means that all the variables occurring in the objective function and the constraints are of
the first degree in the problems under consideration and the term programming means
the process of determining a particular course of action.
USES OF L.P.P :
L.P.P. techniques are used in many industrial and economic problems. They
are applied product mix, diet planning, cargo loading, capital budgeting, man power
planning, transportation and assignment problems, oil refineries, airlines, railways,
textile industries, chemical industries , steel industries, food processing industries and
defence establishments are also the uses of L.P.P.
MATHEMATICAL FORMULATION OF L.P.P :
STEP 1: Identify the unknown decision variables, x1, x2, x3 .. xn to be determined.
STEP 2: Identify the objective function (i.e. maximize or minimize) and represent it also
a linear function of decision variables.
STEP 3: Identify all the constraints and representing the maximum availability () or
minimum commitment () or equality(=) in the problem and express them as linear
equations or inequalities of decision variables.
STEP 4: To add the non-negative constraint (xj>0, j=1 to n) from the Consideration . So
that negative values of the decision variables do not have any valid
physical
interpretation.
STEP 5: Express the complete formulation of L.P.P. as a general mathematical model
OBJECTIVE FUNCTION
Max or Min () =c1x1+c2x2+..+cnxn
Subject to the constraints
a11 x1+a12 x2 a1n xn{ or = or } b1
a21 x1+a22 x2 a2n xn or = or }b2
.
. . .
.
. .
. . .
.
. . . .
. .
. . .
am1 x1+am2 x2amn xn or = or }bm
GRAPHICAL METHOD :
Graphical method is used for solving L.P.P . that involve only two variables .
FEASIBLE SOLUTION:
Any non-negative solution which satisfies all constraints is known as a feasible
solution of the problem.
FEASIBLE REGION:
The collection of all feasible solutions is known as a feasible Region.
CONVEX SET:

Dr. B. Venkateswarlu

vit

A set (or region ) is convex if only for any two points on the set. The line segment
joining those points lies entirely in the set. Thus, the collection of feasible solutions in a
L.P.P. form a convex set.
Eg:

This is a convex set

This is not a convex set

CONVEX POLYGON:
A convex polygon is a convex set formed by the intersection of a finite number of
closed planes.
GRAPHICAL SOLUTION PROCEDURE:
STEP 1: Consider each inequality constraint as equation.
STEP 2: Take one variable (say x) in a given equation equal to zero and find the value
of other variable (say y) by solving that equation to get one co-ordinate [say (0,y)] for
the equation.
STEP 3: Take the second variable (say y) as zero in the equation and find the value of
first variable (say x) to get another co-ordinate [say (x,0)] for the equation.
STEP 4: Shade the appropriate areas as given by the constraints. If the constraint is
< type, shade the area below the line. If the constraint is >, shade the area above
the line. If the constraint is = type, do not shade any area and the line itself is the
region.
STEP 5: Repeat step 2 to 4 for other constraints.
STEP 6: Identify the feasible region by locating the area satisfying all constraints i.e.,
common with all the constraint areas. This region will have the optimum solution. If
there is no common area possible then the solution is infeasible.
STEP 7: Put the co-ordinates of each of such vertex in the objective function. Choose
that vertex which achieves the most optimal solution (i.e. in the case of maximization,
the vertex that gives the maximum value of z and in the case of minimization, the vertex
that gives the minimum value of z) solution can be obtains in the following manner.
Vertex
Number
1
2
.
.
.
n

Coordinates Of Vertices Of
Common Shaded Feasible Region
(x1, y1)
(x2,y2)
.
.
.
(xn,yn)

Optimum
Value Of Z
Z1 =
Z2 =
.
.
.
Zn =

NOTE: 1. It is better to leave 3 units at the bottom and also at the left side so that your
graphical solution will be clearly represented.
2. Divide the scale approximately on x axis and y-axis such that you can represent all its
values.
3. It is always better to have same scale on both axis.

Dr. B. Venkateswarlu

vit

MULTIPLE OPTIMAL SOLUTIONS OR ALTERNATE OPTINAL SOLUTIONS:

DEF: For same L.P.P. there may be more than combination of values of the decision
variables yielding the best objective function value. Such combinations of the values of
the decision variables are known as alternate optimum solutions (or) infinite number of
optimal solutions.
(OR)
An L.P.P. having more than one optimal solution is said to have alternative or
multiple optimal solutions. That is the resources can be combined in more than one way
to maximize the profit.
OPTIMAL SOLUTION:
If there is no other superior solution to the solution obtained for a given L.P.
model, then the solution obtained is treated as the optimal solution.
GENERAL LINEAR PROGRAMMING PROBLEMS: (Basic structure of L.P.P)
Def: The L.P involving more than two variables may be expressed as follows.
Max (Z) or Min (Z ) = c1x1 +c2x2+.cnxn
Subject to the constraints
a11 x1+a12 x2 a1n xn { or = or } b1
a21 x1+a22 x2 a2n xn { or = or }b2
.
. . .
.
. .
. . .
.
. . . .
. .
. . .
am1 x1+am2 x2amn xn { or = or } bm

and x1, x2,.xn 0

SLACK VARIABLES:
n

If the constraints of a general L.P.P. be aij x j bi where I = 1 to k

-----------(1)

j 1

then the

non-negative variables xn+i which are introduced to convert the inequalities

n

(1) to the equalities

aij x j xn i = bi, where i=1, 2,.k

are called slack variables.

j 1

The value of those variables can be interpreted as the amount of unused resources.
Eg: 2x1 + 3x2 4 (1)
5x1 + 4x2 6 (2)
(1) =>
=>2x1 +3x2 +x3 = 4
(2) =>
=> 5x1+4x2 +x4 = 6
here x3 and x4 are slack variables.
SURPLUS VARIABLES:
n

If the constraints of a general L.P.P.

aij x j

j 1

then the non-negative variables xn+i

n

bi

where i=k+1

--------------(2)

which are introduced to convert the

inequalities (2) to the equalities aij x j x n i bi ,

i=k+1..

are called surplus

j 1

variables. The value of these variables can be interpreted as the amount over and
above the required level.
Eg:
2x1 + 4x2 5
(1)
3x1 + 4x2 4
(2)
(1)=>
2x1 + 4x2 +x3 = 5
(2)=>

3x1 + 4x2 x4 = 4

Here x3 is slack variable and x4 is surplus variable

3

Dr. B. Venkateswarlu

vit

CONONICAL FORM OF L.P.P:

STEP 1: The objective function is of maximization type
STEP 2: All the constraints are of type
STEP 3: All variables xi are +ve
Max() =c1x1+c2x2+..cnxn
Subject to constraints
a11 x1+a12 x2 a1n xn b1
a21 x1+a22 x2 a2n xn b2
.
. . .
.
. .
.
. . . .
. .
am1 x1+am2 x2amn xn bm

and x1, x2,.xn 0

STANDARD FORM OF L.P.P: (working rules) :

STEP 1: The objective function is of maximization type
STEP 2: All the constraints are expressed as equations
STEP 3: Right hand side of each constraint is +ve
STEP 4: All variables are positive
Max() = c1x1+c2x2+..cnxn
Subject to constraints
a11 x1+a12 x2 a1n xn = b1
a21 x1+a22 x2 a2n xn = b2
.
. . .
.
. .
.
. . .
.
. .
am1 x1+am2 x2amn xn = bn
and x1,x2,.xn 0
MATRIX FORM:
Max(z) = cx
Subject to the constraints
Ax=b and x 0
where c=c1,c2..cn; X=x1,x2.xn
a11
a21
.
A
.
.
.
am1

a12
a22
.
.
.
.
am2

.....
......

.....
.....
.....
.....
.....

a1n
a2n
.
.
.
.
amn

b1
b2
.
B .
.
.
bm

NOTE:
(1) The minimization of a function z is equivalent to the maximization of the negative
expression of this function
i.e., min (z) = - max (-z)
= - max (z) where z= -z
(2) An inequality in one direction can be converted into an inequality in the opposite
direction by multiplying both sides by -1
eg: 5x1+4x2 2
=>
-5x1-4x2 -2
(3) An equality constraint can be expressed as two inequalities
Eg: 2x1+3x2 =5
=>
2x1+3x2 5
2x1+3x2 5
NOTE: Whenever
stack/surplus variables are introduced in the constraints, they
should also appear in the objective function with zero coefficients

Dr. B. Venkateswarlu

vit

SIMPLEX METHOD:
BASIC SOLUTION: Given a system of m linear equations with n variables (m<n). The
solution obtained by setting (n-m) variables equal to zero and solving for the remaining
m variables is called a basic solution.
The m variables are called basic variables and they form basic solution. The
(n-m) variables which are put to zero are called as non-basic variables.
NOTE:

The maximum number of possible basic solutions is given by

n Cm

n!
,
(n m)! m!

where n= number of variables and m= number of constraints

BASIC FEASIBLE SOLUTION:

A feasible solution which is also basic i.e. all basic variables are non-negative
is called basic feasible solution.
Eg: In the above example
x1=0, x2=5/3,
x3=2/3 and
are basic feasible solutions
x1= 2, x2=1,
x3=0
NON-DEGENERATE BASIC SOLUTION:
A basic solution is said to be a non-degenerate basic solutions, if none of the
basic variables is zero
Eg: In the above example.
x1=0, x2 =5/3,

x3 =2/3
are non degenerate basic solutions

x1= 2, x2=1,

x3 = 0

DEGENERATE BASIC SOLUTION:

A basic solution is said to be a degenerate basic solution, if one or more
of the basic variables are zero.
OPTIMUM BASIC FEASIBLE SOLUTION:
A basic feasible solution is a solution which optimize (max or min)in the
objective function.

Dr. B. Venkateswarlu

vit

SIMPLEX METHOD ALGORITHM:

STEP 1: Check whether the objective function is to be maximized or minimized. If it is
to be minimized, then convert it into a problem of maximization, by using the
relation
Min(z)=-max(-z)
STEP 2: Check whether all bis are positive. If any one of the bi's is negative, multiply
both sides of that constraints by -1 so as to make its right hand side positive.
STEP 3: Convert all the inequations of the constraints into equations by introducing
slack and /surplus variables in the constraints. Put the costs of these variables
equal to zero.
STEP 4: Find an initial basic feasible solution and express the above information
(conveniently)in the following simplex table.
Cj

C1

C2

C3

..

CB

B.V

xb

x1

x2

x3

xn+1

xn+2

xn+3

CB1

xn+1

b1

a11

a12

a13

..

CB2

xn+2

b2

a21

a22

a23

CB3

xn+3

b3

a31

a32

a33

Body matrix
Zj

Unit matrix

Z0

Z1

Z2

Z3

..

Zj -Cj

Z1- C1

Z2- C2

Z3- C3

...

Where Cj row denotes the coefficients of the variables in the objective function
CB column denotes the coefficients of the basic variables in the objective.
B .v column denotes the basic variables
XB values of the basic variables
(zj -cj) denotes the net evaluations
zj = CB xj
Z j C j C B x j C j , where z0 = CB XB j=1 to n
j 1ton

STEP 5: Compute the net evaluations (zj - cj), j=1, 2, n by using the relation.
Zj - cj = CB Xj-Cj
STEP 6: Examine the sign of (zj - cj), j=1 to n
Case 1: If all (zj cj) 0 then the current basic possible solution xB is an
optimum basic feasible solution.
CASE 2: If atleast one (zj cj) 0, then proceed onto the next step.
STEP 7: (To find the entering variable) or (key column). If there are more than one
negative (zj cj), then choose the most negative of them. Let it be (zr cr) for
same j = r
CASE 1: If all xir 0 (i=1,2,3,.m) then there is an unbounded solution to the
given L.P.P.
CASE 2: If atleast one xir 0 (i=1,2,..m)then the corresponding vector xr
enters the basis in basic variable column.
6

Dr. B. Venkateswarlu

vit

STEP 8: To find the leaving variable (key row)

Compute the ratio = min {xBi / xir, xir 0}. Let the minimum of these ratios be
(xBk/xkr), then the vector xk will leave the basis in basic variable column. The
common element xkr which is in the kth row and rth column is known as the
leading element or key element or pivotal element of the table.
STEP 9: Convert the leading element to unity by dividing its row by the key element
itself and all other elements in its column to zeros by making use of the relations.
New elements = old elements / key elements
(for key row)
New elements = old elements - (key elements * key column element) / key elements
(for non-key row)
STEP 10: Go to step (5) and repeat the computational procedure until either an
optimum solution is obtained or there is an indication of an unbounded solution.

Dr. B. Venkateswarlu

vit

ARTIFICIAL VARIABLES TECHNIQUES:

1. Big M- Method (or) Penalty method
2. Two Phase Simplex method
ARTIFICIAL VARIABLE:
Sometimes to overcome some mathematical inconvenience, extra variables
are added to the constraints (i.e. to avoid negative variables).These variables are
known as artificial variables.
Eg:
2x1+ 3x2 4
2 x1 + 3x2 +x3
=4
5x1 +x2 3
5 x1+x2
-x4+x5
=3
x1 +2x2 = 7
x1 +2x2
+x6
=7
Here x3 is slack variable, x4 is surplus variables, x5 and x6 are artificial variables.
ALGORITHM FOR BIG-M-METHOD:
STEP 1: Express the L.P.P.
variables, if necessary.

in the standard form by introducing slack/surplus

STEP 2: Introduce the positive artificial variables A1,A2,..An to the left hand side of all
constraints of or = type. The purpose of introducing artificial variables is just to
obtain an initial basic feasible solution .How ever, addition of these artificial variables
causes violation of the corresponding constraints. Therefore we would like to get rid of
these variables and could not allow them to appear in the final solution. To achieve this
we assign a very large penalty (-m for maximization and +m for minimization problems,
where M>0) as the coefficients of the artificial variables in the objective function.
STEP 3: Solve the modified L.P.P. by using simplex method. While making iterations,
using simplex method, one of the following three cases may arise.
CASE 1: If no artificial variable remains in the basis and the optimality condition is
satisfied, then the current solution is the optimal basic feasible solution.
CASE 2: If atleast one artificial variable appeared in the basis at zero level (i.e., with
zero value of xB column )and the optimality condition is satisfied, then the current
solution is an optimum basic feasible solution.
CASE 3: If atleast one artificial variable appears in the basis at non-zero level (i.e. with
positive value in xB column) and the optimality condition is satisfied, then the original
problem has no feasible solution. The solution satisfies the constraints but does not
optimize the objective function since it contains a very large penalty M and is called
pseudo-optimal solution.
[NOTE: While applying simplex method, whenever an artificial variable happens to
leave the basis, we drop that artificial variable and omit all the elements (entries)
corresponding to its column from the simplex table.]
TWO PHASE SIMPLEX METHOD:
The two-phase method is another method to solve a given problem in which
some artificial variables are involved. The solution is obtained in two phases as follows:
Phase 1: In this phase, the simplex method is applied to a specially constructed
auxiliary linear programming problem leading to a final simplex table containing a basic
feasible solution to the original problem.

Dr. B. Venkateswarlu

vit

Step 1: Assign a cost -1 to each artificial variable and a cost zero to all other variables
(in place of their original cost) in the objective function. Thus the new objective function
is Z * = -A1-A2-A3- . . . . . . An, where Ais are the artificial variable.
Step 2: Construct the auxiliary LPP in which the new objective function Z* is to be
maximized subject to the given set of constraints .
Step 3: Solve the auxiliary LPP by simplex method until either of the following three
possibilities arise.
Case 1: Max (Z*)<0 and atleast one artificial variable appears in the optimum
basis at a non-zero level .In this case the given LPP does not possess any
feasible solutions, stop the procedure.
Case 2: Max (Z*)=0 and atleast one artificial variable appears in the optimum
basis at zero level .In this case proceed to phase-2.
Case 3: Max (Z*)=0 and no artificial variable appears in the optimum basis. In
this case proceed to phase-2.
Phase 2: Use the optimum basic feasible solution of phase-1 as a starting solution for
the original L.P.P. .Assign the actual costs to the variables in the objective function and
a zero cost to every artificial variable that appears in the basis at the zero level. Use
simplex method to the modified simplex table at the end of phase-1 an optimum basic
feasible solution is obtained.
NOTE 1: In phase-1, the iterations are stopped as soon as the value of the new
objective function becomes zero because this is the maximum value. There is no need
to continue till the optimality is reached if the value becomes zero earlier than that.
NOTE 2: The new objective function is always of maximization type regardless of
whether the original problem is of maximization or minimization type.
NOTE 3: Before staring phase 2, remove all artificial variables from the table which
were non-basic at the end of phase -2.

Dr. B. Venkateswarlu

vit

DISADVANTAGES OF BIG-M-METHOD OVER TWO PHASE METHOD:

Even though big-M-method is always be used to check the existence of a
feasible solution it may be inconvenient especially when a digital computer is used
because of the manipulation of the constant M. On the other hand 2-phase method
eliminates the constant M from calculation.
UNRESTRICTED (OR) UNCONSTRAINED VARIABLES
In an L.P.P. if any variable is unrestricted (without specifying its sign) it can be
expressed as the difference between two positive variables. The problem can be converted into
an equivalent one involving only positive variables.
DEGENERACY IN L.P.P:
The method of obtaining a degenerate basic solution in a L.P.P. known as degeneracy.
It may arise
CASE 1: At the initial stage .In this case atleast one basic variable is zero in the initial basic
feasible solution.
CASE 2: At any subsequent iteration. In this case more than one basic variable is eligible to
leave the basis and hence in the next iteration one or more basic variable become zero. The
subsequent iteration may not produce any improvement in the objective function and thus
produces degenerate solution. As a result, the simplex iteration are done repeatedly without
improving the solutions. This concept is known as cycling.
METHOD TO RESOLVE THE DEGENERACY:
STEP 1: First select the rows for which the minimum non-negative ratio is same (tie). Suppose
there is a tie between first and third row.
STEP 2: Now re-arrange the columns of the usual simplex table so that columns forming the
original unit matrix came first in proper order.
STEP 3: Find the minimum ratio only for first and fourth rows (tie).
= (elements of the first column of unit matrix / corresponding elements of key column)
Only for the tied rows, i.e., for the first and third rows.
CASE 1: if the third row has the minimum ratio then this row will be the key row and the key
element can be determined by intersecting the key row with key column.
CASE 2: If this minimum is also not unique, then go to next step.
STEP 4: Find the minimum ratio,
= (elements of second column of unit matrix / corresponding elements of key column)
CASE 1: If this minimum ratio is unique for the first row (say), then this row will determine the
key element by intersecting with key column.
CASE 2: If this minimum is still not unique, then go to step 5.
STEP 5: Next find the minimum ratio
= (elements of third column of unit matrix / corresponding elements of key column)
Above step is repeated till the minimum ratio is obtained so as to resolve the degeneracy. After
resolution of this tie, simplex procedure is obtained to obtain the optimum solution.

10

Dr. B. Venkateswarlu

vit

DUAL SIMPLEX METHOD

The regular simplex method starts with a basic feasible but non-optimal solution, and
works towards optimality, whereas the dual simplex method starts with a basic infeasible but
optimal solution and works towards feasibility. Also in regular simplex method we first determine
the entering variable and then the leaving variable. While in the case of dual simplex method we
first determine the leaving variable and then entering variable.
ALGORITHM:
STEP 1: Convert the problem to maximization form, if it is initially in the minimization form.
STEP 2: Convert () type constraints, if any to type by multiplying both sides by -1.
STEP 3: Convert the inequality constraints into equalities by introducing slack variables only.
Find the initial basic solution and express this information in the simplex table.
STEP 4: (optimal condition)
Test the nature of (zj - cj) and xBi
CASE 1: If all (zj - cj) 0 and x bi 0.The current solution is an optimum feasible solution.
CASE 2: If all (zj - cj) 0 and atleast xBi<0, then go to next step.
CASE 3 : If all (zj - cj) < 0 then dual simplex method fails.
STEP 5: (Feasibility condition)
CASE 1: TO FIND THE LEAVING VARIABLE
The leaving variable is the basic variable corresponding to the most negative
value of xBi. i.e., min {xBi, xBi< 0 }.
Let x k be the leaving variable.
CASE 2: TO FIND THE ENTERING VARIABLE
Compute the ratio between (zj cj) row and the key row i.e.
=max {zj-cj / aik, aik< 0 }
(Consider the ratios with negative denominators alone). The entering variable is the nonbasic variable is the non-basic variable corresponding to the maximum ratio . If there is no
such ratio with negative denominator, then the problem does not have a feasible solution.
STEP 6: Convert the leading element to unity by dividing its row by the key element itself and
all other elements in its column to zeros by making use of the following formula.
New element = old element / key element
(for key row)
New element=old element - key row element * key column elements
( for non-key row)
key element
STEP 7: Repeat the procedure until either an optimum feasible solution is obtained (or) there is
an indication of no feasible solution.
DUALITY IN LINEAR PROGRAMMING PROBLEM
Introduction: For every LPP there is a unique LPP associated with it, involving the same data
and closely related optimal solutions. The original (given) problem is then called the Primal
problem while the other is called as dual problem. But in general, the two problems are said to
be duals of each other.
The importance of the duality concept is due to two main reasons. Firstly, if the primal
contains a large number of constraints and a smaller number of variables, the labour of
computation can be considerably reduced by converting in to the dual problem and then solving
it. Secondly, the interpretation of the dual variables from the cost or economic point of view
proves extremely useful in making future decisions in the activities being programmed.

11

Dr. B. Venkateswarlu

vit

FORMULATION OF DUAL PROBLEMS:

There are 2 important forms of primal-dual pairs, namely symmetric form and
unsymmetric form.
STANDARD FORM OF PRIMAL PROBLEMS:
Consider the following L.P.P
Max(z) = c1x1+c2x2+..cnxn
Subject to constraints
a11 x1+a12 x2 a1n xn b1
a21 x1+a22 x2 a2n xn b2
.
. . .
.
. .
.
. . .
.
. .
am1 x1+am2 x2amn xn bm and x1, x2,.xn 0
To construct the dual problem, we use the following rules
RULE 1: The maximization problem in the primal becomes the minimization problem in the dual
and vice-versa.
RULE 2: The maximization problem has type constraints while the minimization problem has
type constraints.
RULE 3: If the primal m constraints and m variables i.e., then the dual will contain n constraints
and m variables. i.e., the transpose of the body matrix of the primal problem gives the
body matrix of the dual and vice-versa.
RULE 4: The constants c1, c2, c3.,cn in the objective function of the primal appear in the
constraints of the dual.
RULE 5: The constants b1, b2, b3..bm in the constraints of the primal appear in the
objective function of the dual.
RULE 6: The variables in both problems are positive.
RELATION OF PRIMAL TO DUAL:
Primal variable
x1
x2
x3 .
Y1
a11
a12
a13 ...
Y2
a21
a22
a23 ..
dual
Y3
a31
a32
a33 .
variables
.

ym
am1
am2
am3

.
c1
c2
c3.

b1
b2
b3

xn
a 1n
a 2n
a 3n

R.H.S. of
primal
constraints

amn

cn

bm

R.H.S of dual constraints

DUAL PROBLEM
Min (z) = b1 y1+b2 y2+..+ bm ym
Subject to the constraints
a11 y1+a12 y2 am1 y m c1
a 12 y1+a22 y2 am2 ym c2
.
. . .
.
. .
. . .
.
. . . .
. .
. . .
a1n y1+a2n y2amn ym cn
and y1, y2,.ym 0

12

Dr. B. Venkateswarlu

vit

SOME IMPORTANT RESULTS IN DUALITY

RESULT 1: The dual of the dual is primal.
RESULT 2: If one is maximization problem then the other is a minimization problem.
RESULT 3: Fundamental theorem of duality (or) shadow prices.
If either the primal or the dual problem has the finite optimal solution then the other
problem also has a finite optimal solution and the values of the objective functions are equal.
i.e., max(z) = min (z). The solution of the other problem can be read from the (zi - cj) row below
the columns of slack/surplus variables. The values of the dual variables are called shadow
prices.
RESULT 4: (Existence theorem)
If either the primal or the dual problem has an unbounded solution, then the other problem has
no feasible solution.
RESULT 5: (Complementary slackness theorem)
CASE 1: If a primal variable is positive, then the corresponding dual constraint is an equation at
the optimum and vice-versa.
CASE 2: If a primal constraint is a strictly inequality, then the corresponding dual variable is
zero at the optimum and vice-versa.
ADVANTAGES OF DUALITY:
1.Solution of the dual helps in checking the accuracy of the primal solution for the computational
errors.
2.Time can be saved by reducing the computational time, if the dual is solved in case the primal
problem contains a large no of constraints and a smaller no of variables.
3.Duality is used to solve an L.P.P. by the simplex method in which the initial solution is
infeasible (i.e. dual simplex method)
4. The dual variables provide an economic interpretation of the final solution of an L.P.P.
5. In economics, duality is useful in the formulation of the input-output systems. It is also used in
engineering, physics and so on.
ASSUMPTIONS IN L.P.P:
PROPORTIONALITY: This assumption is for both the objective functions and the
constraints.
i.e., the contribution of each activity to the value of the objective function z is proportional
to the level of the activity xj as represented by cj xj term in the objective function.
The contribution of each activity to the left hand side of each functional constraint is
proportional to the level of the activity xj, as represented by the aj xj term in the constraint. This
is also called as linearly assumption.
ADDITIVITY:
Every function in a L.P.P. model (whether the objective function or the function on the
left hand side of a functional constraint).is the sum of the individual contributions to the
respective activities (or) total resources are equal to the sum of resources used by the individual
activities.
Eg: Additivity means, if it takes t1 hours on machine A to make product p 1 and t2 hours to
make product p2 then the time on machine A devoted to produce p1 and p2 both is t1+t2,
Provided the time required to change the machine from product p1 to p2 is negligible.
MULTIPLICATIVITY: It requires, (1) if it takes one hours to make a single item on a given
machine, it will take 10 hours to make 10 items. (2) The total profit from selling a given number
of units is the unit profit times the number of units sold.
DIVISIBILITY:
Solution need not be a whole number i.e. decision variable can be in fractional form.

13

Dr. B. Venkateswarlu

vit

CERTAINITY: (OR) DETERMINISTIC:

Coefficients of the objective functions and constants are known, constants do not
change.
FINITENESS:
Activities and constraints are finite number.
OPTIMISATION:
The ultimate objective is to obtain an optimum solution (maximize or minimize)
ADVANTAGES OF L.P.P:
1. It makes a scientific and mathematical analysis of the problem situations.
2. It gives an opportunity to the decision maker to formulate his strategies consistent with
the constraints and the objectives.
3. It deals with changing situations. Once a plan is arrived through the L.P.P. it can also
be reevaluated for changing conditions.
4. By using L.P.P. the decision maker makes sure that he/she is considering the best
solution.
LIMITATIONS OF L.P.P. : (DRAW BACKS)
1. The major limitation of L.P.P. is that it treats all relation ships as linear. But it is not true real
life situations.
2. The decision variables in some L.P.P. would be meaningful only if they have integer values.
But sometimes we get fractional values to the optimal solution.
3. The problems are complex, if the number of variables and constraints are quite large.
4. L.P. deals with only single objective problems whereas in real life situations there may be
more than one objective.
ADVANTAGES OF GRAPHICAL SOLUTION METHOD:
1. Graphical method is used for solving L.P.P. that involve two variables.
2. Simple steps to reach optimal solution
3. Easy interpretation of results.
4. Less time consuming.

14

Dr. B. Venkateswarlu

vit

ASSIGNMENT PROBLEMS:
The assignment problem is a particular case of TP in which the objective is to assign a
number of tasks (or) jobs (or) sources (or) origins to an equal number of facilities machines (or)
destinations at a minimum cost (or) at a maximum profit. Suppose that we have m jobs to be
performed on m machines (one job to one machine) and our objective is to assign the jobs to
the machines at the minimum cost (or max profit) under the assumption that each machine of
each job can perform each machine.
The assigned problem can be stated in the form of m*n matrix.
1. Let c ij = cost of assigning ith job to jth machine
a ij=Supply capacity of souece (job) where ai= 1 for all values of i
bj= Demand of destination, where bj=1 for all values of j.
CONDITION TO BE SATISFIED IN ASSIGNMENT PROBLEM:
Number of rows(m) should be equal to number of columns (n).
CASE 1: If m < n add dummy rows accordingly so that m=n the cell values are taken as zero.
CASE 2: If m > n add dummy columns accordingly so that m=n the cell values are taken as
zero.
The assignment model is represented in a matrix form as follows:
1
2
3
. J
..n
C11
C12
C13
..cij
..c1n
A1=1
C21
C22
C23
.c2j
..c2n
A2=1
C31
C32
C33
.c3j
.c3n
A3=1
Ci1
Ci2
Ci3
.cij
.cin
Ai=1
Cm1
Cm2
Cm3
cmj
..cmn
Am=1
B1=1
B2=1
B3=1
Bj=1
Bn=1
UNIT II : THE ASSIGNMENT METHOD (OR) HUNGERIAN METHOD OR FAST FOOD
TECHNIQUE :
An efficient method for solving as assignment problem as developed by the Hungarian
mathematician D.konig is summarized below.
STEP 1: Determine the cost table from the given problem
CASE 1: If the number of sources is equal to the number of destinations, go to step 3.
CASE 2: If the number of sources is not equal to the number of destinations, go to step2.
STEP 2: Add a dummy source or dummy destination, so that the cost table becomes a square
matrix. The cost entries of dummy source/destination are always zero.
STEP 3: Locate the smallest element in each row of the given cost matrix and then subtract the
same from each element of that row.
STEP 4:In the reduced matrix obtained in step 3,locate the smallest element of each column
and then subtract the same from each element of that column. Each column and each
row now have atleast one zero.
STEP
a)

b)
c)
d)

5: In the modified matrix obtained in step (4),search for an optimal assignment as

follows:
Examine the rows successfully until a row with single zero is found. En rectangle this zero
and cross off(x) all other zeros in its column. Continue in this manner until all the rows have
been taken care of.
Repeat the procedure for each column of the reduced matrix.
If a row and or column has two or more zeros and cannot be chosen by inspection then
assign arbitrary any one of these zeros and cross off all other zeros of that row/column.
Repeat a) through c) above successively until the chain of assigning or cross (x) ends.

STEP 6: To find the optimal solution

15

Dr. B. Venkateswarlu

vit

CASE 1: If the number of assignment is equal to n (the order of the cost matrix), an optimum
solution is reached.
CASE 2: If the number of assignments is less than n (the order of matrix), go to next step.
STEP 7: Draw the minimum number of horizontal and\or vertical lines to cover all the zeros of
the reduced matrix this can be conveniently done by using a simple procedure.
a) Mark () rows that do not have any assigned zero.
b) Mark () columns that have zeros in the marked rows.
c) Mark () rows that have assigned zeros in the marked columns.
d) Repeat (b) and (c) above until chain of marking is completed.
e) Draw lines through all the unmarked rows and marked columns. This gives us the
desired minimum number of lines.
STEP 8: Develop the new revised cost matrix as follows.
a) Find the smallest element of the reduced matrix not covered by any of the lines.
b) Subtract this element from all the uncovered elements and add the same to all the
elements lying at the intersection of any two-line s.
STEP 9: Go to step(6) and repeat the procedure until an optimum solution is obtained.
MULTIPLE OPTIMAL SOLUTIONS:
In an assignment problems the facilities are assigned to jobs on a one-one basis using
Hungarian method.
While making an assignment in the reduced assignment matrix, it is possible to have two or
more ways, to strike of a certain number of zeros.
This indicates that there are more than the required number of independent zero elements. In
such cases, there will be multiple optimal solution with the same total cost of assignment.
UNBALANCED ASSIGNMENT PROBLEMS:
If the number of rows is not equal to number of columns in the cost matrix of the given
assignment problem, then the given assignment problem is said to be unbalanced.
First convert the unbalanced A.P into a balanced one by adding dummy rows or dummy
columns with zero cost elements in the cost matrix, depending upon whether m<n or m>n and
then solved by the Hungarian method.
RESTRICTIONS OR PROHIBITED IN ASSIGNMENT PROBLEMS:
Sometimes due to certain reason, a particular resource (say a man or machine) cannot be
assigned to perform a particular activity (say job). In such cases, the cost of performing that
particular activity by a particular resource is considered to be very large. (Written as M or
infinity) so as to prohibit the entry of this pair of resource activity into the final solution.
MAXIMIZATION IN A.P:
Step 1: When an A.P involves maximizing objective function, such a problem may be solved by
converting it into a minimization problem in either of the following two cases:
CASE 1: Locate the largest pay of element in the assignment table and then subtract all
the elements of the elements of the table from this largest element.
STEP 2: The transferred assignment problem so obtained can be solved by using Hungarian
method.

16

Dr. B. Venkateswarlu

vit

DIFFERENCE BETWEEN TRANSPORTATION AND ASSIGNMENT PROBLEMS

TRANSPORTATION PROBLEM
1.Supply at any source may be any
positive quantity ai.
2. Demand at any destination may be any
positive quantity bj.
3. One or more source to any number of
destinations.
4. Transhipment problem is its.
5. A possibility of degeneracy either at
initial stage or subsequent stages.

ASSIGNMENT PROBLEM
1. Supply at any source (machine) will be
one i.e., ai=1.
2.Demand at any destination may be any
positive quantity bj.
3.One or more source to any number of
destinations.
4.Travelling salesman problem and crew
assignment are its extension.
5.There will not be any Degeneracy in this
problem.

17

Dr. B. Venkateswarlu

vit

TRANS-SHIPMENT PROBLEMS:
A Transportation problem in which available commodity frequently moves from
one source to another source (or) destination before reaching its actual destination is
called a Transshipment problem. Such a problem cannot be solved as such by the
usual Transportation algorithm, but slight modification is required before applying it to
the transhipment problem.

CHARACTERISTICS OF TRANSHIPMENT PROBLEM:

1. A number of sources (m) and destinations (n) in T.P. are m and n respectively. In
transshipment study, however we have (m+n) source and destinations.
2. If Si denotes the ith sources and Dj denotes the jth destinations then commodity
can move along the route.
Si->Di->Dj, Si->Sj->Di->Dj, and Si->Di->Si->Dj, or in the various other ways
Clearly transportation cost from Si to Si is Zero. And the transportation cost from
Si to Sj or Si to Di do not have the symmetrical.
3. In solving the transshipment problem, we first find the optimum solution to the
transportation problem and proceed in the same fashion as in solving the T.P.

TRANSPORTATION

TRANSSHIPMENT

S1

D1

S1

D1

S2

D2

S2

D2

NOTE: The total number of units transported is N, then add this N at supply and
demand at each source as well as destination. Then the problem is solved as
usually with the rules of T.P.

18

Dr. B. Venkateswarlu

vit

TO FIND THE OPTIMUM DUAL SOLUTION BY USING SIMPLEX METHOD:

RULE 1: Corresponding not evaluations of the starting primal variables =
Difference b/w the left and right sides of the dual constraints associated with the
dual starting variable.
RULE 2: Negative of the corresponding net evaluation of the starting dual Variables =
Difference b/w the left and right sides of the primal constraints associated with
the dual starting variables.
RULE 3: If the primal (dual)problem is unbounded then the dual(primal)problem doesnt have
any feasible solution.

19