90

Transportation Problems

IMPORTANT TOPICS COVERED

After studying this chapter, you will be able to understand :
å Introduction
å Important Considerations of Transportation Modes
å Transportation Algorithm (or) MODI Method (modified distribution method) (Test for
optimal solution).
å Degeneracy in Transportation Problems
å Unbalanced Transportation Problems
å Maximization Case in Transportation Problems

Introduction
The problem of transportation or distribution arises due to shipment of goods to the
destination of their requirement from various sources of the origin. The transportation or
distribution model is also a part of linear programming as it also inherits the objective of
minimization of cost or maximization of profit. In the case of large scale production an industry
may have production centres at various locations where from the goods are sent to the
warehouses or godowns which further send these goods to the distribution centres for the
supply to customers.
Generally in transportation or distribution model the procedure of sending goods from
one destination to another is pre-determined along with their respective cost of transportation.
The transportation or distribution model mainly deals that how the cost of
transportation can be minimized or the revenue of transportation can be maximized
by satisfying the requirements of various destinations within the known constraints
of different sources of supply.
The transportation or distribution model was originally originated by F.L. Hitchcock in
1941 in his work ‘the distribution of a product from several sources to numerous locations.’ It
was subsequently developed in 1947 by T.C. Koopans in his work ‘optimum utilization of
transportation system’ and further developed by Dantzig in 1951 towards the formulation and
solution of linear programming problem. It is why the transportation model is regarded as a
specific type of linear programming problem which analyse the transportation of certain
homogeneous goods or services from their different sources of orgin to their different
destinations of requirements.
Important Considerations of Transportation Modes
(1) The concerns using transportation model must have more than one distribution or
supply and requirement centre.
 91

(2) There should not be any restriction in transporting the goods from any source of
origin to any destination of requirement.
(3) The total number of a units available at various sources should be equal to the total
number of units required by various destinations. However, it those are not equal
then efforts are made to balance the demand and supply through unbalanced
transportation model.
(4) The transportation cost per unit of sending the goods from supply sources to
demand centres should be known which should not fluctuate according to quantity
and time.
(5) The demand centres and sources of supply to be studied in transportation model
should be specific and well defined.
Terminology Used
(1) Transportation Problem : It refers to the problem which is related with
reduction in transportation cost of transporting goods from source of origin to a
specific destination of their demand.
(2) Feasible Solution : It refers to the set of positive allocation of goods or services
which simultaneously removes deficiencies. The feasible solution is divided into
two parts :
(a) Basic Feasible Solution : Basic feasible solution of a transportation problem
is that where ‘m’ sources and ‘n’ destinations in which the non-negative
allocation is one less than the sum of rows and columns.
(b) Optimum Feasible Solution : Optimum feasible solution is that which
minimizes the cost or maximizes the revenue of a transportation problem.
(3) Transportation Table : It refers to the table which shows the goods to be
transported from ‘m’ sources of supply to ‘n’ destinations with their respective
transportation cost per unit. This table is used to find the basic feasible or optimum
feasible solution. The format of transportation table is given is under :
Transportation Table
Production Distribution Centres Capacities of
Centres Supply (m)
X Y Z
(Sources)
A Cost per Cost per Cost per
unit unit unit

B Cost per Cost per Cost per

unit unit unit

Demand Rim Total Supply

( n) Requirements
Total Demand
 92

In the above table A and B are the source centres while X, Y, and Z are the demand
centres. Cost per unit is shown is small cells. In supply column ( m) the capacities of
supply of various source centres is exhibited and in the demand row ( n) the rim,
requirements of various demand centres is presented. In the last column or row the
total of supply and demand is shown.
(4) Cells : It is a square that shows the cost or revenue per unit of goods transported
from a source to a destination.
(5) Supply Sources : It shows the quantity of goods produced by production centres
which is available for distribution to the various destinations of their requirements.
(6) Demand Sources : The quantity demand by various destinations is shown in ‘n’
rows. In this source we show total no. of units demanded by each demand centre.
(7) Rim Requirements : The quantity required for distribution is known as rim
requirements.
I. Mathematical Formulation of a Transportation Problem :
Let us assume that there are m sources and n destinations.
Let a i be the supply (capacity) at source i, b j be the demand at destination j, c ij be the
unit transportation cost from source i to destination j and x ij be the number of units shifted
from source i to destination j.
Then the transportation problem can be expressed mathematically as
m n
Minimize Z = å å c ij x if
i=1 j=1
subject to the constraints
n
å x ij = ai , i = 1, 2, 3 ¼ m.
j=1
m
å x ij = bj , j = 1, 2, 3 ¼ n.
i=1
and x ij ³ 0, for all i and j.
Note 1 : The two sets of constraints will be consistent if
m n
å ai = å bj
i=1 j=1
(total supply) (total demand)
which is the necessary and sufficient condition for a transportation problem to have a
feasible solution. Problems satisfying this condition are called balanced transportation
problems.
Note 2 : If å
ai ¹ å
b j , then the transportation problem is said to be unbalanced.
Note 3 : For any transportation problem, the coefficients of all x ij in the constraints are
unity.
 93

Note 4 : The objective function and the constraints being all linear, the transportation
problem is a special class of linear programming problem. Therefore it can be solved by simplex
method. But the number of variables being large, there will be too many calculations. So we can
look for some other technique which would be simpler than the usual simplex method.
Standard Transportation Table :
Transportation problem is explicitly represented by the following transportation table.
Destination
D1 D2 D3 … Dj … Dn Supply

S1 c11 c12 c13 c1 j c1 n a1

Source S2 c 21 c 22 c 23 c2 j c 2n a2
:
:

Si c i1 c i2 c ij c in :
:

Sm c m1 c m2 c mj c mn am

Demand b1 b2 b3 … … … bn Sa i = Sb j

The mn squares are called cells. The unit transportation cost c ij from the i th source to the
j th destination is displayed in the upper left side of the (i, j) th cell. Any feasible solution is
shown in the table by entering the value of x ij in the centre of the ( i, j) th cell. The various a’s
and b’s are called rim requirements. The feasibility of a solution can be verified by summing
the values of x ij along the rows and down the columns.
Definition 1 : A set of non-negative values x ij , i = 1, 2, ¼m; j = 1, 2 ¼ n. that satisfies the
constraints (rim conditions and also the non-negativity restrictions) is called a feasible
solution to the transportation problem.
Note : A balanced transportation problem will always have a feasible solution.
Definition 2 : A feasible solution to a ( m ´ n) transportation problem that contains no
more than m + n - 1 non-negative allocations is called a basic feasible solution (BFS) to the
transportation problem.
The allocations are said to be in independent positions if it is impossible to increase or
decrease any allocation without either changing the position of the allocation or violating the
rim requirements. A simple rule for allocations to be in independent positions is that it is
impossible to travel from any allocation, back to itself by a series of horizontal and vertical
jumps from one occupied cell to another, without a direct reversal of the route. Example
 94
Non-independent positions Non-independent positions

(i) (ii)

Non-independent positions Independent positions

(i) (ii)
Definition 3 : A basic feasible solution to a ( m ´ n) transportation problem is said to be a
non-degenerate basic feasible solution if it contains exactly m + n - 1 non-negative
allocations in independent positions.
Definition 4 : A basic feasible solution that contains less than m + n - 1 non-negative
allocations is said to be degenerate basic feasible solution.
Definition 5 : A feasible solution (not necessarily basic) is said to be an optimal
solution if it minimizes the total transportation cost.
Note : The number of non-basic variables in an m ´ n balanced transportation problem is
atmost m + n - 1.
Note : The number of non-basic variables in an m ´ n balanced transportation problem is
atleast mn - ( m + n - 1)
II. Methods for Finding Initial Basic Feasible Solution
The transportation problem has a solution if and only if the problem is
balanced. Therefore before starting to find the initial basic feasible solution, check
whether the given transportation problem is balanced. If not one has to balance the
transportation problem first. The way of doing this is discussed in section 7.4 page 7.40. In this
section all the given transportation problems are balanced.
Method 1 : North west corner rule :
Step 1 : The first assignment is made in the cell occupying the upper left-hand
(north-west) corner of the transportation table. The maximum possible amount is allocated
there. That is x 11 = min {a1 , b1 }.
Case (i) : If min {a1 , b1 } = a1 , then put x 11 = a1 , decrease b1 by a1 and move
vertically to the 2nd row (i.e.,) to the cell (2, 1) cross out the first row.
Case (ii) : If min {a1 , b1 } = b1 , then put x 11 = b1 , and decrease a1 by b1 and move
horizontally right (i.e.,) to the cell (1, 2) cross out the first column.
Case (iii) : If min {a1 , b1 } = a1 = b1 then put x 11 = a1 = b1 and move diagonally to
the cell (2, 2) cross out the first row and the first column.
 95

Step 2 : Repeat the procedure until all the rim requirements are satisfied.
Method 2 : Least cost method (or) Matrix minima method (or) Lowest cost entry
method :
Step 1 : Identify the cell with smallest cost and allocate x ij = Min {a i , b j }
Case (i) : If min {a i , b j } = a i , then put x ij = a i , cross out the i th row and decrease bi by a i ,
Go to step (2).
Case (ii) : If min {a i , b j } = b j then put x ij = b j cross out the j th column and decrease a i
by b j Go to step (2).
Case (iii) : If min {a i , b j } = a i = b j , then put x ij = a i = b j , cross out either i th row or j th
column but not both, Go to step (2).
Step 2 : Repeat step (1) for the resulting reduced transportation table until all the rim
requirements are satisfied.
Method 3 : Vogel’s approximation method (VAM) (or) Unit cost penalty method :
Step 1 : Find the difference (penalty) between the smallest and next smallest costs in
each row (column) and write then in brackets against the corresponding row (column).
Step 2 : Identify the row (or) column with largest penalty. If a tie occurs, break the tie
arbitrarily. Choose the cell with smallest cost in that selected row or column and allocate as
much as possible to this cell and cross out the satisfied row or column and go to step (3).
Step 3 : Again compute the column and row penalties for the reduced transportation
table and then go to step (2). Repeat the procedure until all the rim requirements are satisfied.
* Example 1 : Determine basic feasible solution to the following transportation
problem using North West Corner Rule :
Sink
A B C D E Supply
P 2 11 10 3 7 4
Origin Q 1 4 7 2 1 8
R 3 9 4 8 12 9
Demand 3 3 4 5 6

Solution : Since a i = b j = 21, the given problem is balanced.

\ There exists a feasible solution to the transportation problem.
2 11 10 3 7 4
3

1 4 7 2 1 8

3 9 4 8 12 9

3 3 4 5 6
 96

Following North west corner rule, the first allocation is made in the cell (1, 1).
Here x 11 = min {a1 , b1 } = min {4, 3} = 3
\ Allocate 3 to the cell (1, 1) and decrease 4 by 3, i.e., 4 - 3 = 1
As the first column is satisfied, we cross out the first column and the resulting reduced
Transportation table is
11 10 3 7 1
1 1
4 7 2 1 8

9 4 8 12 9

3 4 5 6

Here the North west corner cell (1, 2).

So Allocate x 12 = min {1, 3} = 1 to the cell (1, 2) and move vertically to cell (2, 2). The
resulting reduced transportation table is
4 7 2 1 8
2
9 4 8 12 9

2 4 5 6

Allocate x 22 = min {8, 2} = 2 to the cell (2, 2) and move horizontally to the cell (2, 3). The
resulting transportation table is
7 2 1 6
4
4 8 12 9
4 5 6

Allocate x 23 = min {6, 4} = 4 and move horizontally to the cell (2, 4).
The resulting reduced transportation table is
2 1 2
2
8 12 9
5 6

Allocate x 24 = {2, 5} = 2 and move vertically to the cell (3, 4). The resulting reduced
transportation table is
 97

8 12 9
3

3 6

Allocate x 34 = min {9, 3} = 3 and move horizontally to the cell (3, 5), which is
12
6 6

6
Allocate x 35 = min {6, 6} = 6
Finally the initial basic feasible solution is as shown in the following table.

2 11 10 3 7
3 1
1 4 7 2 1
2 4 2

3 9 4 8 12
3 6

From this table we see that the number of positive independent allocations is equal to
m + n - 1 = 3 + 5 - 1 = 7. This ensures that the solution is non degenerate basic feasible.
\ The initial transportation cost
= Rs. 2 ´ 3 + 11 ´ 1 + 4 ´ 2 + 7 ´ 4 + 2 ´ 2 + 8 ´ 3 + 12 ´ 6
= Rs. 153 / -
* Example 2 : Find the initial basic feasible solution for the following
transportation problem by Least Cost Method.

To Supply
1 2 1 4 30
From 3 3 2 1 50
4 2 5 9 20
Demand 20 40 30 10

Solution : Since Sa i = Sb j = 100, the given TPP is balanced.

\ The exists a feasible solution to the transportation problem.
 98

1 2 1 4 30
20
3 3 2 1 50

4 2 5 9 20

20 40 30 10

By least cost method, min c ij = c11 = c13 = c 24 = 1

Since more than one cell having the same minimum c ij , break the tie.
Let us choose the cell (1, 1) and allocate x 11 = min {a1 , b1 } = min {30, 20} = 20 and cross
out the satisfied column and decrease 30 by 20.
The resulting reduced transportation table is
2 1 4
10 10

3 2 1 50

2 5 9 20

40 30 10

Here min c ij = c13 = c 24 = 1

Choose the cell (1, 3) and allocate x 13 = min {a1 , b3} = min {10, 30} = 10 and cross out the
satisfied row.
The resulting reduced transportation table is
3 2 1
10 50

2 5 9 20

40 20 10

Here min c ij = c 24 = 1,
\ Allocate x 24 = min {a 2 , b4} = min (50, 10) = 10 and cross out the satisfied column.
The resulting transportation table is
3 2
20 40

2 5 20

40 20
 99

Here min c ij = c 23 = c 32 = 2. Choose the cell (2, 3) and allocate x 23 = min {a 2 , b3} = min
(40, 20) = 20 and cross out the satisfied column.
The resulting reduced transportation table is
3
20
2
20 20

40
Here min c ij = c 32 = 2. Choose the cell (3, 2) and allocate
x 32 = min {a 3 , b2} = min (20, 40) = 20 and cross out the satisfied row.
The resulting reduced transportation table is
3
20 20

20

Finally the initial basic feasible solution is as shown in the following table.
1 2 1 4
20 10
3 3 2 1
20 20 10
4 2 5 9
20

From this table we see that the number of positive independent allocations is equal to
m + n - 1 = 3 + 4 - 1 = 6. This ensures that the solution is non degenerate basic feasible.
\ The initial transportation cost = Rs. 1 ´ 20 + 1 ´ 10 + 3 ´ 20
+ 2 ´ 20 + 1 ´ 10 + 2 ´ 20
= 20 + 10 + 60 + 40 + 10 + 40
= Rs. 180/-
* Example 3 : Find the initial basic feasible solution for the following
transportation problem by VAM.
Distribution Centres
D1 D2 D3 D4 Availability
S1 11 13 17 14 250
Origin S2 16 18 14 10 300
S3 21 24 13 10 400
Requirements 200 225 275 250
 100

Solution : Since Sa i = Sb j = 950, the given problem is balanced.

\ There exists a feasible solution to this problem.
11 13 17 14 250 (2)
200
16 18 14 10 300 (4)

21 24 13 10 400 (3)

200 225 275 250

(5) (5) (1) (0)
First let us find the difference (penalty) between the smallest and next smallest costs in
each row and column and write them in brackets against the respective rows and columns.
The largest of these differences is (5) and is associated with the first two columns of the
transportation table. We choose the first column arbitrarily.
In this selected column, the cell (1, 1) has the minimum unit transportation cost c11 = 11.
\ Allocate x 11 = min {250, 200} = 200 to this cell (1, 1) and decrease 250 by 200 and cross
out the satisfied column.
The resulting reduced transportation table is
13 17 14 50 (1)
50
18 14 10 300 (4)
24 13 10 400 (3)
225 275 250
(5) (1) (0)
The row and column differences are now computed for this reduced transportation table.
The largest of these is (5) which is associated with the second column. Since c12 = 13 is the
minimum cost, we allocate x 12 = min {50, 225} = 50 to the cell (1, 2) and decrease 225 by 50 and
cross out the satisfied row.
Continuing in this manner, the subsequent reduced transportation tables and the
differences for the surviving rows and columns are shown below :
18 14 10 300 (4)

175

24 13 10 400 (3)

175 275 250

(6) (1) (0)
(i)
 101

14 10 125 (4)
125
13 10 400 (3)

250
(1) (0)
(ii)

13 10 400

125

275 125
(iii)

13 275

275

275
(iv)
Finally the initial basic feasible solution is as shown in the following table.
11 13 17 14
200 50
16 18 14 10
175 125
21 24 13 10
275 125

From this table we see that the number of positive independent allocations is equal to
m + n - 1 = 3 + 4 - 1 = 6. This ensures that the solution is non degenerate basic feasible.
\ The initial = Rs. 11 ´ 200 + 13 ´ 50 + 18 ´ 175
transportation cost
+ 10 ´ 125 + 13 ´ 275 + 10 ´ 125
= Rs. 12075/-
 102

Transportation Algorithm (or) MODI Method (Modified

Distribution Method) (Test for Optimal Solution)
Step 1 : Find the initial basic feasible solution of the given problem by Northwest Corner
rule (or) Least Cost method of VAM.

Step 2 : Check the number of occupied cells. If these are less than m + n - 1, there exists
degeneracy and we introduce a very small positive assignment of Î ( » 0) in suitable independent
positions, so that the number of occupied cells is exactly equal to m + n - 1.
Step 3 : Find the set of values ui , v j ( i = 1, 2, 3, ¼ m; j = 1, 2, 3 ¼ n) form the relation
c ij = ui + v j for each occupied cell ( i, j), by starting initially with ui = 0 or v j = 0 preferably for
which the corresponding row or column has maximum number of individual allocations.
Step 4 : Find ui + v j for each unoccupied cell ( i, j) and enter at the upper right corner of
the corresponding cell ( i, j).
Step 5 : Find the cell evaluations dij = c ij - (ui + v j ) ( dij = upper left-upper right) for
each unoccupied cell ( i, j) and enter at the lower right corner of the corresponding cell ( i, j).
Step 6 : Examine the cell evaluations dij for all unoccupied cells ( i, j) and conclude that
(i) if all dij > 0, then the solution under the test is optimal and unique.
(ii) if all dij > 0, with atleast one dij = 0, then the solution under the test is optimal and
an alternative optimal solution exists.
(iii) if atleast one dij < 0, then the solution is not optimal. Go to the next step.

Step 7 : Form a new B.F.S. by giving maximum allocation to the cell for which dij is most
negative by making an occupied cell empty. For that draw a closed path consisting of horizontal
and vertical lines beginning and ending at the cell for which dij is most negative and having its
other corners at some allocated cells. Along this closed loop indicate + q and - q alternatively at
the corners. Choose minimum of the allocations from the cells having - q. Add this minimum
allocation to the cells with + q and subtract this minimum allocation from the allocation to the
cells with- q.
Step 8 : Repeat steps (2) to (6) to test the optimality of this new basic feasible solution.
Step 9 : Continue the above procedure till an optimum solution is attained.
Note : The Vogels approximation method (VAM) takes into account not only the
least cost c ij but also the costs that just exceed the least cost c ij and therefore yields
better initial solution than obtained from other methods in general. This can be justified by the
above example (4). So to find the initial solution, give preference to VAM unless otherwise
specified.
* Example 4 : Solve the transportation problem :
 103

1 2 3 4 Supply

I 21 16 25 13 11

II 17 18 14 23 13

III 32 27 18 41 19

Demand 6 10 12 15

Solution : Since Sa i = Sb j = 43, the given transportation problem is balanced. \ There

exists a basic feasible solution to this problem.
By Vogel’s approximation method, the initial solution is as shown in the following table :
21 16 25 13 (3) – – –
11

17 18 14 23 (3) (3) (3) (4)

6 3 4

32 27 18 41 (9) (9) (9) (9)

7 12

(4) (2) (4) (10)

(15) (9) (4) (18)
(15) (9) (4) —
— (9) (4) —
That is,
21 16 25 13
11

17 18 14 23
7 3 4

32 27 18 41
7 12

From this table, we see that the number of non-negative independent allocations is
( m + n - 1) = (3 + 4 - 1) = 6. Hence the solution is non-degnerate basic feasible.
 104

\ The inition transportation cost

= Rs. 13 ´ 11 + 17 ´ 6 + 18 ´ 3 + 23 ´ 4 + 27 ´ 7 + 18 ´ 12
= Rs. 796/-
To find the optimal solution
Consider the above transportation table. Since m + n - 1 = 6, we apply MODI method.
Now we determine a set of values ui and v j for each occupied cell ( i, j) by using the
relation c ij = ui + v j . As the 2nd row contains maximum number of allocations, we choose
u2 = 0.
Therefore
c 21 = u2 + v1 Þ 17 = 0 + v1 Þ v1 = 17
c 22 = u2 + v 2 Þ 18 = 0 + v 2 Þ v 2 = 18
c 24 = u2 + v 4 Þ 23 = 0 + v 4 Þ v 4 = 23
c14 = u1 + v 4 Þ 13 = u1 + 23 Þ u1 = - 10
c 32 = u3 + v 2 Þ 27 = u3 + 18 Þ u3 = 9
c 33 = u3 + v 3 Þ 18 = 9 + v 3 Þ v3 = 9
Thus we have the following transportation table :
21 16 25 13
11 u1 = - 10

17 18 14 23
6 3 4 u2 = 0
32 27 18 41
7 12 u3 = 9

v1 = 17 v 2 = 18 v3 = 9 v 4 = 23

Now we find ui + v j for each unoccupied cell ( i, j) and enter at the upper right corner of
the corresponding unoccupied cell ( i, j).
Then we find the cell evaluations dij = c ij - (ui + v j ) (i. e., upper left corner – upper right
corner) for each unoccupied cell ( i, j) and enter at the lower right corner of the corresponding
unoccupied cell ( i, j)

Thus we get the following table :

 105

21 7 16 8 25 - 1 13
11 u1 = - 10
14 8 26
17 18 14 9 23
6 3 4 u2 = 0
5
32 26 27 18 41 32
7 12 u3 = 9
6 9
v1 = 17 v 2 = 18 v3 = 9 v 4 = 23
Since all dij > 0, the solution under the test is optimal and unique.
\ The optimum allocation schedule is given by x 14 = 11, x 21 = 6, x 22 = 3, x 24 = 4,
x 32 = 7, x 33 = 12 and the optimum (minimum) transportation cost.
= Rs. 13 ´ 11 + 17 ´ 6 + 18 ´ 3 + 23 ´ 4 + 27 ´ 7 + 18 ´ 12
= Rs. 796/-
* Example 5 : Obtain on optimum basic feasible solution to the following
transportation problem :
To Available
7 3 2 2
Form 2 1 3 3
3 4 6 5
Demand 4 1 5 10
Solution : Since Sa i = Sb j = 10, the given transportation problem is balanced. \There
exists a basic feasible solution to this problem.
By Vogel’s approximation method, the initial solution is as shown in the following table :
7 3 2 (1) (5) –
2

2 1 3 (1) (1) (1)

1 2
3 4 6 (1) (3) (3)
4 1

(1) (2) (1)

(1) – (1)
(1) – (3)
That is,
 106

7 3 2
2
2 1 3
1 2
3 4 6
4 1

From this table we see that the number of non-negative allocations is

m + n - 1 = (3 + 3 - 1) = 5.
Hence the solution is non-degenerate basic feasible
\The initial transportation cost
= Rs. 2 ´ 2 + 1 ´ 1 + 3 ´ 2 + 3 ´ 4 + 6 ´ 1
= Rs. 29/-
For optimality : Since the number of non-negative independent allocations is m + n - 1,
we apply MODI method.
Since the third column contains maximum number of allocations, we choose v 3 = 0.
Now we determine a set of values ui and v j by using the occupied cells and the relation
c ij = ui + v i .
That is,
7 -1 3 0 2
2 u1 = 2

2 1 3
1 2 u2 = 3

3 4 6
4 1 u3 = 6

v1 = - 3 v2 = - 2 v3 = 0

Now we find ui + v j for each unoccupied cell ( i, j) and enter at the upper right corner of
the corresponding unoccupied cell ( i, j).
Then we find the cell evaluations dij = c ij - (ui + v j ) for each unoccupied cell ( i, j) and
enter at the lower right corner of the corresponding unoccupied cell ( i, j).
Thus we get the following table :
 107

7 -1 3 0 2
2 u1 = 2

8 3
2 0 1 3
1 2 u2 = 3
2
3 4 4 6
4 0 1 u3 = 6

v1 = - 3 v2 = - 2 v3 = 0
Since all dij > 0, with d32 = 0, the current solution is optimal and there exists an
alternative optimal solutuion.
\ The optimal allocation schedule is given by x 13 = 2, x 22 = 1, x 23 = 2, x 31 = 4, x 33 = 1
and the optimum (minimum) transportation cost
= Rs. 2 ´ 2 + 1 ´ 1 + 3 ´ 2 + 3 ´ 4 + 6 ´ 1 = Rs. 29/ -
* Example 6. Find the optimal transportation cost of the following matrix using
least cost method for finding the critical Solution.
Market
A B C D E Available
P 4 1 2 6 9 100
Factory Q 6 4 3 5 7 120
R 5 2 6 4 8 120
Demand 40 50 70 90 90

Solution : Since S a i = S b j = 340, the transportation problem is balanced.

\ There exists a basic feasible solution to this problem.
By using Least cost method, the initial solution is as shown in the following table :
4 1 2 6 9
50 50
6 4 3 5 7
10 20 90
5 2 6 4 8
30 90

\ The initial transportation cost

= Rs. 1 ´ 50 + 2 ´ 50 + 6 ´ 10 + 3 ´ 20 + 7 ´ 90 + 5 ´ 30 + 4 ´ 90
= Rs. 1410/-
 108

For optimality : Since the number of non-negative independent allocations is

( m + n - 1), we apply MODI method :
That is,
4 5 1 2 6 4 9 6
50 50
-1 2 3 u1 = - 1
6 4 2 3 5 5 7
10 20 90 u2 = 0
2 0
5 2 1 6 2 4 8 6
30 90 u3 = - 1
1 4 2
v1 = 6 v2 = 2 v3 = 3 v4 = 5 v5 = 7

Since d11 = - 1 < 0, the current solution is not optimal.

Now, let us form a new basic feasible solution by giving maximum allocation to the cell
( i, j) for which dij is most negative by making an occupied cell empty. Here the cell (1, 1) having
the negative value d11 = - 1. We draw a closed loop consisting of horizontal and vertical lines
beginning and ending at this cell (1, 1) and having its other corners at some occupied cells.
Along this closed loop indicate + q and - q alternatively at the corners, we have
4 1 2 6 9
+q 50 50 - q

6 4 3 5 7
10 20 90
-q +q
5 2 6 4 8
30 90

From the two cells (1, 3), (2, 1) having - q, we find that the minimum of the allocations 50,
10 is 10. Add this 10 to the cells with + q and subtract this 10 to the cells with - q.
Hence the new basic feasible solution is displayed in the following table :
4 1 2 6 9
10 50 40

6 4 3 5 7
30 90

5 2 6 4 8
30 90
 109

We see that the above table satisfies the rim conditions with ( m + n - 1) non-negative
allocations at independent positions. So we apply MODI method.
4 1 2 6 3 9 6
10 50 40 u1 = 0
3 3
6 5 4 2 3 5 4 7
1 2 30 1 90 u2 = 1
5 2 2 6 3 4 8 7
30 90 u3 = 1
0 3 1
v1 = 4 v2 = 1 v3 = 2 v4 = 3 v5 = 6

Since all dij > 0, with d32 = 0, the current solution is optimal and there exists an
alternative optimal solution.
\ The optimum allocation schedule is given by x 11 = 10, x 12 = 50, x 13 = 40, x 23 = 30,
x 25 = 90, x 31 = 30, x 34 = 90 and the optimum (minimum) transportation cost
= Rs. 4 ´ 10 + 1 ´ 50 + 2 ´ 40 + 3 ´ 30 + 7 ´ 90 + 5 ´ 30 + 4 ´ 90
= Rs. 1400/-
Degeneracy in Transportation Problems
In a transportation problem, whenever the number of non-negative independent
allocations is less then m + n - 1, the transportation problem is said to be a degenerate one.
Degeneracy may occur either at the initial stage or at an intermediate state at some subsequent
iteration.
To resolve degeneracy, we allocate an extremely small amount (close to zero) to one or
more empty cells of the transportation table (generally minimum cost cells if possible), so that
the total number of occupied cells becomes ( m + n - 1) at independent positions. We denote this
small amount by Î (epsilon) satisfying the following conditions :
(i) 0 < Î < x ij , for all x ij > 0
(ii) x ij ± Î = x ij , for all x ij > 0
The cells containing Î are then treated like other occupied cells and the problem is solved
in the usual way. The Βs are kept till the optimum solution is attained. Then we let each ή 0.
* Example 7 : Find the non-degenerate basic feasible solution for the following
transportation problem using
(i) North west corner rule
(ii) Least cost method
(iii) Vogle’s approximation method.
 110

To Supply
10 20 5 7 10
13 9 12 8 20
From 4 5 7 9 30
14 7 1 0 40
3 12 5 19 50
Demand 60 60 20 10

Solution : Since S a i = S b j = 150, the given transportation problem is balanced.

\ There exists a basic feasible solution to this problem.

(i) The starting solution by NWC rule is as shown in the following table.
10 20 5 7
10
13 9 12 8
20
4 5 7 9
30
14 7 1 0
40
3 12 5 19
20 20 10

Since the number of non-negative allocations at independent positions is 7 which is less

than ( m + n - 1) = ( 5 + 4 - 1) = 8, this basic feasible solution is a degenerate one.
To resolve this degeneracy, we allocate a very small quantity Î to the unoccupied cell (5,
1) so that the number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic
feasible solution is as shown in the following table.
10 20 5 7
10
13 9 12 8
20
4 5 7 9
30
 111

14 7 1 0
40
3 12 5 19
Î 20 20 10

The initial transportation cost = Rs. 10 ´ 10 + 13 ´ 20 + 4 ´ 30

+ 7 ´ 40 + 3 ´ Î + 12 ´ 20
+ 5 ´ 20 + 19 ´ 10
= Rs. (1290 + 3 Î)
= Rs. 1290/-, as Î ® 0.
(ii) Least cost method : Using this method the starting solution is as shown in the
following table :
10 20 5 7
10

13 9 12 8
20

4 5 7 9
10 20

14 7 1 0
10 20 10

3 12 5 19
50

Since the number of non-negative allocations at independent positions is ( m + n - 1) = 8,

the solution is non-degenerate basic feasible.
The initial transportation cost = Rs. 20 ´ 10 + 9 ´ 20 + 4 ´ 10
+ 5 ´ 20 + 7 ´ 10 + 1 ´ 20 + 0 ´ 10 + 3 ´ 50
= Rs. 760/-
(iii) Vogel’s approximation Method : The starting solution by this method is as shown
in the following table :
 112

10 20 5 7
10

13 9 12 8
20

4 5 7 9
30

14 7 1 0
10 20 10

3 12 5 19
50

Since the number of non-negative allocations is 7 which is less than ( m + n - 1)

= ( 5 + 4 - 1) = 8, this basic solution is a degenerate one.
To resolve this degeneracy, we allocate a very small quantity Î to the unoccupied cell (5,
2) so that the number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic
feasible solution is as shown in the following table.

10 20 5 7
10

13 9 12 8
20

4 5 7 9
30

14 7 1 0
10 20 10

3 12 5 19
50 Î

\The initial transportation cost

= Rs. 10 ´ 10 + 9 ´ 20 + 5 ´ 30 + 7 ´ 10 + 1 ´ 20 + 0 ´ 10 + 3 ´ 50 + 12 ´ Î
= Rs. (670 + 12 Î)
= Rs. 670/- = as Î ® 0.
* Example 8 : Solve the following transportation problem using Vogel’s
method.
 113

Warehouse
A B C D E F Available
1 9 12 9 6 9 10 5
Factory 2 7 3 7 7 5 5 6
3 6 5 9 11 3 11 2
4 6 8 11 2 2 10 9
Requirement 4 4 6 2 4 2

Solution : Since S a i = S b j = 22, the given transportation problem is balanced.

\There exists a basic feasible solution to this problem. By Vogel’s approximation method, the
initial solution is as shown in the following table :
9 12 9 6 9 10
5

7 3 7 7 5 5
4 2

6 5 9 11 3 11
1 1
6 8 11 2 2 10
3 2 4

Since the number of non-negative allocations is 8 which is less than ( m + n - 1)

= ( 4 + 6 - 1) = 9, this basic solution is degenerate one.
To resolve degeneracy, we allocate a very small quantity Î to the cell (3, 2), so that the
number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic feasible solution
is as shown in the following table.
9 12 9 6 9 10
5
7 3 7 7 5 5
4 2

6 5 9 11 3 11
1 Î 1
6 8 11 2 2 10
3 2 4
 114

The initial transportation cost = Rs. 9 ´ 5 + 3 ´ 4 + 5 ´ 2 + 6 ´ 1

+ 5´ Î+ 9 ´ 1 + 6 ´ 3 + 2´ 2 + 2´ 4
= Rs. (112 + 5 Î) = Rs. 112 /-, Î ® 0.
To find the optimal solution
Now the number of non-negative allocations at independent positions is ( m + n - 1). We
apply the MODI method.
9 6 12 5 9 6 2 9 2 10 7 u =0
1
5
3 7 4 7 3
7 4 3 7 7 7 0 5 0 5 u2 = - 2
4 2
3 0 7 5
6 5 9 11 2 3 2 11 7 u3 = 0
1 Î 1
9 1 4
6 8 5 11 9 2 2 10 7 u4 = 0
3 2 4
3 2 3
v1 = 6 v2 = 5 v3 = 9 v4 = 2 v5 = 2 v6 = 7

Since all dij > 0 with d23 = 0, the solution under the test is optimal and an alternative
optimal solution is also exists.
\The optimum allocation schedule is given by x 13 = 5, x 22 = 4, x 26 = 2,x 31 = 1, x 32 = Î,
x 33 = 1, x 41 = 3, x 44 = 2, x 45 = 4 and the optimum (minimum) transportation cost is
= Rs. 9 ´ 5 + 3 ´ 4 + 5 ´ 2 + 6 ´ 1 + 5 ´ Î + 9 ´ 1 + 6 ´ 3 + 2 ´ 2 + 2 ´ 4
= Rs. (112 + 5 Î)
= Rs. 112, as Î ® 0.
* Example 9 : Solve the transportation problem :
To

Supply

1 2 3 4 6
Form 4 3 2 0 8

0 2 2 1 10
Demand 4 6 8 6
 115

Solution : Since S a i = S b j = 24, the given transportation problem is balanced.

\There exists a basic feasible solution.

By using Vogel’s approximation method, the initial solution is as shown in the following
table :
1 2 3 4
6

4 3 2 0
2 6

0 2 2 1
4 6

4
Since the number of non-negative allocations at independent positions is 5, which is less
then ( m + n - 1) = (3 + 4 - 1) = 6, this basic feasible solution is degenerate.
To resolve degeneracy, we allocate a very small quantity Î to the cell (3, 2) so that the
number of occupied cells becomes ( m + n - 1). Hence the non-degenerate initial basic feasible
solution is given by
1 2 3 4
6

4 3 2 0
2 6

0 2 2 1
4 Î 6

The initial transportation cost

= Rs. 2 ´ 6 + 2 ´ 2 + 0 ´ 6 + 0 ´ 4 + 2 ´ Î + 2 ´ 6
= Rs. ( 28 + 2 Î)
= Rs. 28/-, as Î ® 0.
To find the optimal solution
Now the number non-negative allocations at independent positions is ( m + n - 1). We
apply the MODI method.
 116

1 0 2 3 2 4 0
6 u1 = 0
1 1 4
4 0 3 2 2 0
2 6 u2 = 0
4 1
0 2 2 1 0
4 Î 6 u3 = 0
1

v1 = 0 v2 = 2 v3 = 2 v4 = 0

Since all dij > 0 the solution under the test is optimal and unique.
\ The optimum allocation schedule is given by x 12 = 6, x 23 = 2, x 24 = 6, x 31 = Î, x 32 = Î,
x 33 = 6 and the optimum (minimum) transportation cost.
= Rs. 2 ´ 6 + 2 ´ 2 + 0 ´ 6 + 0 ´ 4 + 2 ´ Î + 2 ´ 6
= Rs. ( 28 + 2 Î) = Rs. 28, as Î ® 0.
* Example 10 : Find the optimal solution of the following problem
Destination

X Y Z Supply
P 1 2 0 30
Origin Q 2 3 4 35
R 1 5 6 35
Demand 30 40 30

Solution : Since S a i = S b j = 100, the given transportation problem is balanced.

By using the Vogel’s approximation method, the basic feasible solution is displayed in the
following table.
1 2 0
30
2 3 4
35
1 5 6
30 5

Since the number of non-negative allocations at independent positions is 4 which is less

than( m + n - 1) = 3 + 3 - 1 = 5, this initial solution is degenerate.
 117

To resolve degeneracy we allocate a very small quantity Î to the cell (3, 3), so that the
number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic feasible solution
is given by

1 2 0
30
2 3 4
35
1 5 6
30 5 Î

Now the number of non-negative allocations at independent positions is ( m + n - 1) = 5.

We apply MODI method.
1 -5 2 -1 0
30 u1 = - 6
6 3
2 -1 3 4 4
35 u2 = - 2
3 0
1 5 6
30 5 Î u3 = 0
v1 = 1 v2 = 5 v3 = 6

Since all dij > 0 with d23 = 0, the solution under the test is optimal and there exists an
alternative optimal solution
\ The optimal allocation schedule is given by x 13 = 30, x 22 = 35, x 31 = 30, x 32 = 5,
x 33 = Î and the optimum (minimum) transportation cost.
= Rs. 0 ´ 30 + 3 ´ 35 + 1 ´ 30 + 5 ´ 5 + 6 ´ Î
= Rs. (160 + 6Î)
= Rs. 160/-, as Î ® 0.
* Example 11 : Solve the following transportation problem to minimize the total
cost of transportation.
Destination
1 2 3 4 Supply
1 14 56 48 27 70
Origin 2 82 35 21 81 47
3 99 31 71 63 93
Demand 70 35 45 60 210
 118

Solution : Since S a i = S b j = 210, the given transportation problem is balanced.

\There exists a basic feasible solution to this problem.
By using Vogel’s approximation method, the initial solution is as shown in the following
table :
14 56 48 27
70
82 35 21 81
45 2
99 31 71 63
35 58

Since the number of non-negative allocations is 5, which is less than ( m + n - 1)

= (3 + 4 - 1) = 6, this basic feasible solution is degenerate.
To resolve degeneracy, we allocate a very small quantity Î to the cell (1, 4), so that the
number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic feasible solution
is given in the following table
14 56 48 27
70 Î
82 35 21 81
45 2
99 31 71 63
35 58

To find the optimal solution

Now the number of non-negative allocations at independent positions is ( m + n - 1) = 6.
We apply MODI method.
14 56 -5 48 - 33 27
70 Î u1 = 27
61 81
82 68 35 49 21 81
45 2 u2 = 81
14 - 14
99 50 31 71 3 63
35 58 u3 = 63
49 68
v1 = - 13 v 2 = - 32 v 3 = - 60 v4 = 0

Since d22 = - 14 < 0, the solution under the test is not optimal.
 119

Now let us form a new basic feasible solution by giving maximum allocation to the cell (2,
2) by making an occupied cell empty. We draw a closed loop consisting of horizontal and vertical
lines beginning and ending at this cell (2, 2) and having its other corners at some occupied cells.
Along this closed loop, indicate + q and - q alternatively at the corners.
14 56 48 27
70 Î

82 35 21 81
+q 45 2
-q
99 31 71 63
35 +q
- q 58

From the two cells (2, 4), (3, 2) having - q we find that the minimum of the allocations 2,
35 is 2. Add this 2 to the cells with + q and subtract this 2 to the cells with - q. Hence the new
basic feasible solution is given by
14 56 48 27
70 Î
82 35 21 81
2 45
99 31 71 63
33 60

We see that the above table satisfies the rim conditions with ( m + n - 1) non-negative
allocations at independent position. We apply MODI method of optimality.
14 56 -5 48 - 19 27 u1 = - 40
70 Î
67
61
82 54 35 21 81 67 u2 = 0
2 45
28 14
99 50 31 71 17 63 u3 = - 4
33 60
49 54
v1 = 54 v 2 = 35 v 3 = 21 v 4 = 67

Since all dij > 0, the solution under the test is optimal.
 120

\The optimal allocation schedule is given by x 11 = 70, x 14 =Î, x 22 = 2, x 23 = 45,

x 32 = 33, x 34 = 60 and the optimum (minimum) transportation cost
= Rs. 14 ´ 70 + 27 ´ Î + 35 ´ 2 + 21 ´ 45 + 31 ´ 33 + 63 ´ 60
= Rs. 6798/-, as Î ® 0.
* Example 12 : Solve the following transportation problem, in which a i is the
availability at origin Oi and b j is the requirement at the destination D j and cell
entries are unit costs of transportation from any origin to any destination :

D1 D2 D3 D4 D5 ai

O1 4 7 3 8 2 4

O2 1 4 7 3 8 7

O3 7 2 4 7 7 9

O4 4 8 2 4 7 2

bj 8 3 7 2 2

Solution : Since S a i = S b j = 22, the given problem is balanced.

\ There exists a basic feasible solution to this problem.
By using Vogel’s approximation method, the initial solution is as shown in the following
table :
4 7 3 8 2
1 1 2
1 4 7 3 8
7
7 2 4 7 7
3 6
4 8 2 4 7
Î 2

Since the number of non-negative allocations is 7, which is less than ( m + n - 1)

= ( 4 + 5 - 1) = 8, this basic feasible solution is degenerate.
To resolve this degeneracy, we allocate a very small quantity Î to the cell (4, 3), so that
the number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic feasible
solution is given in the following table
 121

4 7 3 8 2
1 1 2

1 4 7 3 8
7

7 2 4 7 7
3 6

4 8 2 4 7
Î 2

To find the optimal solution : Now the number of non-negative allocations at

independent positions is ( m + n - 1) = 8. We apply MODI method.
4 7 1 3 8 5 2 u1 = 0
1 1 2
6 3
1 4 -2 7 0 3 2 8 -1 u = -3
2
7
6 7 1 9
7 5 2 4 7 6 7 3 u =1
3
3 6
2 1 4
4 3 8 0 2 4 7 1 u = -1
4
Î 2
1 8 6
v1 = 4 v2 = 1 v3 = 3 v4 = 5 v5 = 2

Since all dij > 0, the solution under the test is optimal.
\ The optimal allocation schedule is given by x 11 = 1, x 13 = 1, x 15 = 2, x 21 = 7, x 32 = 3,
x 33 = 6, x 43 = Î, x 44 = 2 and the optimum (minimum) transportation cost
= Rs. 4 ´ 1 + 3 ´ 1 + 2 ´ 2 + 1 ´ 7 + 2 ´ 3 + 4 ´ 6 + 2 ´ Î + 4 ´ 2
= Rs. ( 56 + 2 Î)
= Rs. 56/-, as Î ® 0.
Unbalanced Transportation Problems
If the given transportation problem is unbalanced one, i.e., if S a i ¹ S b j , then convert
this into a balanced one by introducing a dummy source or dummy destination with zero cost
vectors (zero unit transportation costs) as the case may be and then solve by usual method.
When the total supply is greater than the total demand, a dummy destination is included
in the matrix with zero cost vectors. The excess supply is entered as a rim requirement for the
dummy destination.
 122

When the total demand is greater than the total supply, a dummy source is included in
the matrix with zero cost vectors. The excess demand is entered as a rim requirement for the
dummy source.
* Example 13 : Solve the transportation problem
Destination
A B C D Supply
1 11 20 7 8 50
Source 2 21 16 20 12 40
3 8 12 18 9 70
Demand : 30 25 35 40

Solution : Since the total supply ( S a i = 160) is greater than the total demand
( S b j = 130), the given problem is an unbalanced transportation problem. To convert this into a
balanced one, we introduce a dummy destination E with zero unit transportation costs and
having demand equal to 160 – 130 = 30 units.
\ The given problem becomes
Destination
A B C D E Supply
1 11 20 7 8 0 50
Source 2 21 16 20 12 0 40
3 8 12 18 9 0 70
30 25 35 40 30 160

By using VAM the initial solution is as shown in the following table

11 20 7 8 0
35 15

21 16 20 12 0
10 30

8 12 18 9 0
30 25 15

\ The initial transportation cost

= Rs. 7 ´ 35 + 8 ´ 15 + 12 ´ 10 + 0 ´ 30 + 8 ´ 30 + 12 ´ 25 + 9 ´ 15
= Rs. 1160/-
 123

For optimality : Since the number non-negative allocations at independent positions is

( m + n - 1), we apply the MODI method.
11 7 20 11 7 8 0 -4
35 15 u1 = 8
4 9 4

21 11 16 15 20 11 12 0
10 30 u2 = 12
10 1 9
8 12 18 8 9 0 -3
30 25 15 u3 = 9
10 3
v1 = - 1 v 2 = 3 v3 = - 1 v4 = 0 v 5 = - 12

Since all dij > 0, the solution under the test is optimal and unique.
\ The optimum allocation schedule is
x 13 = 35, x 14 = 15, x 24 = 10, x 25 = 30, x 31 = 30, x 32 = 25, x 34 = 15
It can be noted that x 25 = 30 means that 30 units are despatched from source 2 to the
dummy destination E. In other words, 30 units are left undespatched from source 2.
The optimum (minimum) transportation cost
= Rs. 7 ´ 35 + 8 ´ 15 + 12 ´ 10 + 0 ´ 30 + 8 ´ 30 + 12 ´ 25 + 9 ´ 15
= Rs. 1160/-
* Example 14 : Solve the transportation problem with unit transportation costs,
demands and supplies as given below :

Destination
D1 D2 D3 D4 Supply
S1 6 1 9 3 70
Source S 2 11 5 2 8 55
S3 10 12 4 7 70
Demand 85 35 50 45

Solution : Since the total demand ( S bi = 215) is greater than the total supply
( S a j = 195), the given problem is unbalanced transportation problem. To convert this into a
balanced one, we introduce a dummy source S 4 with zero unit transportation costs and having
supply equal to 215 - 195 = 20 unit. \ The given problem becomes
 124

Destination
D1 D2 D3 D4 Supply
S1 6 1 9 3 70
Source S 2 11 5 2 8 55
S3 10 12 4 7 70
S4 0 0 0 0 20
85 35 50 45 215
As this problem is balanced, there exists a basic feasible solution to this problem. By
using Vogel’s approximation method, the initial solution is as shown in the following table.
6 1 9 3
5
65

11 5 2 8
25
30

10 12 4 7
45
25

0 0 0 0
20

\ The initial transportation cost

= Rs. 6 ´ 65 + 1 ´ 5 + 5 ´ 30 + 2 ´ 25 + 4 ´ 25 + 7 ´ 45 + 0 ´ 20
= Rs. 1010/-
For optimality : Since number of non-negative allocations at independent positions is
( m + n - 1), we apply the MODI method :
6 1 9 -2 3 1
65 5 u1 = 6
11 2
11 10 5 2 8 5
30 25 u2 = 10
1 3
10 12 12 7 4 7
25 45 u3 = 12
-2 5
0 0 -5 0 -8 0 -5
20 u4 = 0
5 8 5
v1 = 0 v2 = - 5 v3 = - 8 v4 = - 5

Since d31 = - 2 < 0, the solution under the test is not optimal.
 125

Now let us form a new basic feasible solution by giving maximum allocation to the cell (3,
1) (since d31 is –ve) by making an occupied cell empty. For this, we draw a closed path
consisting of horizontal and vertical lines beginning and ending at this cell (3, 1) and having its
other corners at some occupied cells. Along this closed loop, indicate + q and - q alternatively at
the corners.
We have

6 1 9 3
65 5+q
-q

11 5 2 8
30 25
-q +q

10 12 4 7
+q 25 45
-q

0 0 0 0
20

From the three cells (1, 1), (2, 2), (3, 3) having - q, we find that the minimum of the
allocations 65, 30, 25 is 25. Add this 25 to the cells with + q and subtract this 25 to the cells with
- q. Finally, the new basic feasible solution is displayed in the following table.

6 1 9 3
40 30
11 5 2 8
5 50
10 12 4 7
25 45
0 0 0 0
20

We see that the above table satisfies the rim conditions with ( m + n - 1) non-negative
allocations at independent positions. Now we check for optimality
 126

6 1 9 -2 3 3
40 30 11 0 u1 = 6
11 10 5 2 8 7
1 5 50 1 u2 = 10
10 12 5 4 2 7
25 7 2 45 u3 = 10

0 0 -5 0 -8 0 -3
20 5 8 3 u4 = 0
v1 = 0 v2 = - 5 v3 = - 8 v4 = - 3

Since all dij > 0 with d14 = 0, the solution under the test is optimal and an alternative
optimal solution exists.
\ The optimum allocation schedule is given by
x 11 = 40, x 12 = 30, x 22 = 5, x 23 = 50, x 31 = 25, x 34 = 45, x 41 = 20.
If can be noted that x 41 = 20 means that 20 units are despatched from the dummy source
S 4 to the destination D1 . In other words, 20 units are not fulfilled for the destination D1 .
The optimum (minimum) transportation cost
= Rs. 6 ´ 40 + 1 ´ 30 + 5 ´ 5 + 2 ´ 50 + 10 ´ 25 + 7 ´ 45 + 0 ´ 20
= Rs. 960/-
* Example 15 : Solve the transportation problem with unit transportation costs
in rupees, demands and supplies as given below :

Destination
D1 D2 D3 Supply
(unit)
A 5 6 9 100
Origin B 3 5 10 75
C 6 7 6 50
D 6 4 10 75
Demand 70 80 120
(units)

Solution : Since the total supply ( S a i = 300) is greater than the total demand
( S b j = 270), the given transportation problem is unbalanced.
To convert this into a balanced one, we introduce a dummy source D 4 with zero unit
transportation costs and having equal to 300 - 270 = 30 units. \ This given problem becomes
 127

Destination
D1 D2 D3 D4 Supply
A 5 6 9 0 100
Origin B 3 5 10 0 75
C 6 7 6 0 50
D 6 4 10 0 75
Demand 70 80 120 30 300

By using VAM the initial solution is given by

5 6 9 0
100

3 5 10 0
70 5

6 7 6 0
20 30

6 4 10 0
75

Since the number of non-negative allocations is 6, which is less than ( m + n - 1)

= 4 + 4 - 1 = 7, this basic feasible solution is degenerate.
To resolve this degeneracy, we allocate a very small quantity Î to the cell (2, 4), so that
the number of occupied cells becomes ( m + n - 1). Hence the non-degenerate basic feasible
solution is given in the following table
5 6 9 0
100
3 5 10 0
70 5 Î
6 7 6 0
20 30
6 4 10 0
75

Now the number of non-negative allocations at independent positions is ( m + n - 1). We

apply MODI method.
 128

5 6 6 8 9 0 3 u1 = 3
100
-1 -2 -3
3 5 10 6 0 u2 = 0
70 5 4 Î
6 3 7 5 6 0 u3 = 0
3 2 20 30
6 2 4 10 5 0 -1 u = -1
4
4 75 5 1
v1 = 3 v2 = 5 v3 = 6 v4 = 0

Since there are some dij < 0, the current solution is not optimal.
Since d14 = - 3 is the most negative, let us form a new basic feasible solution by giving
maximum allocation to the corresponding cell (1, 4) by making an occupied cell empty. We draw
a closed loop consisting of horizontal and vertical lines beginning and ending at this cell (1, 4)
and having its other corners at some occupied cells. Along this closed loop indicate + q and - q
alternatively at the corners.
We have
5 6 9 -q 0
100 +q

3 5 10 0
70 5 Î

6 7 6 0
20 30
+q -q
6 4 10 0
75

From the two cells (1, 3), (3, 4) having - q, we find that the minimum of the allocations
100, 30 is 30. Add this 30 to the cells with + q and subtract this 30 to the cells with - q. Hence the
new basic feasible solution is given in the following table.
5 6 9 0
70 30
3 5 10 0
70 5 Î
6 7 6 0
50
6 4 10 0
75
 129

We see that the above table satisfies the rim conditions with ( m + n - 1) non-negative
allocations at independent positions. So we apply MODI method.
5 3 6 5 9 0 u1 = 0
2 1 70 30
3 5 10 9 0 u2 = 0
70 5 1 Î
6 0 7 2 6 0 -3 u3 = - 3
6 5 50 3
6 2 4 10 8 0 -1 u4 = - 1
4 75 2 1
v3 = 3 v2 = 5 v3 = 9 v4 = 0

Since all dij > 0, the current solution is optimal and unique.
The optimum allocation schedule is given by
x 13 = 70, x 14 = 30, x 21 = 70, x 22 = - 5, x 24 = Î, x 33 = 50, x 42 = 75 and the optimum
(minimum) transportation cost
= Rs. 9 ´ 70 + 0 ´ 30 + 3 ´ 70 + 5 ´ 5 + 0 ´ Î + 6 ´ 50 + 4 ´ 75
= Rs. 1465/-
Maximization Case in Transportation Problems
So far we have discussed the transportation problems in which the objective has been to
minimize the total transportation cost and algorithms have been designed accordingly.
If we have a transportation problem where the objective is to maximize the total profit,
first we have to convert the maximization problem into a minimization problem by multiplying
all the entries by - 1 (or) by subtracting all the entries from the highest entry in the given
transportation table. The modified minimization problem can be solved in the usual manner.
* Example 16 : Solve the following transportation problem to maximize profit
Profits (Rs.)/Unit
Destination
A B C D Supply
1 40 25 22 33 100
Source 2 44 35 30 30 30
3 38 38 28 30 70
Demand 40 20 60 30

Solution : Since the given problem is of maximization type, first convert this into a
minimization problem by subtracting the cost elements (entries or C ij ) from the highest cost
element ( c ij = 44) in the given transportation problem. Then the given problem becomes.
 130

Destination
A B C D Supply
1 4 19 22 11 100
Source 2 0 9 14 14 30
3 6 6 16 14 70
Demand 40 20 60 30
This modified minimization problem is unbalanced ( Sa i = 200, Sb j = 150 and Sa i ¹ Sb j ).
To make it balanced, we introduce a dummy destination E with demand ( 200 - 150) = 50 units
with zero costs c ij . Hence the balanced minimization transportation problem becomes
Destination
A B C D E Supply
1 4 19 22 11 0 100
Source 2 0 9 14 14 0 30
3 6 6 16 14 0 70
Demand 40 20 60 30 50 200
Since Sa i = Sb j = 200, there exists a basic feasible solution to this problem and is
displayed in the following table by using VAM. [Try last cost method]
4 19 22 11 0
10 60 30

0 9 14 14 0
30

6 6 16 14 0
20 50

Since the number of non-negative allocations at independent position is 6, which is less

than ( m + n - 1) = (3 + 5 - 1) = 7, this initial solution is degenerate.
To resolve degeneracy, we allocate a very small quantity Î to the cell (3, 3), so that the
number of occupied cells becomes ( m + n - 1). Hence the initial solution is given by
4 19 22 11 0
10 60 30
0 9 14 14 0
30
6 6 16 14 0
20 Î 50
 131

Now the number of non-negative allocations at independent positions is ( m + n - 1). We

apply MODI method for optimal solution.
4 19 12 22 11 0 6
10 60 30
u1 = 0
7 -6
0 9 8 14 18 14 7 0 2
30 u2 = - 4
1 -4 7 -2
6 -2 6 16 14 5 0
20 Î 50 u3 = - 6
8 9
v1 = 4 v 2 = 12 v 3 = 22 v 4 = 11 v5 = 6

Since d15 , d23 , d25 are less than zero, the current solution under the test is not optimal.
Here d15 = - 6 is the most negative value of dij .
Let us form a new basic feasible solution by giving maximum allocation to the cell (1, 5)
by making an occupied cell empty. For this, we draw a closed path consisting of horizontal and
vertical lines beginning and ending at this cell (1, 5) and having its other corners at some
occupied cells. Along this closed loop, indicate + q and - q alternatively at the corners.
4 19 22 11 0
10 60 - q 30 +q

0 9 14 14 0
30

6 6 16 14 0 -q
20 +q Î 50

From the two cells (1, 3), (3, 5) having - q, we find that the minimum of 60, 50 is 50. Add
this 50 to the cells with + q and subtract this 50 to the cells with - q. Hence the new basic
feasible solution is displayed in the following table.
4 19 22 11 0
10 10 30 50
0 9 14 14 0
30
6 6 16 14 0
20 50
 132

We see that the above table satisfies the rim conditions with ( m + n - 1) non-negative
allocations at independent positions.
Now we apply the MODI method for optimality.
4 19 12 22 11 0
10 10 30 50 u1 = 0
7
0 9 8 14 18 14 7 0 -4
30 u2 = - 4
1 -4 7 4
6 -2 6 16 14 5 0 -6
20 50 u3 = - 6
8 9 6
v1 = 4 v 2 = 12 v 3 = 22 v 4 = 11 v5 = 0

Since d23 = - 4 < 0, the current solution is not optimal.

Let us form a new basic feasible solution by giving maximum allocation to the cell (2, 3)
by making an occupied cell empty. For this, we draw a closed loop consisting of horizontal and
vertical lines beginning and ending at this cell (2, 3) and having its other corner at some
occupied cells. Along this closed loop, indicate + q and - q alternatively at the corners.
4 19 22 11 0
20 10 30 50
+q -q
0 9 14 14 0
30
+q
-q

6 6 16 14 0
50
20

From the two cells (1, 3), (2, 1) having - q, we find that the minimum of the allocations 10,
30 is 10. Add this 10 to the cells with + q and subtract this 10 to the cells with - q. Hence the new
basic feasible solution is displayed in the following table.
4 19 22 11 0
20 30 50
0 9 14 14 0
20 10
6 6 16 14 0
20 50
 133

Now the number non-negative allocations at independent positions is ( m + n - 1). We

apply MODI method for the optimality
4 19 8 22 18 11 0
20 30 50 u1 = 0
11 4
0 9 4 14 14 7 0 -4
20 10 u2 = - 4
5 7 4
6 2 6 16 14 9 0 -2
20 50 u3 = - 2
4 5 2

v1 = 4 v2 = 8 v 3 = 18 v 4 = 11 v 5 = 0
Since all dij > 0, the current solution is optimal and unique.
\ The optimum allocation schedule is given by
x 11 = 20, x 14 = 30, x 15 = 50, x 21 = 20, x 23 = 10, x 32 = 20, x 33 = 50.
The optimum profit
= Rs. 40 ´ 20 + 33 ´ 30 + 0 ´ 50 + 44 ´ 20 + 30 ´ 10 + 38 ´ 20 + 28 ´ 50
= Rs. 5130/-
* Example 17 : Solve the following transportation problem to maximize profit.

Destination
A B C D Supply
1 15 51 42 33 23
Source 2 80 42 26 81 44
3 90 40 66 60 33
Demand 23 31 16 30 100

Solution : Since the given problem is of maximization type, we convert this in to

minimization problem by multiplying the profit costs c ij by - 1.
Destination
A B C D Supply
1 - 15 - 51 - 42 - 33 23
Source 2 - 80 - 42 - 26 - 81 44
3 - 90 - 40 - 66 - 60 33
Demand 23 31 16 30 100
 134

Since S a i = S b j = 100, there exists a basic feasible solution to this problem and is
displayed in the following table by using VAM.
- 15 - 51 - 42 - 33
23
- 80 - 42 - 26 - 81
6 8 30
- 90 - 40 - 66 - 60
17 16

Since the number of non-negative allocations at independent positions is ( m + n - 1) = 6,

we apply MODI method for optimal solution.
- 15 - 89 - 51 - 42 - 65 - 33 - 90 u1 = - 9
23
74 23 57
- 80 - 42 - 26 - 56 - 81 u2 = 0
6 8 30
30
- 90 - 40 - 52 - 66 - 60 - 91 u3 = - 10
17 16
12 31
v1 = - 80 v 2 = - 42 v 3 = - 56 v 4 = - 81

Since all dij > 0, the current solution is optimal and unique.
\ The optimum allocations are given by x 12 = 23, x 21 = 6, x 22 = 8, x 24 = 30, x 31 = 17,
x 33 = 16
\ The optimum profit
= Rs. 51 ´ 23 + 80 ´ 6 + 42 ´ 8 + 81 ´ 30 + 90 ´ 17 + 66 ´ 16
= Rs. 7005/-
MOST IMPORTANT QUESTIONS
1. What is a general transportation problem ? Explain with an example step by step
solution of a transportation problem.
2. Explain the terms ‘Initial Feasible Solution’, ‘Basic solution’, ‘Feasible Solution’ in the
context of transportation problem.
3. In case the objective function of a transportation problem is expressed in terms of profits
rather than costs, how the computational procedure is different from the standard
procedure.
4. Explain the technique used for solving a transportation problem and testing its
optimality.
5. What is meant by degeneracy in a transportation problem ? How is degeneracy resolved
in such problem ?
 135

6. How is optimality analysis done in a transportation problem ? Illustrate your answer

with an example.
7. What is Initial Feasible Solution of a transportation problem ? Explain with the help of
an example the North west corner method and Least cost entry method for obtaining
initial feasible solution.
8. Is it possible to obtain more than one optimal allocations in a transportation problem ?
What is its significance ?
9. Explain the Vogel’s approximation method of solving a transportation problem. How is it
an improvement over the NWCM ?
10. What is an unbalanced transportation problem ? How is such a problem handled and
solution obtained?
11. Explain the modified distribution method (MODI) of step by step improvement of initial
feasible solution of a transportation problem.
12. Explain the meaning of the following terms in transportation problem :
(a) Unbalanced Transportation problem. (b) Alternate solution.
(c) Optimal solution. (d) Cell evaluation.
13. Show that a transportation problem can be regarded as a particular case of linear
programming problem.
14. What are transportation problems ? How can a transhipment problem be solved as a
transportation problem ? Explain.
15. What is Degeneracy ? How it can be resolved in case of transportation problem ?
16. What is the difference between degeneracy in simplex problem and degeneracy in
transportation problem ?
17. Can degeneracy arise after you have started the optimality test if so how it can be
resolved ?
18. Given the choice which method you will prefer to test optimality.
(i) Stepping Stone Method (ii) Modified Distribution Method.
EXERCISE
1. Find the Initial Basic Feasible Solution to the following transportation problem by :
(a) Least cost method, (b) North-west corner rule.
State which of the methods is better
Table
To Supply
2 7 4 5
From 3 3 1 8
5 4 7 7
1 6 2 14
Demand 7 9 18

(Ans. Min. cost 83/- LCEM Min. Cost 102/- NWCM. ; LCEM is better)
 136

2. The appex company is the distribution for television receivers. It owns three warehouses
with stocking capacity as follows :
Warehouse Location A B C
Sets in Stock 100 25 75

It has the following order for set deliveries :

Market Location X Y Z
Orders 80 30 90

Delivery costs for warehouses to each customer are largely a function of mileage or
distance. The per unit costs have been determined to be :

X Y Z
A 5 10 2
B 3 7 5
C 5 8 4

The deliveries could be made in many ways but the distributor would like to deliver the
T.V. sets in a way that would minimise the delivery cost. Give the distribution schedule.
Use VAM only. (Ans. AX = 5 ´ 10 = 50; CX = 5 ´ 45 = 225; AZ = 2 ´ 90 = 180;
CY = 8 ´ 30 = 240; BX 3 ´ 25 = 75; Total Cost Rs. 770)
3. A transport company engaged in carrying parcels has three branches to serve five
customers. The distance (kms) from each branch to each of the customers is given below :

Branches Customers No. of trucks available

A B C D E
I 10 8 12 9 3 15
II 4 4 6 6 7 12
III 15 7 11 13 8 16
No. of Trucks Required 8 8 4 7 6

(i) Find out using VAM allocation of trucks from Branches to customers in order to
minimise the total cost of transportation assuming that the cost is proportional to
distance. What is the minimum required kms. to be run by the trucks.
(ii) If an arrival of a VVIP blocks the traffic from Branch I to customer C and D and
from branch II to customer D and E, then what should be the optimal allocation in
order to minimise total transportation cost ? (Use VAM)
(Ans. (i) Min. Distance = 195 kms. (ii) Minimum Transportaion Cost = 221)
 137

4. A company has four factories situated in different locations and five warehouses in
different cities. The matrix of transportation cost is given below :
Factories

Warehouse I II III IV Requirement

A 4 8 7 6 150
B 9 5 8 8 50
C 6 5 8 7 40
D 5 8 6 3 60
E 7 6 5 8 200
Capacity (units) 100 80 120 100

Find the Optimum Transportation Schedule using VAM only.

(Ans. Total Cost = Rs. 1870)
5. Obtain the Initial BFS to the following Transportation Problem by VAM.

D1 D2 D3 D4 D5 ai
A1 5 7 10 5 3 5
A2 8 6 9 12 14 10
A3 10 9 8 10 15 10
bj 3 3 10 5 4 25

(Ans. Total Cost = Rs. 183)

6. A steel company has three furnaces and five rolling mills. Transportation cost (rupees
per quintal) for sending steel from furnaces to rolling mills are given in the following
table :

Furnaces M1 M2 M3 M4 M5 Availability (Q)

A 4 2 3 2 6 8
B 5 4 5 2 1 12
C 6 5 4 7 3 14
Requirement (Quintal) 4 4 10 8 8

How should they meet the requirement ? Use VAM. (Ans. Total Cost Rs. 96)
7. A cement factory manager is considering the best way to transport cement from his three
manufacturing centres P, Q and R to depots A, B, C, D and E. The weekly production and
demand along with transportation costs per ton are given below :
 138

A B C D E Tons
P 4 1 3 4 4 60
Q 2 3 2 2 3 35
R 3 5 2 4 4 40
Tons 22 45 20 18 30 135

What should be the distribution programme? (Ans. Total Cost = Rs. 290)
8. Solve the following problem and test its optimality.

Project A Project B Project C Plant

Plant W 4 8 8 56
Plant X 16 24 16 82
Plant Y 8 16 24 77
Project Requirement 72 92 41 215
205

(Ans. Total Cost 2,504)

9. The relevant data on demand, supply and profit per unit of a product manufactured and
sold by a company are given below :
Marketing Centre
Factory 1 2 3 4 5 Supply
A 5 8 14 7 8 100
B 2 6 7 8 7 20
C 3 4 5 9 8 60
D 4 10 7 8 6 20
Demand 45 65 70 35 15

Given that transportation from A to 3, and D to 2 are not allowed due to certain reasons.
Find out using VAM, Optimal method of Transportation from factories to marketing
centres. (Ans. Max. Profit Rs. 1,450/- by VAM 1,460 optimality)
10. A company has decided to manufacture some or all of five new products at three of its
plants. The production capacity of each of these three plants is as follows :

Plant No. Production capacity in total number of units

1 40
2 60
3 90
 139

Sales potential of five products is as follows :

Product no. 1 2 3 4 5
Market potential in units 30 40 70 40 60

Plant no. 3 cannot produce product no. 5. The variable cost per unit for the respective
plant and product combination is given on the next page.

Product No. 1 2 3 4 5
Plant No. 1 20 19 14 21 16
Plant No. 2 15 20 13 9 16
Plant No. 3 18 15 18 20 —

Based on above data determine the optimum product to plant combination. Test
optimality by Modi method to minimize cost. (Ans. Total Cost = Rs. 2,700)