Assignment Problem

The assignment problem is a particular case of transportation problem in which the number of
jobs/origins/sources is equal to the number of facilities/destinations/machines/persons. The objective is to
maximize total profit of allocation or to minimize the total assignment cost. It is completely the
degenerate form of a transportation problem.

Mathematically:
n n
Minimize  c
i 1 j 1
ij xij subject to constraints

n n

 xij  1, for i 1, 2,........and  xij  1, for j  1, 2, ......with xij  0 or 1

Hungarian Assignment Algorithm

1. If the no of rows are not equal to the no of columns and vice versa, then a dummy row/dummy
column must be added with zero cost elements.
2. Find the smallest cost in each row of the cost matrix. Subtract this smallest cost element from all
other elements in that row. Therefore, there will be at least one ZERO in every row of the matrix
which is called reduced matrix.
3. In the reduced matrix from STEP-2 find the smallest cost in each column of the cost matrix.
Subtract this smallest cost element from all other elements in that column. Therefore, there will
be at least one ZERO in every column.
4. Determine the optimum assignment as follows:
a. Check the rows successively until a row with exactly one ZEROs is found. Box that
ZERO and strike out all other ZEROs in the respective column. Proceed in this manner
till all the rows are examined. If a row has two or more ZEROs left that row and proceed
for the next row.
b. Repeat the above procedure for columns.
c. Repeat the above procedures till all ZEROs either BOXED or CROSSED.
d. If all the rows having single BOX ZERO then the current solution is the optimal solution.
Otherwise proceed to next step. If a ZERO cell arbitrarily chosen then there may exists
another optimal solution.
5. a. Mark () to those rows where no assignment (no BOX ZERO) has been made.
b. Mark () to those columns which have crossed ZEROs in the marked rows.
c. Mark () to those rows which is not marked before which have assignments (BOX ZEROs) in
marked columns.
d. The process may be repeated until no more rows/columns can be checked.
e. Draw horizontal rows in unmarked rows and vertical lines in marked columns.
6. If the minimum no of lines passing through all the ZEROs is equal to the no of rows/columns, the
optimum solution is attained by an arbitrarily allocation in the positions of the ZEROs not crossed
before. Otherwise go to the next step.
7. Revise the cost matrix as follows:

1
 a. Find the elements that are not covered by the lines. Choose minimum those elements.
Subtract that element from all uncovered elements and add that element to the
intersection elements of the lines.
b. Other elements crossed by the lines will remain unchanged.
8. Go to previous step (STEP-4) and continue the procedure till an optimal solution is obtained.
Illustrations:
Find the Optimal Assignment of
I II III IV
A 10 5 13 15
B 3 9 18 3
C 10 7 3 2
D 5 11 9 7

Answer: 1. Row-Wise Subtraction:

I II III IV
A 5 0 8 10
B 0 6 15 0
C 8 5 1 0
D 0 6 4 2

2. Column-Wise Subtraction:

I II III IV
A 5 0 7 10
B 0 6 14 0
C 8 5 0 0
D 0 6 3 2

3. BOX a single ZERO starting from first row and cross () all other zeros in its columns, we get

I II III IV
A 5 0 7 10
B 0 6 14 0
C 8 5 0 0
D 0 6 3 2

Since each row and column has exactly one assignment, the current solution is the optimal solution. That
is
AII, BIV, CIII and DI and the total assignment cost is 16.

2
Illustrations:
Find the Optimal Assignment of
I II III IV V
A 10 5 13 15 16
B 3 9 18 13 6
C 10 7 2 2 2
D 7 11 9 7 12
E 7 9 10 4 12

Answer:
Table-1 (Row-wise Subtraction)
I II III IV V
A 5 0 8 10 11
B 0 6 15 10 3
C 8 5 0 0 0
D 0 4 2 0 5
E 3 5 6 0 8
Table-2 (Column-wise Subtraction)
I II III IV V
A 5 0 8 10 11
B 0 6 15 10 3
C 8 5 0 0 0
D 0 4 2 0 5
E 3 5 6 0 8
Table-3 (Allocations)
# Check for Single ZEROs in Rows.BOX it (in this case color) and CROSS all other ZEROs in
corresponding columns
I II III IV V
A 5 0 8 10 11
B 0 6 15 10 3
C 8 5 0 0 0
D 0 4 2 0 5
E 3 5 6 0 8

As no of BOX ZEROs are not equal to no. of rows/no of columns, current assignment is not an optimal
assignment.
Table-4 (Drawing lines using () marks) (Step-5)

I II III IV V
A 5 0 8 10 11
B 0 6 15 10 3 
C 8 5 0 0 0
D 0 4 2 0 5 
E 3 5 6 0 8 
 

3
As the no of lines (4) is not equal to the no. of rows/no. of columns, arbitrary assignment cannot be found,
go to next step.
Table-5 (Choose the smallest element from all uncovered elements (in this case 2), subtract it all other
uncovered elements and add to the intersection elements)
I II III IV V
A 7 0 8 12 11
B 0 4 13 10 1
C 10 5 0 2 0
D 0 2 0 0 3
E 3 3 4 0 6
In this table a new assignment will be done.
I II III IV V
A 7 0 8 12 11
B 0 4 13 10 1
C 10 5 C 2 0
D C 2 0 C 3
E 3 3 4 0 6
So the assignment is
AII @ 5,
B  I @3,
C V @ 2,
D III @ 9,
E IV @ 4
The total assignment cost is 23 hours.
Variations:
1. Unbalanced Assignment Problem: If the no. of rows is not equal to the no. of columns (Not a
Square matrix) the assignment problem is called the unbalanced assignment problem. By adding
dummy row (s)/column(s) with ZERO cost value, unbalanced assignment problem can be
converted to balanced assignment problem.
2. Maximization Problem: If an assignment problem is given in maximization form, we can
convert them to minimization problem by following TWO ways:
a. By subtracting each element from the highest element of the matrix.
b. By multiplying the matrix elements by -1.
3. Multiple Optimal Solutions: While making an assignment in the reduced assignment matrix, it
is possible to have two/more ways of strike off certain no. of ZEROs. Such situation leads to
multiple solutions with same optimal cost.
4. Impossible Assignment: Cells in which assignments are not allowed are assigned a very heavy
cost (written as M). Such cells are prohibited to enter into the final solution.

4
Examples:
1. (Unbalanced Assignment)
A B C D E
M1 9 11 15 10 11
M2 12 9 - 10 9
M3 - 11 14 11 7
M4 14 8 12 7 8

2. (General Approach)
I II III IV V
A 2 9 2 7 1
B 6 8 7 6 1
C 4 6 5 3 1
D 4 2 7 3 1
E 5 3 9 5 1
3. (Impossible Assignments)

A B C D E
M1 7 7 - 4 8
M2 9 6 4 5 6
M3 11 5 7 - 5
M4 9 4 8 9 4
M5 8 7 9 11 3

4. (Multiple Solutions)

I II III IV
A 2 3 4 5
B 4 5 6 7
C 7 8 9 8
D 3 5 8 4