Assignment Model: Prototype Example

The JOB SHOP COMPANY has purchased three new machines of different types. There are four
available locations in the shop where a machine could be installed. Some of these locations are more
desirable than others for machines because of their proximity to work centers that will have a heavy
workflow to and from these machines. (There will be no workflow between the new machines.)

Therefore, the objective is to assign the new machines to the available locations to minimize the
total cost of materials handling. The estimated cost in dollars per hour of materials handling
involving each of the machines is given in Table 9.24 for the respective locations. Location 2 is not
considered suitable for machine 2, so no cost is given for this case.

• To formulate this problem as an assignment problem, we must introduce a dummy machine

for the extra location.

• Also, an extremely large cost M should be attached to the assignment of machine 2 to

location 2 to prevent this assignment in the optimal solution.

• The resulting assignment problem cost table is shown in Table 9.25.

Step I: # Rows = # Column =>We have balanced the model by adding a dummy row

Step II: It should be a minimization

Step III

Row Reduction: Subtract the smallest element from element of the row

2 5 1 0

2 M 0 7

0 2 5 1

0 0 0 0

Column Reduction – Not required

Step IV: Make assignments

2 5 1 0

2 M 0 7

0 2 5 1

0 0 0 0

# Assignments = 4 = # Rows (Columns)

 WE have reached an optimum solution

 Stop here

Optimum assignment plan

Machine Location Cost ($)

1 4 11

2 3 13

3 1 5

Dummy 2 0

-------------------------------------------------

Total 29 ($)

------------------------------------------------

Location 2 remains unallocated

Assignment model is a special case of transportation model if

1) If each supply = 1
2) If each demand = 1
3) If no. of sources = no. of destinations

In this condition the transportation model becomes an assignment model

Example

Job
Worker Supply
A B C D

1 40 51 67 1
45

2 57 42 63 55 1

3 49 52 48 64 1

4 41 45 60 55 1

Demand 1 1 1 1 4

Here each supply = 1 = each demand

Also no. of workers = 4 = no. of jobs = 4

Hence this is one of the assignment model

Solution to an Assignment Problem

• Can be solved by complete enumeration – ruled out since the number of alternatives is very
large (equal to n!)

• Can be formulated and solved as a transportation problem – not efficient because the
solution would be severely degenerate

• Can be formulated and solved an LPP (a 0-1 IPP) – also not an efficient way of solving the
problem
 • Can be solved by Hungarian Assignment Method (HAM)

• HAM is most efficient of all methods

1) Enumeration Method
a. A list of all possible assignments among the given resources (men, machine etc.) and
activities (jobs, sales areas, etc.) is prepared.
b. Assignment involving the minimum cost ( or maximum profit), time or distance is
selected
c. If an assignment problem involves n workers / jobs, then in total there are n!
possible assignments.
d. However, when n is large, the method is unsuitable for manual calculations.
2) Transportation method
a. Can be solved by MODI, however, every basic feasible solution of general
assignment problem that has a square matrix of order n should have m + n – 1
assignments
b. Solution is degenerate and requires n-1 number of dummy allocations (deltas of
epsilons) to remove it which makes the procedure computationally inefficient for
solving an assignment problem

Actual no of allocations = 4
4+4-1 = 7
This method creates a severe degeneracy and becomes very difficult and complex

3) Simplex – The general mathematical model of the assignment problem involves nxn decision
variables and n + n or 2n equalities. Solving such an assignment problem manually is difficult

4) Hungarian Method

a. Developed by Hungarian mathematician D. Konig

b. It is an efficient method of finding the optimal solution

c. Works on the principle of reducing the given cost matrix to a matrix of opportunity
costs

d. Opportunity costs show relative penalties associated with assigning a resource to an

activity

e. Hungarian method reduces the cost matrix to the extent of having at least one zero
in each row and column so as to make optimal assignments

Hungarian Assignment Method

Step 1

• Balance the problem if it is unbalanced

• Place an M as the cost element if some assignment is prohibited

• Convert into equivalent min problem if it is a max problem

 • For this conversion, subtract each value of the matrix from a constant value, K (
generally the largest value in the given matrix)

Step 2

• Obtain zeros in every row and column by row reductions and column reductions

Step 3

• Cover all zeros by least number of horizontal/ vertical lines

• If number of lines = n move to step 4 else, make adjustment: select least uncovered
value; subtract it from every uncovered value; and add to each one at intersection of
lines

• Again cover zeros with lines to find if you could go to 4

Step 4

• Make assignments by considering zeros

HAM: Making Assignments

• Select a row/column with a single zero

• Make an assignment at zero by placing a square around it

• Cross out zero/s, if any, in the corresponding column/row

• Continue in this manner until all assignments are made

• If there are multiple zeros left in all rows and columns, select a zero arbitrarily, make
assignment and cancel other zeros in the row and column involved

• Multiple zeros in all rows and columns remaining indicates multiple optimal solutions

Ex1 Solve the given problem using Hungarian Method

Programers A B C

Programes

1 120 100 80

2 80 90 110

3 110 140 120

1) Check whether the Assignment model is a balanced or not, if no. of rows = no. of column
then the model is a balanced AM. Since in this example, no, of rows = 3 = no. of columns
hence the model is a balanced one
2) Calculate the Row minima matrix: Subtract the smallest element from all the elements in
each row
 Row minima matrix

Programers A B C

Programes

1 120-80 = 40 100-80 = 20 80-80=0

2 80-80 = 0 90-80 = 10 110-80 = 30

3 110-110 = 0 140-110 = 30 120-110 = 10

Programers A B C

Programes

1 40 20 0

2 0 10 30

3 0 30 10

Check now if you have at least one zero in each column, if yes then no need to reduce it
column wise otherwise reduce it column wise as well
Column Minima Matrix

Programers A B C

Programes

1 40 20-10=10 0

2 0 10-10=0 30

3 0 30-10=20 10

Programers A B C

Programes

1 40 10 0

2 0 0 30

3 0 20 10
 3) Make assignments
4) If the no. of assignments = no. of rows or columns, then we have reached an optimum
solution. (no. of allocations = 3 = no. of rows hence the solution is optimum)
Stop iterating and write down the optimum assignment plan and the minimum time
Optimum assignment plan:
Programmers Programme Time
A -> 3 110
B -> 2 90
C -> 1 80
Total 280 minutes
The solution is unique optimum as there are unique zeroes which are assigned for each row
and column.

Example 2 (Heera Gupta Page 329 Ex 4.6-2)

• Four different jobs can be done on four different machines. The set-up and take-down time
costs are assumed to be prohibitively high for changeovers. The matrix below gives the cost
in rupees of producing job i on machine j.

• How should the jobs be assigned to the various machines so that the total cost is
minimized?

Jo Machines No of rows = 4 = no. of columns => the

bs given assignment model is a balanced
model
M M M M
1 2 3 4 2) Reduce it row wise (Row minima) –
Row Reduction
J1 5 7 1 6
1 Jo Machines
bs
J2 8 5 9 6 M M M M
1 2 3 4
J3 4 7 1 7
0 J1 0 2 6 1

J4 1 4 8 3 J2 3 0 4 1
0

Solution:1) Check whether the assignment J3 0 3 6 3

model is balanced one or not=> if no. of
rows = no. of columns J4 7 1 5 0

3) Reduce it column wise (Column minima) – Column Reduction

 Jo Machines 4) Make assignments
bs
Jobs Machines
M M M M
1 2 3 4
M1 M2 M3 M4

J1 0 2 2 1
J1 0 2 2 1

J2 3 0 0 1
J2 3 0 0 1

J3 0 3 2 3
J3 0 3 2 3

J4 7 1 1 0
J4 7 1 1 0

# Assignments = 3 ≠ # Rows (4)

 Non-optimum
 Move to step 5 for the improvement in the solution.

Step 5: Improvement in the solution

Jobs Machines

M1 M2 M3 M4

J1 0 2 2 1

J2 3 0 0 1

J3 0 3 2 3

J4 7 1 1 0

Since no. of assignment = 3 ≠ 4(no of rows) hence the solution is non-optimum, we need to improve
the solution further and go to step 5.

(5) Steps of improvement in the solution

(a) the row with unassigned zero

(b) in the ticked row find crossed zeros and tick the corresponding column

(c) in the ticked column find the assigned zero and tick the corresponding row

(d) draw minimum no. of lines to cover maximum zeros

 (i) Draw the lines for unticked rows

(ii) Draw the lines for ticked columns

(iii) no. of lines drawn = 3 ≠ 4 which confirms that our solution is non-optimum

(e) identify the smallest uncovered element (1)

(f) subtract this smallest uncovered element (1) from all uncovered elements (as highlighted
in yellow)

(g) add it (1) to the elements at the intersection of the lines (as highlighted in green)

(h) write all other elements as it is

Revised solution

Jobs Machines

M1 M2 M3 M4

J1 0 1 1 0

J2 4 0 0 1

J3 0 2 1 2

J4 8 1 1 0

(6) Make assignments

No of assignments = 3 ≠ 4 (no of rows) hence this is non-optimum solution, move to step (5) again to
improve the solution.

Revised solution J4 8 0 0 0
o Machines
b
s Since no. of assignments = 4 = no of rows =>
M M M M
the given solution is optimum. Stop iterating
1 2 3 4
Optimum solution (Since we have multiple
J1 0 0 0 0 zeros, it has multiple optimum solution)

Jobs Machines Cost (Rs.)

J2 5 0 0 2
J1 M1 5
J3 0 1 0 2 J2 M2 5

J3 M3 10
J4 M4 3 Total Rs. 23

Alternate solution

Jo Machines
bs Jobs Machines Cost (Rs.)
M M M M J1 M2 7
1 2 3 4
J2 M3 9
J1 0 0 0 0 J3 M1 4

J4 M1 3
J2 5 0 0 2
Total Rs. 23
J3 0 1 0 2

J4 8 0 0 0

The given problem has multiple optimum solutions where the total minimum cost remains same i.e.
Rs. 23 but the assignment of jobs to machines is altered as shown in the above table.