Page 1 of 16

Name :
Course and Year :
Subject : Management Science
Class Schedule :
Instructor :

Module 7: Unbalanced and Degenerate Transportation Problems;

Transportation – Maximization Problems and Linear
Programming: Assignment Method

Intended Learning Outcomes

 Describe the characteristics of unbalanced and degenerate transportation problems.
 Solve unbalance transportation problems
 Solve degenerate transportation problems
 Solve maximization transportation problem
 Discuss the process of assignment method
 Solve assignment method problems

Activity
As we start this lesson please review the steps in solving minimization process
that we have discussed in the previous chapters, using the Stepping stone and
MODI method.

Analysis
In the previous chapters, we have discussed how to solve minimization
transportation problems using the Stepping Stone and MODI Methods. However, it
must be noted that in the examples solved in the previous module, the total demand
and the total supply are always equal. This is not the always the situation in real life.
Sometimes there is deficiency in supply or lower supply. Thus, in this module such
situation will be dealt with. Also, this module will tackle transportation problem with
degeneracy problem. Another type of linear programming problem – assignment
problem, will also be discussed in this module.

Abstraction
\

1. The Unbalanced Transportation Problem:

 In actual practice, it seldom happens that the quantity demanded is just
equal to the quantity supplied.
 Under normal situation, transportation problems come as unbalanced
problems.
 Page 2 of 16

 Stepping stone and MODI methods can only handle balanced tables
 In order to solve unbalanced transportation problem, they must first be
converted into balanced problems.
 To convert unbalanced table to balanced one, the following shall be done:
- Use a dummy column / demand (if demand is deficient, supply is
greater) – add additional column to the right
- Use a dummy row / supply (if supply is deficient, demand is greater) –
add additional row below
- A dummy is something we pretend to exists, although in reality it does
not.
- The cost of transportation for dummy cells is zero.

2. Sample problems – UNBALANCED PROBLEMS

Sample 1: Supply is greater than demand (deficient demand)

Minimize:
A B C Suppl
y
1 4 8 8
76
2 16 24 16
82
3 8 16 24
77
Deman 235
72 102 41
d 215
The above table is unbalanced, since the total supply is 235 units, while the
total demand is 215.
To convert the table to a balanced one, create a dummy demand / dummy
column. So, a new column must be added to the right. The cost of
transportation for each dummy cell is zero.

Solution:

A new table with a dummy demand, and the initial distribution of units is shown
below:

A B C Dumm Suppl
y y
1 4 8 8 0
76
2 16 24 16 0
82
3 8 16 24 0
77
Deman 235
72 102 41 20
d 235
 Page 3 of 16

The above table shows a surplus supply of 20 units which is entered in the
dummy demand. (Now you can continue solving for the optimum table by the
MODI or stepping stone methods).

Further solution (using Stepping Stone Method):

PROBLEM 1 TABLE 1:

Computation of total cost (For Table 1)

Occupied Computatio Total
Cells n Cost
1-A 72 x 4 = 288
1-B 4x8= 32
2-B 82 x 24 = 1, 968
3-B 16 x 16 = 256
3-C 41 x 24 = 984
3 – Dummy 20 x 0 = 0
Total Cost 3, 528

Computation for Improvement

Vacant Cost Reduction for
Cells improvement
1-C -8+16 -24 + 8 = - 8
1 – Dummy -8 + 16 – 0 + 0 = 8
2-A -4 + 8 – 24 + 16 = 4
2-C -24 + 16 - 24 + 16 = -16
2-Dummy -24 + 16 – 0 + 0 = -8
3-A -16 + 8 – 4 + 8 = -4
Transfer 3-C (41 units) to 2-C
TABLE 2:
A B C Dummy Supply
1 4 8 8 0
76
72 4
2 16 24 16 0
82
41 41
8 16 24 0
 Page 4 of 16

3 57 20 77
235
Computation of72
total cost102
(For Table 41
2) 20
235
Occupied Computatio Total
Cells n Cost
1-A 72 x 4= 288
1-B 4x8= 32
2-B 41 x 24 = 984 Checking:
2-C 41 x 16 = 656 Previous Cost: 3, 528
3-B 57 x 16 = 912 Present Cost: 2, 872
3 – Dummy 20 x 0 = 0 Difference: 656
Divide: Number of units transferred:41
Total Cost 2, 872
Answer:16 (should be the absolute
Computation for Improvement value of cost of improvement)
Vacant Cost Reduction for
Cells improvement
1-C -8 + 24 – 16 + 8 = 8
1 – Dummy -0 + 16 – 8 + 0 = 8
2-A -24 + 8 – 4 + 16 = -4
2-Dummy -0 + 16 – 24 + 0 = -8
3-A -16 + 8 – 4 + 8 = - 4
3-C -16 + 24 – 16 + 24 = 16
Transfer 3-Dummy (20 units) to 2 - Dummy
TABLE 3:

A B C Dummy Supply
1 4 8 8 0
76
72 4
2 16 24 16 0
82
21 41 20
3 8 16 24 0
77
77
235
72 102 41 20
215
Computation of total cost (For Table 3)
Occupied Computatio Total
Cells n Cost
1-A 72 x 4= 288
1-B 4x8= 32 Checking:
2-B 21 x 24 = 504 Previous Cost: 2,872 (Table 2)
2-C 41 x 16 = 656 Present Cost: 2, 712 (Table 3)
2- Dummy 20 x 0 = 0 Difference: 160
3-B 77 x 16 = 1, 232 Divide: Number of units transferred:20
Total Cost 2, 712 Answer:8 (should be the absolute value
of cost of improvement)
Computation for Improvement
Vacant Cost Reduction for
Cells improvement
 Page 5 of 16

1-C -8+24-16+8 =8
1 – Dummy -0+24-8+0 = 16
2-A -24+8-4+16 = -4 TABLE 4:
3-A -16+8-4+8 =-4
3-C -16+24-16+24 = 16
3- Dummy -0+24 - 16 + 0 = 8
Transfer 2-B (21 units) to 2-A

A B C Dummy Supply
1 4 8 8 0
76
51 25
2 16 24 16 0
82
21 41 20
3 8 16 24 0
77
77
235
72 102 41 20
215

Computation of total cost (For Table 4)

Occupied Computation Total
Cells Cost
1-A 51 x 4 = 204 Checking:
1- B 25 x 8 = 200 Previous Cost: 2,712 (Table 3)
2-A 21 x 16 = 336 Present Cost: 2, 628 (Table 4)
2-C 41 x 16 = 656 Difference: 84
2-Dummy 20 x 0 Divide: Number of units transferred:21
0= Answer:4 (should be the absolute value
3-B 77 x 16 = 1,232 of cost of improvement)
Computation for Improvement
Total Cost 2,628
Vacant Cost Reduction for
Cells improvement
1-C -8+24-16+8 =8
1 – Dummy -0+24-8+0 = 16
2-B -16+4-8+24 = 4
3-A -4+8-16+8 = -4
3-C -16+24-16+24 = 16
3- Dummy -0+24 - 16 + 0 = 8
Transfer 1-A (51 units) to 1 – 3

TABLE 5:

A B C Dummy Supply
1 4 8 8 0
76
76
2 16 24 16 0
82
21 41 20
3 8 16 24 0
77
51 26
72 102 41 20
 Page 6 of 16

235
215

Computation of total cost (For Table 5)

Occupied Computation Total
Cells Cost Checking:
2- B 76 x 8 = 608 Previous Cost: 2, 628 (Table 4)
Present Cost: 2, 424 (Table 4)
2-A 21 x 16 = 336
Difference: 204
2-C 41 x 16 = 656
Divide: Number of units transferred:51
2-Dummy 20 x 0 0
Answer:4 (should be the absolute value
=
of cost of improvement)
3-A 51 x 8 408
=
3-B 26 x 16 = 416
Total Cost 2,424
Computation for Improvement
Vacant Cost Reduction for
Cells improvement
1-A -8+16-8+4 =4
1-C -8+24-16+8 =8
1 – Dummy -0+24-8+0 = 16
2-B -16+4-8+24 = 4
3-C -16+24-16+24 = 16
3- Dummy -0+24 - 16 + 0 = 8
OPTIMUM TABLE (All results are positive) Decision:
1 to A: 76
2 to A: 21
2 to C: 41
3 to A: 51
3 to B: 26
Minimum Cost: 2, 424
Sample 2: Demand is greater than supply (deficient supply)
Minimize:
A B C Suppl
y
1 4 8 8
56
2 16 24 16
82
3 8 16 24
77
Deman 80 102 41
 Page 7 of 16

d 215
223
To convert the table to a balanced table, create a dummy supply.
Solution:
A B C Suppl
y
1 4 8 8
56
56
2 16 24 16
82
24 58
3 8 16 24
77
44 33
Dummy 0 0 0
8
8
Deman 223
80 102 41
d 223
The above table shows a demand surplus of 8 units, entered in the dummy supply. Use either
stepping stone or MODI method.

3. Degenerate Transportation Problems

- Review: In stepping stone and MODI method, to ensure that you can work on
improvement computation, the number of occupied cells should be equal to
number of rows + number of columns minus one.
- Also, in making a route, it should land in occupied cells.
- If the total number of used cells is not equal to the number of rows plus the
number of columns minus one, degeneracy in transportation occurs.
- Degeneracy may occur in an initial solution or in subsequent solutions.
- A degenerate table is one which cannot be improved due to lack of entries.

4. Sample Problems - Degeneracy

Sample 3: Degeneracy on initial solution

- Degeneracy in the initial solution arises when in using the Northwest Method
of initial distribution, both column requirement and row requirement are satisfied
simultaneously, thus breaking the stair-step pattern. To resolve this, we assign
a Zero entry to one of the unused cells.

Minimize:

A B C Suppl
y
1 4 8 8
55
35 20
2 16 24 16
25
25
3 8 16 24
30
 Page 8 of 16

30
Deman 110
35 45 30
d 110

# rows + # columns – one (constant) = number of cells that should be occupied

3+3 – 1 = 5 (4 occupied cells – degeneracy problem)
Solution:
TABLE 1

A B C Suppl
y
1 4 8 8
55
35 20
2 16 24 16
25
25
3 8 16 24
30
0 30
Deman 110
35 45 30
d 110
An example of
transportation initial table with degenerate initial solution is shown above. By the
use of the Northwest distribution, the entries are shown in the table. While the
number of rows plus the number of columns minus one is supposed to be five,
the number of occupied cells in the table is only 4. Hence, the table is
degenerate. If we try to compute for the vacant cells 1-C, 2-C, 3-B and 3-A, no
route (rectangular or zigzag) is possible.

A zero entry may be assigned to cell 2-C or 3-B. Either case will give the same
optimum solution.
The new table with zero entry assigned to an unused cell is shown below.
TABLE 1:

A B C Suppl
y
1 4 8 8
55
35 20
2 16 24 16
25
25
3 8 16 24
30
0 30
 Page 9 of 16

Deman 110
35 45 30
d 110

We can now solve the problem, treating the zero entry just like an ordinary non-
zero entry.
COST (TABLE 1)
3
1-A 5 4 140
2
1-B 0 8 160
2
2-B 5 24 600
3-B 0 16 0
3
3-C 0 24 720
TOTAL COST 1,620.00
Computation for Improvement
Vacant Cost Reduction for
Cells improvement
1-C -8+16-24+8=-8
2-A -24+8-4+16=-4
2-C -24+16-24+16 = -16
3-A -4+8-16+8 = -4
Transfer 2-B (25 units) to 2-C

TABLE 2:
 Page 10 of 16

COST (TABLE 2)
1-A 35 4 140
1-B 20 8 160
2-C 25 16 400
3-A 25 16 400
3-C 5 24 120
TOTAL COST 1,220.00
IMPROVEMENT
1- - 1
C 24 6 -8 8 -8
2- - 2 - - 1 1
A 16 4 16 8 4 6 2
2- - 2 - 2 1
B 16 4 16 4 6
3- -
A 16 8 -4 8 -4
TRANSFER 3-C (5 UNITS) TO 1-C
TABLE 3

A B C Supply
4 8 8
1 35 15 5 55
16 24 16
2 25 25
8 16 24
3 30 30
Deman 110
35 45 30
d 110

COST (TABLE 3)
 Page 11 of 16

1-A 35 4 140
1-B 15 8 120
1-C 5 8 40
2-C 25 16 400
3-B 30 16 480
TOTAL COST 1,180.00

IMPROVEMENT
1
2-A -16 8 -4 6 4
2
2-B -8 8 -16 4 8
3-A -16 8 -4 8 -4
2
3-C -8 8 -16 4 8
Transfer 3-B (30 units) to 3-A

A B C Supply
4 8 8
1 5 45 5 55
16 24 16
2 25 25
8 16 24
3 30 30
Deman 110
35 45 30
d 110
COST
1-A 5 4 20
1-B 45 8 360
1-C 5 8 40
2-C 25 16 400
3-A 30 8 240
TOTAL COST 1,060.00

IMPROVEMENT
2-A -4 8 -16 16 4
2-B -16 8 -8 24 8
3-B -8 4 -8 16 4
3-C -8 4 -8 24 12
OPTIMUM TABLE
 Page 12 of 16

Sample 4: Degeneracy problem in subsequent solution

- Some transportation problems begin with an initial table that is not degenerate,
but in the process of finding the optimum solution, degeneracy arises. This
happens when a tie exists between two or more entries that represent the
smallest on the path of improvement with minus sign.

The following table has an initial solution which is not degenerate. Using the MODI
method, we will find out that we can improve the solution by transferring some units to
cell 3-B.

Minimize:

W1 = 4 W2 = 8 W3 = 12
A B C Supply
V1 = 0 1 4 8 8
95
50 45
V2 =-4 2 8 4 8
65
35 30
V3 =4 3 16 4 16
35
35
195
Demand 50 80 65
195
Vacant (Cost – W-V)
Cells
1-c 8 -12 -0 = -
4
2-a 8-4-(-4) = 8
3-a 16-4-4= 8
The second table 3-b 4-8-4 = -8 after the transfer of 35 to cell 3_B is
shown below:
Table II.
W1 = W2 = W3 =
A B C Supply
V1 = 1 4 8 8
95
50 45
V2 = 2 8 4 8
65
65
V3 = 3 16 4 16
35
35 - The
195 above
Demand 50 80 65
195 table
is
 Page 13 of 16

degenerate because only 4 cells are occupied, while the rows plus the number of
columns minus one is 5.
- To resolve this when there are two cells that will remain vacant, it is advisable to
indicate 0 in one of the two vacated cells. Assume that only one quantity was
removed or transferred, and the other one was subtracted, thus leaving 0 in one
cell. The affected cells are:

Thus Table II is shown below:

W1 = 4 W2 = 8 W3 = 4 Using
A B C Supply the
MODI
V1 =0 1 4 8 8
95
50 45
V2 = 4 2 8 4 8
65
0 65
V3 = -4 3 16 4 16
35
35
195
Demand 50 80 65
195
method, find the optimum table.

5. Summary on the Iteration of the Transportation Table

a. A transportation table is workable only if the total demand quantity equals the
total supply quantity, and the number of occupied cells equals number of rows
plus number of columns minus one.
b. If demand is not equal to supply, add a dummy supply row if demand is greater;
and add a dummy demand column if supply is greater.
c. In case of degeneracy in an initial solution, use a zero entry in a vacant cell such
that a “stair-step” chain remains.
d. In case of degeneracy in a subsequent table, retain a zero difference in one of
the vacated cells.

6. Transportation Problem: Maximization Process

- Maximization transportation problems have the same procedure of solving as
that of minimization, except that instead of finding the greatest negative value in
the computation for improvement, we find the greatest positive value. The
reason is obvious, positive value in the improvement indicates the greatest
contribution to the objectives.

7. Linear Programming: Assignment Method

- Another special-purpose algorithm used in linear programming is the Assignment
Method.
- The problem is concerned with allocating the jobs to each of the workers for
minimum cost.
- There are three main steps to follow in solving as assignment problem:
 Page 14 of 16

1. Subtract the smallest cost from each entry in each row. If each zero can
now be assigned one-to-one correspondence with the “workers”, an
optimal solution is reached. If it cannot, go on step 2.
2. Subtract the smallest cost in each column. If the zero entries can now be
distributed one-to-one correspondence with the “workers”, an optimal
solution is reached. If it cannot, go on step 3.
3. Cover the zero entries by vertical or horizontal lines, using the least
number of lines possible. (This can be done by covering first the row or
column having the most number of zeros). Subtract the smallest
uncovered cost from each uncovered cost but add it to the entry found at
the intersection of the lines. If an assignment is already possible, an
optimal solution is reached. If not, repeat step 3.
- An assignment is optimum if the number of lines used is equal is equal
to the number of rows or the number of columns.

8. Sample Problems – Assignment Problems

Sample 5: Assignment Method

Four engineers are to work on 4 projects of PSV Construction Company. The problem is
to decide which engineer should be assigned to which project. Each engineer charges
different fees on each project, due to distances of the projects and the complexity of
the work. The cost of assigning particular engineers to particular projects are as
follows.

Engineer Projects
s A B C D
1 1 8 1 7
1 0
2 6 5 3 8
3 4 8 1 9
0
4 1 1 5 7
1 0

The objective is to find the least cost of assignment.

Solution:
For easy computation, since all entries have 3 zero-endings, we may temporarily
eliminate the three zeros first.
RESULT AFTER REMOVING THREE ZEROS IN EACH ENTRY.
1 8 1 7
1 0
6 5 3 8
4 8 1 9
0
1 1 5 7
1 0
 Page 15 of 16

RESULT AFTER SUBTRACTING THE SMALLEST ENTRY IN EACH

Lowest is 7 in the previous table, deduct 7 from all entries
4 1 3 0
3 2 0 5 Lowest is 3 in the previous table, deduct 7 from all entries

0 4 6 5 Lowest is 4 in the previous table, deduct 7 from all entries

6 5 0 2
Lowest is 5 in the previous table, deduct 7 from all entries

SINCE THE ZEROS CANNOT YET BE ASSIGNED TO EACH ENGINEER/Project,

PROCEED
TO THE NEXT STEP (no zero yet in project B)

RESULT AFTER SUBTRACTING THE SMALLEST ENTRY IN EACH COLUMN.

4 0 3 0 Lowest is 0 in the previous table, deduct 0 from all entries
3 1 0 5
Lowest is 1 in the previous table, deduct 1 from all entries
0 3 6 5
6 4 0 2 Lowest is 0 in the previous table, deduct 0 from all entries

Lowest is 0 in the previous table, deduct 0 from all entries

COVERING ZERO ENTRIES WITH THE LEAST IN NUMBER OF

VERTICAL AND HORIZONTAL LINES.

4 0 3 0
Smallest uncovered entry is 1, so subtract all uncovered
3 1 0 5 entries with 1 and add it to the intersection of the lines.
0 3 6 5
6 4 0 2

RESULT AFTER SUBTRACTING 1, THE SMALLEST UNCOVERED FROM

EACH UNCOVERED ENTRY AND ADD IT TO THE INTERSECTION OF
THE LINES.

5 0 4 0
3 0 0 4
0 2 6 4
6 3 0 1

COVERING ZERO ENTRIES WITH THE LEAST IN NUMBER OF

VERTICAL AND HORIZONTAL LINES.
EN A B C D
G
1 5 0 4 0
2 3 0 0 4
3 0 2 6 4
4 6 3 0 1
3–A
1–D
4-C
2-B
 Page 16 of 16

This table is optimum since the least number of vertical or horizontal lines that can be
drawn is equal to the number of columns, and each Zero can now be assigned to each
engineer.

Assigning the projects to the engineers: (giving each zero to each engineer)
Options:
Engineer 1 can go to project B or D
Engineer 2 can go to project B or C
Engineer 3 can go to project A
Engineer 4 can go to project C

Decision:
Engineer 1 to D P7, 000
Engineer 2 to B 5, 000
Engineer 3 to A 4, 000
Engineer 4 to C 5, 000
Minimum Cost 21, 000

Explanation:
Since engineers 4 and 3 cannot go to any project except C and A respectively, then engineer
2 must go to project B, while engineer 1 has no other choice but D.

Application

Write your answers in any clean piece of paper. Take a picture and submit it via Google
Classroom. Ensure that every sheet of your submission is labeled with your name.
Show your solution and box-in your final answer.

PROBLEM 1: Fill up the initial table and use the stepping stone method to
come up with a decision
DISTRIBUTORS
PLANTS A B C SUPPLY
1 20 10 5
500
2 5 10 20
500
DEMAND 400 300 400
PROBLEM 2:
The table below has indicated per unit profit at the right hand corner of each
cell. Find the right decision for maximum profit. Find the maximum profit. Use
the MODI Method.
 Page 17 of 16

PLANTS DISTRIBUTORS SUPPLY

A B C D
70 20 10 40
1 25
20 60 30 20
15
2

3 30 40 50 20
45
4 50 40 30 50
25
DEMAND 20 10 40 40
PROBLEM 3:
Determine the optimal assignment of the management teams to the 4 projects
Team Projects
s A B C D
1 190 140 120 180
0 0 0 0
2 100 120 150 100
0 0 0 0
3 160 900 110 140
0 0 0
4 200 160 130 170
0 0 0 0

*** NOTHING FOLLOWS ***