Transportation Model and its Variants

Definition:
Transportation problem is a special kind of linear
programming problem in which goods are transported from
the sources to the destinations with respect to the supply and
demand of the sources and destinations respectively so that
the total transportation cost is minimum.
Representation of transportation problem:
Destinations
Sources Supply
D1 D2 D3 D4
C11 C12 C13 C14 a1
S1

S2 C21 C22 C23 C24 a2

S3 C31 C32 C33 C34 a3

b1 b2 b3 b4
Demands
Types of Transportation Problem:
Transportation problem is of 2 types
1) Balanced Transportation Problem
2) Unbalanced Transportation Problem
Balanced Transportation Problem:
Total supply (S) = Total Demand (D)
Destinations
1 2 3 4 supply
source
1 10 22 0 22 8

2 15 20 12 8 13
3 20 12 10 15 11

Demands 5 11 8 8 32

Total Supply (S) = Total Demand (D) =32

Unbalanced Transportation Problem:
Total supply (S) ≠ Total Demand (D)
• Convert the unbalanced problem to balanced problem by inserting rows and
columns according to the situation.

supply
40 25 22 33 100

44 35 30 30 30

38 38 28 30 70

40 20 60 30 Demand

Total supply (S)=200

Total demand (D)=150
So, Supply is not equal to demand. To make it equal insert a column of value 50
with unit cost cell value 0.
So transportation table becomes

supply
40 25 22 33 0 100
44 35 30 30 0 30
38 38 28 30 0 70
40 20 60 30 50 Demand
Methods for solving the transportation Problem:
1. Northwest-corner method
2. Least-cost method
3. Vogel approximation method(VAM)
Methods are used
• For finding the basic feasible solutions
• For finding the total transportation cost
• Checking the optimality
Northwest-corner method
Step-1 Select the northwest corner cell and then find out the value of
demand and supply for that cell.
Step-2 From the value of demand and supply the minimum value will be
allocating to the selected cell.
Step-3 After allocating which one (either supply and demand) will be zero
cancelled that row or column accordingly.
Step-4 Repeat the process till all the demands and supply both will be zero.
Problem: Solve the problem using Northwest corner method and find out the
basic feasible solutions and total transportation cost.
Destinations
sources 1 2 3 4 Supply
1 10 2 20 11 15
2 12 7 9 20 25
3 4 14 16 18 10
5 15 15 15 Demands
Feasible solutions are:
X11 =5; X12=10; X22=5; X23=15; X24=5; X34=10
If no of feasible solutions are equal to or less than (m+n-1)
then feasible solutions are called as basic feasible solutions.
Where m= no of rows, n= no of columns
So in this problem m=3, n=4
So m+n-1=3+4-1=6= no of feasible solutions
Total Transportation cost=
(5x10)+(10x2)+(5x7)+(15x9)+(5x20)+(10x18)
= 50+20+35+135+100+180
=Rs. 520
 • Problem2 Solve the transportation problem using Northwest corner
method and find out the basic feasible solutions and total transportation
cost
Destinations

Source 1 2 3 4 Supply

1 10 20 5 7 10

2 13 9 12 8 20

3 4 5 7 9 30

4 14 7 1 9 40

5 3 12 5 19 50

60 60 20 10

Demands
 1

10 20 5 7
13 9 010
12 8 10
4 5 7 9 20
14 7 1 9 30
3 12 5 19 40
20 50
10
60 60
Feasible solutions are:
X11 =10; X21=20; X31=30; X42=40; X52=20; X53=20;X54=10
Now put m+n-1=5+4-1=8
If no of feasible solutions are equal to or less than (m+n-1)
then feasible solutions are called as basic feasible solutions.
Basic Feasible solutions are:
X11 =10; X21=20; X31=30; X42=40; X52=20; X53=20;X54=10
Total Transportation cost=
(10x10)+(20x13)+(30x4)+(40x7)+(20x12)+(20x5)+(10x19)
= 100+260+120+280+240+100+190
=Rs. 1290
• Least Cost Method:
• Step-1: Select the least cost cell and then find out the corresponding value
of supply and demand for that cell.
• Step-2: Assign the supply or demand value, which is minimum to the
selected cell. After assigning the minimum supply or demand cancelled
that supply row or demand column which is zero.
• Step-3: Repeat the process till all the supply and demand becomes zero.
Problem: Solve the problem using Northwest corner method and find out the
basic feasible solutions and total transportation cost.

Destinations
sources 1 2 3 4 Supply
1 10 2 20 11 15
2 12 7 9 20 25
3 4 14 16 18 10
5 15 15 15 Demands
Solution
Feasible solutions are:
X12 =15; X23=15; X24=10; X31=5; X34=5
Now put m+n-1=3+4-1=6
If no. of feasible solutions are equal to or less than (m+n-1) then feasible
solutions are called as basic feasible solutions.
Basic Feasible solutions are:
X12 =15; X23=15; X24=10; X31=5; X34=5

Total Transportation cost=

(15x2)+(15x9)+(10x20)+(5x4)+(5x18)
= 30+135+200+20+90
= Rs. 475
Vogel Approximation Method (VAM):
• Step-1: Find out the row difference(Difference between two smallest nos) in the
same row and column difference (Difference between two smallest nos) in the
same column.
• Step-2: After finding the row difference and column difference select the
maximum no and corresponding to that find out the least cost cell and assign the
minimum value of either demand or supply. Cancelled hat row or column of having
zero value.
• Step-3: Repeat the process till all the supply and demand becomes zero.
Problem: Solve the problem using vogel approximation method and find out
the basic feasible solutions and total transportation cost.
Destinations
sources 1 2 3 4 Supply
1 10 2 20 11 15
2 12 7 9 20 25
3 4 14 16 18 10
5 15 15 15 Demands
Solution
Optimality Test:
After finding the basic feasible solutions optimality can be checked using UV
method.
Steps followed:
• Rows are represented as U and columns are represented as V
• For finding the value of U and V put Ui+ Vj= Cij for allocated cell.
• For finding the penalty value put the formulae Ui+ Vj-Cij for unallocated
cell.
• Optimality condition : The penalty values are either all zero or negative. If
not then it is not optimized. Go for further iteration.
• Make a closed loop by choosing the cell having maximum positive value
and the closed loop only covers the allocated cell.
• After making loop give +ve sign to the first cell where it is started and then
–ve sign to the next cell, then +ve alternatively.
• From the allocated –ve sign cells see allocated values which is minimum ,
subtract that value from the –ve cell and add that value to the +ve sign
cell.
• Repeat the process till getting all –ve or zero value in the penalty.
Final table:
Basic feasible solutions:
x11=5;x12=10;x22=5;x23=15;x23=5;x34=10
Transportation cost:
(5x10)+(10x2)+(5x7)+(15x9)+(5x20)+(10x18)= 520
for optimality check:
For finding the value of U and V
Formulae Ui+ Vj=Cij (For allocated cell only)
Put U1=0
U1+V1=C11
0+V1=10, V1=10
U1+V2=C12;
0+V2=2, V2=2
U2+V2=C22
U2+2=7, U2=5
U2+V3=C23
5+V3=9, V3=4
U2+V4=C24
5+V4=20, V4=15
U3+V4=C34,
U3+15=18, U3=3
Ui +Vj= Cij ; i=1,2,3; j= 1,2,3,4
U1+V1=10 -> 0+V1=10-> V1=10
U1+V2=2-> 0+V2=2-> V2=2
U2+V2=7->U2+2=7-> U2=5
U2+V3=9-> 5+V3=9-> V3=4
U2+V4=20-> 5+V4=20-> V4=15
U3+V4=18-> U3+15=18-> U3=3
Finding the penalty value for unallocated cell:
Pij= Ui+Vj-Cij
P13=U1+V3-C13=0+4-20=-16
P14=U1+V4-C14=0+15-11=4
P21=U2+V1-C21=5+10-12=3
P31=U3+V1-C31=3+10-4=9
P32=U3+V2-C32=3+2-14=-9
P33=U3+V3-C33=3+4-16=-9
Making the closed loop
Final Table
For finding the value of U and V
Formulae Ui+Vj=Cij (For allocated cell only)
Put U1=0
U1+V2=C12
0+V2=2, V2=2
U2+V2=C22;
U2+2=7, U2=5
U2+V3=C23
5+V3=9, V3=4
U2+V4=C24
5+V4=20, V4=15
U3+V4=C34
U3+15=18, U3=3
U3+V1=C31
3+V1=4, V1=1
Finding the value of U and V
For finding penalty value for unallocated cell:
Pij= Ui+Vj-Cij
P11=U1+V1-C11=0+1-10=-9
P13=U1+V3-C13=0+4-20=-16
P14=U1+V4-C14=0+15-11=4
P21=U2+V1-C21=5+1-12=-6
P32=U3+V2-C32=3+2-14=-9
P33=U3+V3-C33=3+4-16=-9
• CLOSED LOOP
Penalty for unallocated cell
P11=U1+V1-C11=0-3-10=-13
P13=U1+V3-C13=0+4-20=-16
P21=U2+V1-C21=5-3-12=-10
P24=U2+V4-C24=5+11-20=-4
P32=U3+V2-C32=7+2-14=-5
P33=U3+V3-C33=7+4-16=-5
All penalty values are –ve . So optimalityreached.
Basic solutions are X12=5,X14=10,X22=10,X23=15,X31=5,X34=5
Total Transportation cost=
(5x2)+(10x11)+(10x7)+(15x9)+(5x4)+(5x18)
=10+110+70+135+20+90= Rs 435
ASSIGNMENT PROBLEM:
Assign the jobs to the right operator to minimize the processing time and
cost.
Types of assignment problem(Balanced and Un balanced problem)
Balanced problem: No of rows = no of columns
Unbalance assignment problem : no of rows is not equal to no of columns
To make it balanced enter the row or column accordingly with zero entry.
Balance assignment problem is solved by Hungarian method
Procedure for Hungarian method
Phase 1
Step-1 Row and column deduction
Subtract the minimum value of each row from the entries of that row.
Step-2
Subtract the minimum value of each column from the entries of that column
Phase 2
Step-1:
Draw the minimum no. of lines to cover all the zeros of the
matrix.
Procedure:
Row scanning
• Starting from the 1st row , ask the following questions.
– Is exactly one zero in that row , if yes mark a square
around the zero entry and draw a vertical line passing
through that zero. Otherwise skip that row.
• After row scanning check whether all the zeros are covered
with lines if yes go to step-2 otherwise go for column
scanning.
Column scanning:
• Starting from the 1st column ask the following questions:
• Is exactly one zero in that column , if yes mark a square
around that zero entry and draw a horizontal line passing
through that zero, otherwise skip the column.

• After column scanning check whether all the zeros are

covered with lines
• Step-2
• Check whether the no. of square marked is equal to the no of
rows of the matrix . If yes go to step 5 otherwise go to step -3
• Step-3
• Identify the minimum value of undeleted cell values
• Add the minimum undeleted value to the
intersection points of the present matrix.
• Subtract the minimum undeleted cell value from
all the undeleted cell values
• All other entries are remain same.
Step-4 Go to step-1
Step-5
Treat the solution as marked by the squares as the
optimal solution. Optimality condition: If no of rows
= no of square
Solve the problem using Hungarian method and find out the
total processing time in hrs. after assigning the jobs to the
operator. operators

Jobs 1 2 3 4 5
1 9 11 14 11 7
2 6 15 13 13 10
3 12 13 6 8 8
4 11 9 10 12 9
5 7 12 14 10 14
• After row reduction
• Row Scanning
• Column Scanning

Optimality check(no of rows = no of square blocks)

• No of square blocks (4) is not equal to no of rows(5)

No of square blocks (5)= no of rows(5) .So optimality reached.
Jobs Operators processing time(hrs)
1 5 7
2 1 6
3 3 6
4 2 9
5 4 10
-----
38 hrs
Assignment problem: Solve the assignment problem using Hungarian method
and calculate the total no of processing time in hrs.
operators
Jobs 1 2 3 4 5
1 32 38 40 28 40 28
2 40 24 28 21 36 21
3 41 27 33 30 37 27
4 22 38 41 36 36 22
5 29 33 40 35 39 29

4 10 12 0 12
19 3 7 0 15
14 0 6 3 10
0 16 19 14 14
0 4 11 6 10
0 0 6 0 10
4 10 6 0 2
19 3 1 0 5
14 0 0 3 0
0 16 13 14 4
0 4 5 6 0

Optimality condition: no of square blocks=no of rows

4 9 5 0 2
19 2 0 0 5
15 0 0 4 1
0 15 12 14 4
0 3 4 6 0

No of blocks=no of rows
Jobs Operators processing
time(hrs.)
1 4 28
2 3 28
3 2 27
4 1 22
5 5 39
-----
144 hrs