Steps of Transportation and Assignment

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16

Steps for solution of Transportation Problems

1. Find an initial basic feasible solution by:

a. Northwest Corner Rule, or


b. Row-Minima method, or
c. Column-Minima Method, or
d. Least Cost Method, or
e. Vogel’s Approximation Method. (preferred method)

2. Test the solution for optimality by:

a. Stepping stone Method, or


b. MODI (Modified Distribution) Method or UV Method. (preferred method)

If optimality has been achieved, stop.

3. If optimality has not been achieved, improve the solution.

4. Retest for optimality and continue as above till optimality is achieved.

In examination, we may follow the Vogel’s Approximation Method, if nothing is stated about method to be followed.

Ex. A car manufacturing company has plants at cities A, B and C. Its destination centres are located at cities X and Y.
The capacity of three plants during the next quarter is 1000, 1500 and 1200 cars. The quarterly demand of the two
destination centres is 2300 and 1400 cars. The transportation cost per car per km is Rs. 2/-. The chart below shows the
distance in km between the plants and the distribution centres.

Distribution Centres
X Y
A 1000 2690
Plants B 1250 1350
C 1275 850

How many cars should be transported from which plant to which destination centre to minimize cost?

Ans: We first make the transportation cost matrix as follows: -

Distribution Centres
Supply
X Y
2000/- 5380/-
A 1000
2500/- 2700/-
Plants B 1500
2550/- 1700/-
C 1200

Demand 2300 1400 3700

Here our objective is to find out the values of Xij i.e. the number of cars to be transported from plant (source) i to
distribution centre (destination) j, so that the cost of transportation is minimized i.e.

Minimize, Z = 2000 X11 + 5380 X12 + 2500 X21 + 2700 X22 + 2550 X31 + 1700 X32 = Cij Xij ; where, Cij is
the unit cost of transportation from the ith source to jth destination.

1
Distribution Centres
Supply
X Y
2000/- 5380/-
A 1000
X11 X12
2500/- 2700/-
Plants B 1500
X21 X22
2550/- 1700/-
C 1200
X31 X32
Demand 2300 1400 3700

1. Northwest Corner Rule: Beginning with the (1, 1) cell i.e. the northwest corner; allocate to x11 as many units as
possible without violating the constraints. Thereafter, continue by moving one cell to the right, if supply remains,
or, if not, one cell down. At each step, allocate as much as possible to the cell (variable) till the process is
complete. The allocation may be zero, but not negative.

Distribution Centres
Supply
X Y
A 1000
1000
Plants B 1500
1300 200
C 1200
1200
Demand 2300 1400 3700

2. Row-Minima Method: In this method we allocate maximum possible in the lowest cost cell of the first row. The
idea is to exhaust the capacity of the first source or the demand at destination centre is satisfied or both. Continue
the process for the other reduced transportation costs until all the supply and demand conditions are satisfied.

Distribution Centres
Supply
X Y
2000/- 5380/-
A 1000
1000
2500/- 2700/-
Plants B 1500
1300 200
2550/- 1700/-
C 1200
1200
Demand 2300 1400 3700

3. Column Minimum Method: In this method, we start with the first column and allocate as much as possible in the
lowest cost cell of the column, so that either the demand of the first destination center is satisfied or the capacity of
the 2nd plant is exhausted or both. There are three cases: -
a. It the demand of the first distribution centre is satisfied, cross off the first column and move to the second
column to the right.
b. If the supply (capacity) of the ith plant is satisfied, cross off the ith row and reconsider, the first column
with the remaining demand.

2
c. If the supply (requirement) of the first distribution centre as also the capacity of the ith plant are
completely satisfied, make a zero allotment in the second lowest cost cell of the first column. Cross off
the column as well as the ith row and move to the second column.

Distribution Centres
Supply
X Y
2000/- 5380/-
A 1000
1000
2500/- 2700/-
Plants B 1500
1300 200
2550/- 1700/-
C 1200
1200
Demand 3700

4. Least Cost Method: In this method, we allocate as much as possible in the lowest cost cell and then move to the
next lowest cell / cells and so on.

Distribution Centres
Supply
X Y
2000/- 5380/-
A 1000
1000
2500/- 2700/-
Plants B 1500
1300 200
2550/- 1700/-
C 1200
1200
Demand 2300 1400 3700

5. Vogel’s Approximation Method (VAM): For each row or column having some supply or some demand
remaining, calculate its difference, which is the nonnegative difference between the two smallest shipping cost cij
associated with unassigned variable in that row or column. Consider that row or column having the largest
difference; in case of a tie, arbitrarily choose one. In this row or column, locate the unassigned variable (cell)
having the smallest unit shipping cost and allocate to it as many units as possible without violating the constraints.
Recalculate the new differences and repeat the above procedure until all demands are satisfied.

Distribution Centres
Supply
X Y
2000/- 5380/-
A 1000 (3380/-)
1000
2500/- 2700/-
Plants B 1500 (200/-)
2550/- 1700/-
C 1200 (850/-)
Demand 2300 1400 3700
(500/-) (1000/-)

3
Distribution Centres
Supply
X Y
A 1000
1000
2500/- 2700/-
Plants B 1500 (200/-)
2550/- 1700/-
C 1200 (850/-)
1200
Demand 2300 1400 3700

(50/-) (1000/-)

Distribution Centres
Supply
X Y
A 1000
1000
2500/- 2700/-
Plants B 1500 (200/-)
1300 200
C 1200
1200
Demand 2300 1400 3700

A nondegenerate basic feasible solution will be characterized by positive values for exactly (m + n – 1) basic
variables [m = number of rows and n = number of column]. If the initial basic feasible solution does not yield the
same number of positive values, additional variables with zero allocations must be designated as basic. This choice
is arbitrary, but to the point: BASIC VARIABLES CANNOT FORM A LOOP and preference is usually given
to variables with the lowest associated shipping costs.

Test for Optimality:

Assign one (any one) of the ui or vj in tableau the value zero and calculate the remaining ui or vj so that for each basic
variable ui + vj = cij. Then, for each nonbasic variable, calculate the quantity (cij - ui - vj). If all these latter quantities
are nonnegative, the current solution is optimal; otherwise, the current solution is not optimal.
(S = Source; D = Destination)

D1 D2 .. Dn Supply ui
S1 a1 u1
S2 a2 u2
. .. ….
Sm am um
Demand b1 b2 … bn
vj v1 v2 …. vn

4
Improving the Solution: Consider the nonbasic variable corresponding to the most negative of the quantities (cij - ui -
vj) calculated in the test for optimality; it is made the incoming variable. Construct a loop consisting exclusively of this
incoming variable (cell) and current basic variables (cells). Then allocate to the incoming cell as many units as
possible such that, after appropriate adjustments have been made to the other cells in the loop, the supply and demand
constraints are not violated, all allocations remain nonnegative, and one of the old basic variables has been reduced to
zero (whereupon it cease to be zero).

+ Incoming
- Variable

+
-

+ -

5
Illustration: National Oil Company has three refineries and four depots. Transportation costs per ton, capacities
and requirements are as follows.

Destination Supply (Capacity)


D1 D2 D3 D4 in tons
S1 5 7 13 10 700
Source S2 8 6 14 13 400
S3 12 10 9 11 800
Demand (requirement) in tons 300 600 700 400

Determine optimal allocation of output.

Ans. We introduce a dummy source S4 for transforming the given unbalanced cost matrix into a balanced one and
for the cells relating to the dummy source we assume the unit costs as zero.
We start with the following transformed cost matrix and use the Vogel’s Approximation Method (VAM) to find
the initial basic feasible solution (IBFS).

Destination Row difference


Supply
1st 2nd 3rd
D1 D2 D3 D4
S1 5 7 13 10 700 2
Source
S2 8 6 14 13 400 2
S3 12 10 9 11 800 1
S4 0 0 0 0 100 0
30 60 70 40
Demand 2,000
0 0 0 0
1st 5 6 9 10*

Column 2n
difference d
3r
d

We consume the full supply from source S4 by allocating the quantity of 100 ton to destination D4 and remove the
source S4 from further consideration.

Destination Row difference


Balance
Supply 1st 2nd 3rd
D1 D2 D3 D4
S1 5 7 13 10 700 2
Source
S2 8 6 14 13 400 2
S3 12 10 9 11 800 1

Balance 30 60 70 30
1,900
Demand 0 0 0 0
1st
Column 2n
3 1 4* 1
difference d
3r
d

6
We meet the full requirement of destination D3 by allocating the quantity of 700 ton from source S3 and remove the
source D3 from further consideration.

Destination Row difference


Balance
Supply 1st 2nd 3rd
D1 D2 D4
S1 5 7 10 700 2
Source
S2 8 6 13 400 2
S3 12 10 11 100 1

Balance 30 60 30
1,200
Demand 0 0 0
1st
Column 2n
difference d
3r
3* 1 1
d

We meet the full requirement of destination D1 by allocating the quantity of 300 ton from source S1and remove the
source D1 from further consideration.

Destination Row difference


Balance
2n
Supply 1st d 3rd 4th 5th
D2 D4
S1 7 10 400 3
Source
S2 6 13 400 7*
S3 10 11 100 1

Balance 60 30
900
Demand 0 0
1st
2n
d
Column
difference 3r
d
4th 1 1
5th

We consume the full supply from source S2 by allocating the quantity of 400 ton to destination D2 and remove the
source S2 from further consideration.

Destination Row difference


Balance
2n
Supply 1st d 3rd 4th 5th
D2 D4
Source S1 7 10 400 3

7
S3 10 11 100 1

Balance 20 30
500
Demand 0 0
1st
2n
d
Column
difference 3r
d
4th
5th 3* 1

We meet the full requirement of destination D2 by allocating the quantity of 200 ton from source S1 and remove the
source D2 from further consideration.

Destination Row difference


Balance
2n
Supply 1st d 3rd 4th 5th
D4
S1 10 200
Source

S3 11 100

Balance 30
300
Demand 0
1st
2n
d
Column
difference 3r
d
4th
5th

Finally, we allocate 200 tons from source S1 and 100 tons from source S3 to the destination D4 to complete the
allocation.

So the initial basic feasible solution (IBFS) becomes:-

Destination
Supply
D1 D2 D3 D4
30 20 20
S1 5 7 10 700
0 0 0
Source
40
S2 6 400
0
70 10
S3 9 11 800
0 0
10
S4 0 100
0
30 60 70 40
Demand 2,000
0 0 0 0

8
Let us now test for optimality of this initial basic feasible solution.
Here, [m + n – 1] = [4 + 4 -1] = 7. As there are seven positive allocations in the IBFS, so we can start the optimality
test.

Let us take u1 = 0 as there are maximum allocations in the first row. Considering only the basic variables, we find out
the values of all ui and vj as shown below using the formula, [cij = ui + vj].

Destination
Supply ui
D1 D2 D3 D4
30 20 20
S1 5 7 10 700 0
0 0 0
Source
40
S2 6 400 -1
0
70 10
S3 9 11 800 1
0 0
10
S4 0 100 - 10
0
30 60 70 40
Demand 2,000
0 0 0 0
vj 5 7 8 10

Now we calculate the (cij - ui - vj) values [bold, italics and also underscore values] for the nonbasic variables as
follows: -

Destination
Supply ui
D1 D2 D3 D4
S1 13 5 700 0
Source
S2 8 4 14 7 13 4 400 -1
S3 12 6 10 2 800 1
S4 0 5 0 3 0 2 100 - 10
30 60 70 40
Demand 2,000
0 0 0 0
vj 5 7 8 10

As the (cij - ui - vj) values for all the nonbasic cells are nonnegative, we have reached the optimal solution in the IBFS
itself.

As the source R4 is a dummy, hence the destination D4 will only receive a total quantity of 300 tons instead of its
requirement of 400 tons. The optimum allocations are

R1D1 = 300 tons; R1D2 = 200 tons; R1D4 = 200 tons;


R2D2 = 400 tons;
R3D3 = 700 tons; R3D4 = 100 tons.

Optimal cost = 300 x 5/- + 200 x 7/- + 200 x 10/- + 400 x 6/- + 700 x 9 /- + 100 x 11/- = 14,700/-.

Ex: Solve the following transportation problem by VAM.

9
Ans:

Illustration of Maximization in Transportation Problem: There are certain types of transportation problems where
the objective function is to be maximized. These problems can be solved by converting the maximization problem into
a minimization problem.

Surya Roshni limited has three factories – X, Y and Z. It supplies goods to four dealers A, B, C and D spread all over
the country. The production capacities of these factories are 200, 500 and 300 per month respectively. The unit net
return (profit) for each factory-dealer combination is given in the following table.

Dealer Production
Factory
A B C D Capacity
X 12 18 6 25 200
Y 8 7 10 18 500
Z 14 3 11 20 300
Demand 180 320 100 400

Determine a suitable allocation to maximize the total net return.

Ans: Maximization transportation problem can be converted into minimization transportation problem by subtracting
each transportation cost from maximum transportation cost. Here, the maximum transportation cost is 25. So subtract
each value from 25. The revised transportation problem is shown below: -

Revised transportation problem


10
Dealer Production
Factory
A B C D Capacity
X 13 7 19 0 200
Y 17 18 15 7 500
Z 11 22 14 5 300
Demand 180 320 100 400

An initial basic feasible solution is obtained by VAM method and subsequently we find the final solution which is
shown below: -

Dealer Production
Factory
A B C D Capacity
X 200 200
Y 120 100 280 500
Z 180 120 300
Demand 180 320 100 400

Total maximum net return = [200 x 18 + 120 x 7 + 100 x 10 + 280 x 18 + 180 x 14 + 120 x 20] = Rs. 15,400.

11
Steps for solution of Assignment Problems (Hungarian Method)

This means scheduling of workers to jobs on a one-to-one basis. The number of workers n is presumed to be equal to
the number of jobs (if required, create either a fictitious worker or job.)

1. In each row of the cost or time matrix (tableau), locate the smallest element and subtract it from every element in
that row. Record the results in new tableau.
Repeat this procedure for each column (the column minimum is determined from the new table after the row
subtraction). Record the results in a revised table (cost matrix).

2. Determine whether there exists a feasible assignment involving only zero costs in the revised cost matrix. In other
words, find if the revised matrix has n zero entries no two of which are in the same row or column. If such an
assignment exists, it is optimal. If no such exists, go to step 3.

3. Cover all zeros in the revised cost matrix with as few horizontal and vertical lines as possible. Each horizontal line
must pass through an entire row, each vertical line must pass through an entire column, the total number of lines in
this minimal covering will be smaller than n.
Locate the smallest number in the revised cost matrix not covered by a line. Subtract this number from every
element not covered by a line and add it to every element covered by two lines.

4. Return to step 2.

Example: A company is faced with problem of assigning five jobs to five machines; each job must be done only on
one machine; the cost of processing each job on each machine is given in the following table (in Rs.).
Machines
M1 M2 M3 M4 M5
J1 7 5 9 8 11
J2 9 12 7 11 10
Jobs J3 8 5 4 6 9
J4 7 3 6 9 5
J5 4 6 7 5 11

Determine the optimal assignment of jobs to machines so that in will result in minimum total cost.
Ans.
1. Select the minimum element in each row and subtract this element from every element in that row. We get

Machines
M1 M2 M3 M4 M5
J1 2 0 4 3 6
J2 2 5 0 4 3
Jobs J3 4 1 0 2 5
J4 4 0 3 6 2
J5 0 2 3 1 7

2. Next, select the minimum element in each column and subtract this element from every element in that column.
We get
Machines
M1 M2 M3 M4 M5
J1 2 0 4 2 4
J2 2 5 0* 3 1
Jobs J3 4 1 0 1 3
J4 4 0 3 5 0
J5 0* 2 3 0 5

As a result of this step there would be at least one zero in each of the rows and columns in the reduced cost matrix.

12
3. Now we attempt to make a complete set of assignments using only a single zero element in each row or column. In
other words, find if the reduced cost matrix has n zero entries with no two of which are in the same row or
column. If such an assignment exists, it is optimal. Otherwise we proceed to the next step.

Machines
M1 M2 M3 M4 M5
J1 2 0* 4 2 4
J2 2 5 0* 3 1
Jobs J3 4 1 0 1 3
J4 4 0 3 5 0*
J5 0* 2 3 0 5

Since row J1 contains only single zero, therefore assignment is made in the cell J1M2 (marked as 0*) and other zero(s)
appearing in the corresponding column M2 is crossed out (marked as 0)
Similarly row J2 contains only single zero, therefore assignment is made in the cell J2 M3 and other zero(s) appearing
in the corresponding column M3 is crossed out (marked as 0).
For row J3 there is no assignable (left out) zero, hence no assignment can be made.
Row J4 contains one assignable (left out) zero; therefore assignment is made in the cell J4 M5.
Row J5 contains two assignable zeros; assignment is made in the first cell J5M1and other zero(s) appearing in the
corresponding row J5 is crossed out (marked as 0).
Thus it is possible to make only four of the five necessary assignments using the zero element positions. So the
optimum solution has not been reached.

4. Cover all zeros in the reduced cost matrix with as few horizontal and vertical lines as possible. We get

Machines
M1 M2 M3 M4 M5
J1 2 0* 4 2 4
J2 2 5 0* 3 1
Jobs J3 4 1 0 1 3
J4 4 0 3 5 0*
J5 0* 2 3 0 5

The total number of lines in this minimal covering (will be smaller than n which is 5 in this case) is four.

5. Locate the smallest number in the reduced cost matrix not covered by a line. In this case this is 1. Subtract this
number (i.e.1) from every element not covered by a line and add it to every element covered by two lines. We get
the following revised cost matrix.

Machines
M1 M2 M3 M4 M5
J1 2 0 5 2 4
J2 1 4 0 2 0
Jobs J3 3 0 0 0 2
J4 4 0 4 5 0*
J5 0 2 4 0 5

6. Now we make fresh assignments on this revised cost matrix. Proceeding in the usual way as explained in step 3 we
get
Machines
M1 M2 M3 M4 M5
Jobs J1 2 0* 5 2 4
J2 1 4 0* 2 0

13
J3 3 0 0 0* 2
J4 4 0 4 5 0*
J5 0* 2 4 0 5

The optimum solution has been reached since all the five assignment have been made.

7. The optimum solution is as follows: -

Assign job To machine Cost in Rs.


J1 M2 5
J2 M3 7
J3 M4 6
J4 M5 5
J5 M1 4
Minimum total cost 27

Ex. A 400 meter swimming relay involves four different swimmers, who successfully swim 100 meter backstroke,
breaststroke, butterfly and freestyle. A coach has six very fast swimmers (Kaushik, Arindam, Tapas, Suman, Kanak and
Mohan), whose expected times (in second) in the individual events are given in the following table: -

4 x 100 meter Swimming Relay


Swimmer
Backstroke Backstroke Butterfly Freestyle
Kaushik 65 73 63 57
Arindam 67 70 65 58
Tapas 68 72 69 55
Suman 67 75 70 59
Kanak 71 69 75 57
Mohan 69 71 66 59

How should the coach assign swimmers to the 4 x 100 meter relay so as to minimize the sum of times?
Ans.

Optimum i.e. Minimum total time, Z* = C11 + C23 + C34 + C52 = 65 + 65 + 55 + 69 = 254 Secs.

Ex.
5. An Airline Company has drawn up a new flight schedule involving five flights. To assists in allocating five pilots
to the flight, it has asked to state their preferences score by giving each flight a number out of 10. The higher the
number, the greater is the preference. Certain of these flights are unsuitable to some pilots owing to domestic
reasons. These have been marked with “X”.

Flight Number
Pilots 1 2 3 4 5
A 8 2 X 5 4
B 10 9 2 8 4
C 5 4 9 6 X
14
D 3 6 2 8 7
E 5 6 10 4 3

Ans: At first, we assign nil value to the cells marked “X”. Then we deduct the numbers in each cell from 10 and form a
revised minimization matrix as follows: -

So the allocation should be A1; B2; C3; D4 and E5.

Ex.
6. Priyanshu Enterprise has three factories at locations A, B and C which supplies three warehouses located at D, E
and F. Monthly factory capacities are 10, 80 and 15 units respectively. Monthly warehouse requirements are 75,
20 and 50 units respectively. Unit shipping costs are given below:

Warehouse
Factory
D E F
A 5 1 7
B 6 4 6
C 3 2 5

Determine the optimal distribution for Priyanshu Enterprise, using any know algorithm for:

i. Case I: when there is no penalty cost for not satisfying demand at any warehouse.
ii. Case II: when the penalty costs for not satisfying demand at warehouse D, E and F are Rs. 5, Rs. 3 and Rs.
2 per unit respectively.
iii. Case III: when the penalty costs for not satisfying demand at warehouse D, E and F are Rs. 2, Rs. 3 and
Rs. 4 per unit respectively.
iv. Case IV: when the penalty costs for not satisfying demand at warehouse D, E and F are Rs. 5, Rs. 2 and
Rs. 4 per unit respectively.

Ans:
Case Transportation Model Optimal Solution

15
I

II

III

IV

16

You might also like