Week16 /Lecture15

Transportation models: Special cases

Introduction
If the basic feasible solution of a transportation problem with m origins and n
destinations has fewer than m + n 1 positive xij (occupied cells), the problem is
said to be a degenerate transportation problem.
egenerac! can occur at two stages"
1. #t the initial solution
$. uring the testing of the optimal solution
%o resolve degenerac!, we ma&e use of an artificial 'uantit! (d). %he 'uantit! d is
assigned to that unoccupied cell, which has the minimum transportation cost.
%he use of d is illustrated in the following example.
Example
Factory Dealer Supply
1 2 3 4
A 2 2 2 4 1000
B 4 6 4 3 700
C 3 2 1 0 900
Requirement 900 800 500 400
Solution.
#n initial basic feasible solution is obtained b! (atrix (inimum (ethod.
Table 1
Factory Dealer Supply
1 2 3 4
A 2 4 1000
B 4 4 3 700
C 3 2 900
Requirement 900 800 500 400
)umber of basic variables * m + n 1 * + + , 1 * -
.ince number of basic variables is less than -, therefore, it is a degenerate
transportation problem.
%o resolve degenerac!, we ma&e use of an artificial 'uantit!(d). %he 'uantit! d is
assigned to that unoccupied cell, which has the minimum transportation cost.
In the above table, there is a tie in selecting the smallest unoccupied cell. In this
situation, !ou can choose an! cell arbitraril!. /e select the cell 0$ as shown in
the following table.
Table 2
Factory Dealer Supply
1 2 3 4
A 2 4 1000
B 4 4 3 700
C 3 900 + d
Requirement 900 800 + d 500 400 2600 + d
)ow, we use the stepping stone method to find an optimal solution.
Calculating opportunity cot
Unoccupied
cells
Increase in cost per unit of
reallocation
Remars
A3 +2 2 + 2 1 = 1 Cost Increases
A4 +4 2 + 2 0 = 4 Cost Increases
B1 +4 6 + 2 2 = 2 Cost Decreases
B3 +4 6 + 2 1 = 1 Cost Decreases
B4 +3 6 + 2 0 = 1 Cost Decreases
C1 +3 2 + 2 2 = 1 Cost Increases
%he cell 11 is having the maximum improvement potential, which is e'ual to 2$.
%he maximum amount that can be allocated to 11 is 344 and this will ma&e the
current basic variable corresponding to cell 1$ non basic. %he improved solution is
shown in the following table.
Table !
Factory Dealer Supply
1 2 3 4
A 2 4 1000
B 6 4 3 700
C 3 900
Requirement 900 800 500 400 2600
%he optimal solution is
$ 5 $44 + $ 5 644 + , 5 344 + $ 5 d + 1 5 744 + 4 5 ,44 * 7+44 + $d.
)otice that d is a ver! small 'uantit! so it can be neglected in the optimal
solution. %hus, the net transportation cost is 8s. 7+44
Unbalanced Transportation Problem
.o far we have assumed that the total suppl! at the origins is e'ual to the total
re'uirement at the destinations.
.pecificall!,
.i * j
1ut in certain situations, the total suppl! is not e'ual to the total demand. %hus,
the transportation problem with une'ual suppl! and demand is said to be
unbalanced transportation problem.
If the total suppl! is more than the total demand, we introduce an additional
column, which will indicate the surplus suppl! with transportation cost 9ero.
.imilarl!, if the total demand is more than the total suppl!, an additional row is
introduced in the table, which represents unsatisfied demand with transportation
cost 9ero. %he balancing of an unbalanced transportation problem is illustrated in
the following example.
Example
!lant "are#ouse Supply
"1 "2 "3
A 28 17 26 500
B 19 12 16 300
Demand 250 250 500
Solution"
%he total demand is 1444, whereas the total suppl! is 644.
.i : j
%otal suppl! : total demand.
%o solve the problem, we introduce an additional row with transportation cost
9ero indicating the unsatisfied demand.
!lant "are#ouse Supply
"1 "2 "3
A 28 17 26 500
B 19 12 16 300
Unsatisfied demand 0 0 0 200
Demand 250 250 500 1000
;sing matrix minimum method, we get the following allocations.
!lant "are#ouse Supp
ly
"1 "2 "3
A 17 500
B 19 300
Unsatisfied demand 0 0 200
Demand 250 250 500 1000
Initial baic #eaible olution
74 5 $6 + ,74 5 $- + $74 5 1$ + 74 5 1- + $44 5 4 * 1-<44.
Maximization In A Transportation Problem
%here are certain t!pes of transportation problems where the objective function is
to be maximi9ed instead of being minimi9ed. %hese problems can be solved b!
converting the maximi9ation problem into a minimi9ation problem.
Example
.ur!a 8oshni =td. has three factories 2 5, >, and ?. It supplies goods to four
dealers spread all over the countr!. %he production capacities of these factories
are $44, 744 and +44 per month respectivel!.
Factory Dealer Capacity
A B C D
$ 12 18 6 25 200
% 8 7 10 18 500
& 14 3 11 20 300
Demand 180 320 100 400
etermine a suitable allocation to maximi9e the total net return.
Solution.
(aximi9ation transportation problem can be converted into minimi9ation
transportation problem b! subtracting each transportation cost from maximum
transportation cost.
@ere, the maximum transportation cost is $7. .o subtract each value from $7.
%he revised transportation problem is shown below.
Table 1
Factory Dealer Capacity
A B C D
$ 13 7 19 0 200
% 17 18 15 7 500
& 11 22 14 5 300
Demand 180 320 100 400
#n initial basic feasible solution is obtained b! matrix2minimum method and is
shown in the final table.
$inal table
Factory Dealer Capaci
ty
A B C D
$
13 7 19 200
%
7 500
&
22 14 300
Demand 180 320 100 400
T%e maximum net return i
$7 5 $44 + 6 5 64 + 3 5 +$4 + 14 5 144 + 1, 5 144 + $4 5 $44 * 1,$64.
Prohibited Routes
.ometimes there ma! be situations, where it is not possible to use certain routes
in a transportation problem. Aor example, road construction, bad road conditions,
stri&e, unexpected floods, local traffic rules, etc. /e can handle such t!pe of
problems in different wa!s"
# ver! large cost represented b! ( or is assigned to each of such routes,
which are not available.
%o bloc& the allocation to a cell with a prohibited route, we can cross out
that cell.
%he problem can then be solved in its usual wa!
Example
0onsider the following transportation problem.
Factory "are#ouse Supply
"1 "2 "3
F1 16 12 200
F2 14 8 18 160
F3 26 16 90
Demand 180 120 150 450
Solution.
#n initial solution is obtained b! the matrix minimum method and is shown in the
final table.
$inal Table
Factory "are#ouse Supply
"1 "2 "3
F1
200
F2
18 160
F3
16 90
Demand 180 120 150 450
Initial baic #eaible olution
1- 5 74 + 1$ 5 174 + 1, 5 ,4 + 6 5 1$4 + $- 5 <4 * -,-4.
%he minimum transportation cost is 8s. -,-4.
Time Minimizing Problem
.uccinctl!, it is a transportation problem in which the objective is to minimi9e the
time. %his problem is same as the transportation problem of minimi9ing the cost,
expect that the unit transportation cost is replaced b! the time tij.
Step
1. etermine an initial basic feasible solution using an! one of the following"
)orth /est 0orner 8ule
(atrix (inimum (ethod
Bogel #pproximation (ethod
$. Aind %& for this feasible plan and cross out all the unoccupied cells for which tij
%&.
+. %race a closed path for the occupied cells corresponding to %&. If no such closed
path can be formed, the solution obtained is optimum otherwise, go to step $.
Example 1
%he following matrix gives data concerning the transportation times tij
Destination
'ri(in D1 D2 D3 D4 D) D* Supply
'1 25 30 20 40 45 37 37
'2 30 25 20 30 40 20 22
'3 40 20 40 35 45 22 32
'4 25 24 50 27 30 25 14
Demand 15 20 15 25 20 10
Solution.
/e compute an initial basic feasible solution b! north west corner rule which is
shown in table 1.
Table 1
Destination
'ri(in D1 D2 D3 D4 D) D* Supply
'1 40 45 37 37
'2 30 25 40 20 22
'3 40 20 40 22 32
'4 25 24 50 27 14
Demand 15 20 15 25 20 10
@ere, t11 * $7, t1$ * +4, t1+ * $4, t$+ * $4, t$, * +4, t+,* +7, t+7 * ,7, t,7 *+4, t,-
* $7
0hoose maximum from tij, i.e., %1 * ,7. )ow, cross out all the unoccupied cells
that are %1.
%he unoccupied cell (C+-) enters into the basis as shown in table $.
Table 2
0hoose the smallest value with a negative position on the closed path, i.e., 14.
0learl! onl! 14 units can be shifted to the entering cell. %he next feasible plan is
shown in the following table.
Table !
Destination
'ri(in D1 D2 D3 D4 D) D* Supply
'1 40 37 37
'2 30 25 40 20 22
'3 40 20 40 32
'4 25 24 27 25 14
Demand 15 20 15 25 20 10
@ere, %$ * (ax($7, +4, $4, $4, $4, +7, ,7, $$, +4) * ,7. )ow, cross out all the
unoccupied cells that are %$.
Table &
1! following the same procedure as explained above, we get the following revised
matrix.
Table 6
Destination
'ri(in D1 D2 D3 D4 D) D* Supp
ly
'1 37 37
'2 30 25 20 22
'3 20 32
'4 25 24 27 25 14
Demand 15 20 15 25 20 10
%+ * (ax($7, +4, $4, $4, +4, ,4, +7, $$, +4) * ,4. )ow, cross out all the
unoccupied cells that are %+.
)ow we cannot form an! other closed loop with %+.
@ence, the solution obtained at this stage is optimal.
%hus, all the shipments can be made within ,4 units.