Transportation Problem in LPP

• Special case of LPP

• Objective is to minimize cost like
• Supplying products from various sources to
the defined destinations
• Satisfying both demand the supply limits and
the demand requirement.
 Model Assumption
• The model assumes that the distribution cost
on a given route is directly proportional to the
number of units distributed on that routes
• Transportation model can extended to other
areas like
o Inventory control
o Employment scheduling
o Machine assignment
o Plants location
o Product mix and others
Note that
• The transportation models are primarily
concerned with the optimal way in which a
product produced at different factories or
plants (called supply origins) can be
transported to a number of warehouses or
customers (called demand destinations
 Cont..
• There are special procedures or methods to
solve transportation problems
o North –west corner method
o Least cost method
o Vogel's Approximation Methods (VAM)
• These adopted due to large number of
constraints which brings complexity of using
simplex methods or graphical method
 Formulation Transportation Problem
• Mathematical model of transportation
Min 𝑍 = 𝑚 𝑛
𝑖=1 𝑗=1 𝑐𝑖𝑗 𝑥𝑖𝑗
𝑚
S. t 𝑗=1 𝑥𝑖𝑗 ≤ 𝑠𝑖 for (𝑖 = 1,2 … 𝑚)…sup constraints
𝑛
𝑖=1 𝑥𝑖𝑗 ≥ 𝑑𝑗 for (𝑗 = 1,2 … 𝑛)…dem constraints
𝑥𝑖𝑗 ≥ 0 𝑓𝑜𝑟 𝑖 = 1,2. . 𝑚 𝑎𝑛𝑑 𝑗 = 1,2 … 𝑛
• Objective function and constraints equations are all
linear in 𝑥𝑖𝑗
• If a problem has the constraints given above and is a
maximization problem, it is still a transportation
problem
 Cont.…
Where by
• 𝑠𝑖 = quantity of product available at origin
where by 𝑖 = 1,2 … 𝑚
• 𝑑𝑖 = quantity of product required at the
destination 𝑗 where by 𝑗 = 1,2 … 𝑛
• 𝑐𝑖𝑗 = Cost of transporting one unit of product
from origin 𝑖 to destination 𝑗
• 𝑥𝑖𝑗 = Quantity transported from origin 𝑖 to
destination 𝑗
 Cont..
• There be 𝑚 sources and 𝑛 destination.
• The objective is to determine the number of
units to be transported from source "𝑖" to
destination "𝑗” so that the total
transportation cost is minimum.
 General Algorithm to a TP
1. Formulate the modal in matrix form
2. Standardize the TP
3. If the TP is Min case then it is in standard
from
4. If it is Max case convert it in to standard form
o Multiply each elements of each cells by (-1)
o Subtract the element of each cell from the highest
element and placing in the respective cell
5. Obtaining balanced TP
o It is balance if supply = demand constraints
 Cont..
• If the condition is not satisfied, then the
problem is called unbalanced TP
o It is balanced by adding dummy row/ column
• Find IBFS and BFS to balanced TP by using
either of the following methods
o North –west corner method
o Least cost method
o Vogel's Approximation Methods (VAM)
 To find BFS to balanced TP
The following are needed
• Testing for optimality by the following method
o Stepping stone method
o Modified distribution method (MODI)
• Updating the solution
• Moreover the above step is repeated until the
optima solution is reached
 Cont.
What makes difference of the three methods?
• The quality of the IBFS they produce
o A better initial solution yields a smaller objective
value
o Generally VAM produce best IBFS than others
o However the North-West Corner method
produces the worst, but involves least
computation than others
North West Corner Method (NWC)
• Basic variables are selected from the NWC (i.e., top
left corner).
• The method starts at the northwest-corner cell
(route)
• Advantage of the NWC is that, it is very simple and
easy to apply
• However, it is not sensitive to costs and consequently
yields poor initial solutions
 The NWC Summary of Steps
1. Allocate as much as possible to the cell in the
upper left-hand corner, subject to the supply
and demand by subtracting the allocated
amount.
2. Allocate as much as possible to the next
adjacent feasible cell.
3. Repeat step 2 until all rim requirements are
met
Example of TP
City 1 City 2 City 3 City 4 Supply
Plant 1 𝑥11 8 𝑥12 6 𝑥13 10 𝑥14 9 35
Plant 2 𝑥21 9 𝑥22 12 𝑥23 13 𝑥24 7 50
Plant 3 𝑥31 14 𝑥32 9 𝑥33 16 𝑥34 5 40
Demand 45 20 30 30 125

Total supply = Total demand =125

Therefore it is a balanced TP
𝑚 𝑛
𝑠𝑖 = 𝑑𝑗
𝑖=1 𝑗=1
 Solution of TP
• Steps
o Determine the IBFS
o Test for optimization
• A general TP with 𝑚 sources and 𝑛 destination
has 𝑚 + 𝑛 constraint questions.
• The starting basic solution consist of 𝑚 + 𝑛 − 1
• If 𝑚 + 𝑛 − 1 variables yields a basic solution,
the cell corresponding to this set contain no
loop
Example: Consider the following Tableau
For SRTM
Sun Ray Transportation Model
Mills
1 2 3 4 Supply
𝑠1 10 10 20 11 15
𝑠2 12 7 9 25
𝑠3 4 14 16 16 10
Demand 5 15 15 15 50
Use North-West corner method to find the IBFS of the
above TP and its corresponding transportation cost.
m+n-1= 3+4-1= 6 variables
Solution to the TP to get IBFS
Sun Ray Transportation Model
Mills
1 2 3 4 Supply
5 10 15-5 =10
𝑠1 10 2 20 11 10-10=0
25-5=20
5 15 20-15=5
𝑠2 12 7 9 20 5-5=0
10
𝑠3 4 14 16 18 10-10=0
15-10=5
Demand 5-5=0 5-5=0 15-15=0 15 50
Therefore, the initial basic solution
• 𝑥11 = 5, 𝑥12 =10, 𝑥22 =5, 𝑥23 =15, 𝑥24 =5 and
𝑥34 =10
• The corresponding transportation cost is
• The OF is given by Min Z = 𝑚 𝑛
𝑖=1 𝑗=1 𝑐𝑖𝑗 𝑥𝑖𝑗
• Z= (5*10)+(10*2)+(5*7)+(15*9)+(5*20)+(18*10)
• Min of Z=520 $
Example
Powerco has three electric power plants that supply
the needs of four cities. Each power plant can supply
the following numbers of kwh of electricity: plant 1, 35
million; plant 2, 50 million; and plant 3, 40 million. The
peak power demands in these cities as follows (in kwh):
city 1, 45 million; city 2, 20 million; city 3, 30 million;
city 4, 30 million. The costs of sending 1 million kwh of
electricity from plant to city is given in the table below.
To minimize the cost of meeting each city’s peak power
demand, formulate a balanced transportation problem
in a transportation tableau and represent the problem
as a LP model.
 TP in a Transportation Tableau
To
From City 1 City 2 City 3 City4 Supply
Plant 1 8$ 6$ 10$ 9$ 35
Plant 2 9$ 12$ 13$ 7$ 50
Plant 3 14$ 9$ 16$ 5$ 40
Demand 45 20 30 30 125

Total supply & total demand both equal 125:

“balanced transportation problem”.
Representation of the problem as a LP
model
• 𝑥𝑖𝑗 : number of (million) kwh produced at plant i and sent to city j.
• Objective function will be
• Min 𝑧 = 8 𝑥11 + 6 𝑥12 + 10 𝑥13 + 9 𝑥14 + 9 𝑥21 + 12 𝑥22 + 13 𝑥23 +
7 𝑥24 + 14 𝑥31 + 9 𝑥32 + 16 𝑥33 + 5 𝑥34
𝑥11 + 𝑥12 + 𝑥13 + 𝑥14 ≤ 35 (supply constraints)
𝑥21 + 𝑥22 + 𝑥23 + 𝑥24 ≤ 50
𝑥31 + 𝑥32 + 𝑥33 + 𝑥34 ≤ 40
𝑥11 + 𝑥21 + 𝑥31 ≥ 45 (demand constraints)
𝑥12 + 𝑥22 + 𝑥32 ≥ 20
𝑥13 + 𝑥23 + 𝑥33 ≥ 30
𝑥14 + 𝑥24 + 𝑥34 ≥ 30
𝑥𝑖𝑗 ≥ 0 (i = 1, 2, 3; j = 1, 2, 3, 4)
 Balancing an unbalanced TP
Excess Supply
• If total supply exceeds total demand, we can
balance a TP by creating a dummy demand
point.
• Since dummy demand point are not real, they
are assigned a cost of zero.
• These indicate unused supply capacity
Unmet Demand
• If total supply is less than total demand,
actually the problem has no feasible solution.
• To solve the problem it is sometimes
necessary to allow the possibility of leaving
some demand unmet.
• This means that a dummy supply point should
be introduced
Example
• From previous example, Suppose that demand
for city 1 is 40 million kwh. Formulate a
balanced transportation problem.
• Total demand is 120, total supply is 125.
• To balance the problem, we would add a
dummy demand point with a demand of 5
million kwh.
• From each plant, the cost of shipping 1 million
kwh to the dummy will be 0.
 Balanced TP
To

From City 1 City 2 City 3 City4 Dummy Supply

Plant 1 8$ 6$ 10$ 9$ 0 35
Plant 2 9$ 12$ 13$ 7$ 0 50
Plant 3 14$ 9$ 16$ 5$ 0 40
Demand 40 20 30 30 5 125
 Least Cost Methods
• This method finds a better initial basic solution
by concentrating on the cheapest routes than
NWC.
• Row/column corresponding to which 𝑪𝒊𝒋 is min
• Then assign 𝒙𝒊𝒋 its largest possible value to the
cell with smallest unit cost.
• cross out row 𝑖/column 𝑗 and reduce the 𝒔𝒊 or
𝒅𝒊 accordingly.
• Look for un crossed out cells with smallest unit
cost and repeat the process.
• If there are 2 or more min cost then we should
select the row/column corresponding to the
lower numbered column.
• If they appear in the same row we select the
lower numbered column.
Example: Use LCM to get IBFS and its
associated cost of the following TP
 cont..
Sun Ray Transportation Model

Mills
From M1 M2 M3 M4 Supply
15 0 15-15=0
S1 10 2 20 11 0-0=0
15 10 25-15=10
S2 12 7 9 20 10-10=0
5 5 10-5=5
S 3 4 14 16 18 5-5=0
15-0=15
15-5=10
Demand 5-5=0 15-15=0 15-15=0 10-10=0 50
 Cont.
• Therefore, the IBS that has the total lower cost
are 𝑥12 = 15, 𝑥14 = 0, 𝑥23 = 15, 𝑥24 = 10,
𝑥31 = 5, 𝑥34 = 5 (these are like pivots)
• The corresponding cost is
• Z= (1*15)+(0*11)+(9*15)+(10*20)+(5*4)+(5*18)
• Min Z= 460 $
• This min cost of transportation is lower than the
corresponding cost of the solution developed
by the NWCM
Practice for the NWCM and LCM
Use both NWCM and LCM to find the IBS and the
corresponding costs, ultimately comment on the
results.
Transportation Model
d1 d2 d3 d4
2 3 5 6
S1 5
2 1 3 5
S2 10
2 8 4 6
S 3 15
Demand 12 8 4 6
Vogel’s Approximation Method (VAM)
• This is an improved version of the LCM
• This generally produces better solution
o Step 1: Begin by computing for each row (and
column) a “penalty” equal to the difference
between the two smallest costs in the
row/column.
o Step 2: Identify the row/column with the largest
penalty among all rows and columns.
o If the penalties corresponding to two or more columns
or rows are equal we select the topmost row and the
extreme left column the remaining row/column are
assigned zero supply.
 Cont..
• Step 3: Select 𝒙𝒊𝒋 as basic variable if 𝒄𝒊𝒋 is the
minimum cost in the row or column with
largest penalty.
o Choose as the first basic variable 𝒙𝒊𝒋 , the variable
in this row or column that has the smallest
transport cost.
• Step 4: Back to step 2, is now performed in
uncrossed out rows and columns until all basic
variables are satisfied.
Example: Use the VAM to Find the IBS
and it Corresponding Cost
Sun Ray Transportation Model

From M1 M2 M 3 M4 Supply

S1 10 2 20 11 15

S2 12 7 9 20 25

S3 4 14 16 18 10

Demand 5 15 15 15 50
Find row and column penalties
Sun Ray Transportation Model
From M1 M2 M 3 M 4 Supply Penalty
S1 10 2 20 11 15 10-2=8
S2 12 7 9 20 25 9-7=2
5
S3 4 14 16 18 10-5=5 14-4=10
Demand 5-5=0 15 15 15
Penalty 10-4=6 7-2=5 16-9=7 18-11=7
• Because row 3 has the largest penalty =10 and
Cell(3,1) has smallest unit cost.
• Assign 5 units to the cell 𝒙𝟑𝟏
• Column one is satisfied and must be crossed out.
Find new row and column penalties
Sun Ray Transportation Model
From M1 M2 M 3 M 4 Supply Penalty
15
S1 10 2 20 11 15-15=0 11-2=9
S2 12 7 9 20 25 9-7=2
5
S3 4 14 16 18 10-5=5 16-14=2
Demand 5-5=0 15-15=0 15 15
Penalty - 7-2=5 16-9=7 18-11=7
• Row 1 has highest penalty
• Allocate cell 𝒙𝟏𝟐 = 𝟏𝟓
• It has sissified both row 1 and column 2
 Cont.
Sun Ray Transportation Model
From M1 M2 M 3 M4 Supply Penalty
15
S1 10 2 20 11 15-15=0 -
15
S2 12 7 9 20 25-15=10 20-9=11
5
S3 4 14 16 18 10-5=5 18-16=2
Demand 5-5=0 15 15-15=0 15 50
Penalty - - 16-9=7 18-11=7
• Row produced highest penalty =11
• Assign cell 𝒙𝟐𝟑 = 𝟏𝟓 and cross out column 3
• Reduce unit 15 in row 2
 Cont.
Sun Ray Transportation Model
From M1 M2 M 3 M4 Supply Penalty
15 0 15-15=0
S1 10 2 20 11 0-0=0 -
15 10 25-15=10
S2 12 7 9 20 10-10=0 -
5 5 10-5=5
S3 4 14 16 18 5-5=0 -
15-10=5
Demand 5-5=0 15-15=0 15-15=0 5-5=0
Penalty - - - 18-11=7
• On column 4 remained, apply LCM to the column so
assign 𝑥14 = 0, 𝑥34 = 5 𝑎𝑛𝑑 𝑥24 = 10
Identify IBS and corresponding cost
Sun Ray Transportation Model
From M1 M2 M 3 M4 Supply
15 0
S1 10 2 20 11 15
15 10
S2 12 7 9 20 25
5 5
S3 4 14 16 18 10
Demand 5 15 15 15
Therefore, the IBS that has the total lower cost are 𝑥12 = 15,
𝑥14 = 0, 𝑥23 = 15, 𝑥24 = 10, 𝑥31 = 5 and 𝑥34 = 5
The corresponding cost is
Z= (2*15)+(0*11)+(9*15)+(10*20)+(5*4)+(5*18)= 460 $
 Testing for optimization
Modified distribution method (MODI)
• The IBFS obtained from NWCM, LCM and VAM
can be tested for optimality
• Some times is called transportation simplex
methods
•