Chapter 2

Linear Programming: Special

Purpose Algorithm

I. OBJECTIVES

At the end of the lesson, the students should be able to:

 Describe the nature of transportation problem through downloading and studying the file

uploaded by the teacher in EDMODO;

 Solve transportation problems manually and interpret the results;

 Set up transportation problems in the general linear programming format; and

 Appreciate transportation problem through the use of EDMODO.

TRANSPORTATION MODEL

The transportation model is a valuable tool in analyzing and modifying existing transportation

systems or the implementation of new ones. In addition, the model is effective in determining

resource allocation in existing business structures.

METHODS IN ESTABLISHING THE INITIAL AND OPTIMAL FEASIBLE SOLUTION

There are (3) ways to establish the initial solution of a transportation problem namely:

Northwest Corner Rule (NCR) Greedy Method or Minimum Cost Method (MCM) and Vogel’s

approximation Method (VAM).

A. Initial Feasible Solution

1. Northwest Corner Rule (NCR). A procedure for obtaining an initial feasible solution to a

transportation problem that starts with allocating units to the upper left–hand corner and ends in the

lower right corner of any transportation problem.

2. Greedy Method or Minimum Cost Method (MCM). A systematized procedure used to find an

initial feasible solution to a transportation problem; it is easy to use and provides good (but not

optimal) solution. This method allocates as much as possible to the least-cost/highest-profit cell. Ties

may be broken arbitrarily. Rows and columns that have been completely allocated are not

considered, and the process of allocation is continued. The procedure is completed when all row and

column requirements are addressed.

3. Vogel’s approximation Method (VAM). An algorithm that an initial feasible solution to a

transportation problem by considering the “penalty cost” of not using the cheapest available route. It

usually produces an optimum solution in 80% of the sample problems tested.

B. Optimal Feasible Solution

On the other hand, there are two (2) ways to establish the optimal solution of a transportation

problem namely: Stepping Stone Method and Modified Distribution method.

1. Stepping Stone Method (SSM). A procedure for determining if a solution to a transportation

problem is optimal that involves tracing closed paths from each unused square through stone

squares. To develop an optimal solution using SSM in transportation problem involves evaluating

each unused square to determine whether a shift into it is advantageous from, a total-cost/total-profit

stand point. If it is, the shift is made, and the process is repeated. When all squares have been

evaluated and appropriate shifts made, the problem is solved. The term stepping stone appeared in

early descriptions of the method, in which unused squares were referred to as “water” and used cells

as “stones”- from the analogy of walking on a path of stones half-submerged in water.

2. Modified Distribution Method (MODI). A procedure for determining the per-unit cost/profit

change associated with assigning flow to an unused square in the transportation problem. Paths or
loops in this method are determined mathematically. The tableau is modified with U (row) and V

(column) variables. Allocated cell costs C ij=U i+ V j.

INITIAL FEASIBLE SOLUTION

In this section we will deal with the initial solutions for transportation problems particularly the

Northwest Corner Rule (NCR), Greedy Method or Minimum Cost Method (MCM), and Vogel’s

approximation Method (VAM).

Example: The WSS Company sells desktop computers to IT companies in Metro Manila, and ships

them from three distribution warehouses located in three (3) different areas. The company is able to

supply the following numbers of desktop computers to IT companies by the beginning of the year.

Distribution Supply
Warehouse
Warehouse 1 150
Warehouse 2 200
Warehouse 3 50
Total 400

IT companies have ordered desktop computers that must be delivered and installed by the beginning

of the year: IT Company Demand

(desktop computers)
AUS Link 100
SJS Networking Inc. 80
RFS Data Limited 220
Total 400

The shipping costs per desktop computer from each distributor to each company are as follows:
From At
A B C
1 7 5 9
2 10 12 10
3 6 3 14

With cost minimization as a

criterion, WSS Company wants to determine how many desktop computers should be shipped from

each warehouse to each IT company. Find the initial solution using 9a) Northwest Corner Rule, (b)

Solution:

In order to solve a transportation problem it is necessary to follow the following steps.

Step 1: Establish the Mathematical Model for the problem.

 Let 1= Warehouse 1 A= AUS Link

2= Warehouse 2 B= SJS Networking Inc.

3= Warehouse 3 C= RFS Data Limited

Step 2: Establish the transportation tableau.

Destination
14
36
12
10
o To
579
r From A B C Supply
i
g 150
1
i
n 200
2

50
3

Demand 100 80 220 400

7
12
X 2X XA 2 C
X
5 X 13BAB 3 C 9
X 1 CObserve the variable representation in each square.
10
X2 B To
10
From A B C Supply
6
X3 A 150
3 14
1

200
2

50
3

Deman 100 80 220 400

d
Objective Function:

Minimize: C=7 X 1 A +5 X 1 B +9 X 1 C +10 X 2 A +12 X 2 B + 10 X 2C + 6 X 3 A +3 X 3 B+ 14 X 3 C

Constraints:

Subject to: X 1 A + X 1 B + X 1C =150

X 2 A + X 2 B + X 2C =200

X 3 A + X 3 B + X 3 C =50

X 1 A + X 2 A + X 3 A =100

X 1 B+ X 2 B + X 3 B=80

X 1 C + X 2C + X 3 C =220 X ij ≥
A. Northwest Corner Rule Solution:

 Steps in Establishing the Northwest Corner Rule (NCR)

1. Start in the upper left-hand corner (origin 1, destination A) and allocate as many units as possible
to this cell. That is, use as much supply from origin A is possible, to satisfy demand at destination A. This
means that the amount allocated is the minimum of supply at 1 and demand at A.
2. Reduce the available supply at the current origin and unsatisfied demand at the current
destination by the amount of the allocation.
3. Identify the first origin with available supply. This is either the current origin or the one directly
 Step 1: Beginning
37
14
12
1069 5 X 1 A which is the upper left corner of the tableau we allocate 10 units,
in cell
100
exhausting demand in column A and leaving 50 units of supply in row 1.

To
From A B C Supply
50
150
1

200
2

50
3

Demand 100 80 220 400

Step 2: Staying
579
63
10
14
12
10 in a row 1, move to
50
100
cell X 1 B, where supply is now 50 and demand is 80 units. Allocate 50 units to this cell, exhausting the

supply of row 1 To and leaving 30

From A B C Supply
units in column 50 B.
150
1

200
2

50
3
30
Demand 100 80 220 400

Step 3: Staying in column B, move down to cell X 2 B, where supply is 200 units. Allocate 30 units to

this cell, exhausting

To
demand in From A B C Supply column B and

leaving 170 units 150 of supply in row

1
2. 170
200
2
7
10
12
10
5 30
50
100 9 50
3
14
6
Demand 100 80 220 400
Step 4: Staying in row 2, move to cell X 2 C and allocate 170 units, exhausting that row’s supply and

leaving 50 units in column C.

Step 5: Staying in column C, move down to cell X 3 C and allocate 50 units, exhausting both the row

and column quantities.

To
From A B C Supply
7 5 9
50 150
100
1
10 12 10
30 170 200
2
6 3 14
50 50
3

Deman 100 80 220 400

d

The total transportation cost for Northwest Corner Rule

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700
X1 B 50 x 5 250
X2 B 3 x 12 360
X2C 170 x 10 1,700
X3C 50 x 14 700
Total Transportation Cost 3,710

In terms of minimizing or maximizing total transportation cost, this solution may or may not be

optimal. The main drawback of the northwest corner rule method is that it does not consider cell

(route) costs/ profits in making allocation. Consequently, if this allocation is optimal, that can be

attributed to chance rather than to the method used.

B. Greedy Method or Minimum Cost Method (MCM)

 Steps in establishing the Greedy Method or Minimum Cost Method (MCM)

Step 1: Beginning with the lowest cost X 3 B, in which the cost is 3, allocates 50 units. Cross out that

row 3, and we change the column B to 30 units.

To
From A B C Supply
7 5 9
150
1
10 12 10
200
2
6 3 14
50
3 50
30
Demand 100 80 220 400
Step 2: Of the cells that remain, X 1 B has the lowest cost. Allocate 30 units in cell X 1 B, the cells in

column B are crossed out along with the total. The remaining demand for the row 1 is 120 units, so

that column total

14
3
12
1069 must be adjusted accordingly.
57
30
50
To

From A B C Supply
120

150

200

50

Demand 100 80 220 400

Step 3: Of the remaining cells, X 1 A has the lowest cost, then assigned 100 units. Because the column

1 has been satisfied, the cells in the column must be crossed out, and 100 units must be subtracted

from the supply of row 1.

To

From A B C Supply
7 5 9 120
100 30
150

1 20
10 12 10

200

2
6 3 14
50 50

Demand 100 80 220 400

Step 4: Cell X 1 C has the next lowest cost, so it is next in line for allocation. The remaining supply is 20

units, and the remaining demand is 220 units. Consequently, the quantity 20 is placed in X 1 C . This

completes the use of supply for row 1, and leaves a demand of 200 units for column C.

To

From A B C Supply
7 5 9 120
100 30 20
150

1 20
10 12 10

200

2
6 3 14
50 50

3
200
 Demand 100 80 220 400

Step 5: The last remaining cell X 2 C , then, receives a quantity of 200 unuits, canceling the remaining

supply and demand for both its row and column.

To

From A B C Supply
7 5 9 120
100 30 20
150

1 20
10 12 10
200
200

2
6 3 14
50 50

3
200

Demand 100 80 220 400

The total transportation cost for Greedy Method

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700

X1 B 30 x 5 150

X1C 20 x 9 180

X2C 200 x 10 2,000

X3 B 50 x 3 150
Total Transportation Cost 3,180

C. Vogel’s Approximation Method (VAM)

 Steps in Establishing the Vogel’s Approximation Method (VAM)

1. For each row with an available supply and each column with an unfilled demand, calculate an
opportunity/ penalty cost by subtracting the smallest entry from the second smallest entry for a
minimization calculate an opportunity profit by getting the difference between the highest and second
highest and second highest entry.
2. Identify the row or column with largest opportunity/penalty cost for minimization or opportunity/
penalty profit for maximization. (Ties may be broken arbitrarily.)
Opportunity Cost is the cost of the opportunities that are sacrificed in order to take action for

minimization problem. It is also called as penalty cost (opportunity/ penalty profit for maximization

problem.

Step 1: Determine the difference of the lowest and second lowest cost alternatives for each row and

column. Identify the row or column with largest opportunity cost.

Opportunity Cost for the First Allocation

Row/Column 2nd Lowest Cost Lowest Cost Opportunity Cost

1 7 5 2 Largest
difference
2 10 10 0
 3 6 3 3

A 7 6 1

B 5 3 2

C 10 9 1

Step 2: Allocate 50 units in cell X 3 B that contains the lowest cost in row 3 and cross out row 3.

Compute for the next opportunity cost.

To

From A B C Supply
7 5 9

150

1
10 12 10

200

2
6 3 14
50 50

3
30

Demand 100 80 220 400

Step 3: Then identify the difference of the lowest and second lowest cost alternatives for each

remaining row and column. Identify the row or column with largest opportunity cost.

Opportunity Cost for the Second Allocation

Row/Column 2nd Lowest Cost Lowest Cost Opportunity Cost

1 7 5 2 Largest
difference
2 10 10 0
 A 10 7 3

B 12 5 7

C 10 9 1

Step 4: Allocate 30 units in cell X 1 B that contains the lowest cost in column B and cross out column B.

Compute for the next opportunity cost.

To

From A B C Supply
7 5 9 120
30
150

1
10 12 10

200

2
6 3 14
50 50

Demand 100 80 220 400

Step 5: Again select the largest difference between the lowest and second lowest cost alternatives for

each row or column. Identify the row or column with largest opportunity cost.

Opportunity Cost for the Third Allocation

Row/Column 2nd Lowest Cost Lowest Cost Opportunity Cost

1 9 7 2 Largest
difference
2 10 10 0

A 10 7 3

C 10 9 1
Step 6: allocate 100 units in cell X 1 A that contains the lowest cost in column A and cross out column

A. If there are only two unused squares, select the lowest cost and allocate the necessary amount.

To

From A B C Supply
7 5 9 20
100 30
150

1
10 12 10

200

2
6 3 14
50 50

Demand 100 80 220 400

Step 7: In the previous tableau the lowest cost is in X 1 C . Allocate 20 units in X 1 C and then allocate

200 units in X 2 C .

To

From A B C Supply
7 5 9 20
100 30 20
150

1
10 12 10
200
200

2
6 3 14
50 50

Demand 100 80 220 400

The total transportation cost for Vogel’s Approximation Cost Method

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700

X1 B 30 x 5 150

X1C 20 x 9 180

X2C 200 x 10 2,000

X3 B 50 x 3 150
Total Transportation Cost 3,180

In general, the NCR, Greedy Method, and VAM do not guarantee to provide directly an optimal

solution. They simply provide an initial feasible solution. There are some cases when these

procedures provide an optimal solution, we would not know that it is optimal unless it was tested

using improvement index computation in either the SSM or MODI.

Enrichment Exercise
Establish the initial feasible solutions of the LP model using NCR, Greed Method and VAM.

Minimize:
15 X 1 A +24 X 1 B +12 X 1 C +19 X 1 D +10 X 2 A +11 X 2 B +13 X 3C +13 X 2 D +16 X 3 A + 21 X 3 B +11 X 3 C +26 X 3 D

Subject to: X 1 A + X 1 B + X 1C + X 1 D =140

X 2 A + X 2 B + X 2C + X 2 D =160
X 3 A + X 3 B + X 3 C + X 3 D =50
X 1 A + X 2 A + X 3 A =100
X 1 B+ X 2 B + X 3 B=100
X 1 C + X 2C + X 3 C =50
X 1 D + X 2 D + X 3 D=100
X ij ≥0
MINIMIZATION OPTIMAL FEASIBLE SOLUTION

In this section covers the optimal solution set of a minimization transportation model for both

the Stepping Stone Method (SSM) and Modified Distribution Method (MODI).

Example: The WSS Company sells desktop computers to IT companies in Metro Manila, and ships

them from three distribution warehouses located in three(3) different areas. The company is able to

supply the following numbers of desktop computers to IT companies by the beginning of the year:

Distribution Warehouse Supply

Warehouse 1 150

Warehouse 2 200

Warehouse 3 50
Total 400

IT companies have ordered desktop The shipping costs per desktop

computers that must be delivered and computer from each distributor to

installed by the beginning of the year: each company are as follows:

IT Company Demand

(desktop computers)
AUS Link 100
From At
SJS Networking Inc. 80
A B C
1 7 5 9
RFS Data Limited 220
Total 400
2 12 12 10

3 6 3 14

With cost minimization as a criterion, WSS Company wants to determine how many desktop

computers should be shipped from each warehouse to each IT company. Find the minimum optimal
solution using Northwest Corner Rule for the initial solution and the optimal solution using Stepping

Stone Method (SSM) and Modified Distribution (MODI).

Solution:

Stepping Stone Method (SSM)

In order to solve a transportation problem it is necessary to follow the following steps.

Step 1: Establish the Mathematical Model for the problem.

Minimize: C=7 X 1 A +5 X 1 B +9 X 1 C +10 X 2 A +12 X 2 B + 10 X 2C + 6 X 3 A +3 X 3 B+ 14 X 3 C

Subject to: X 1 A + X 1 B + X 1C =150 X 1 A + X 2 A + X 3 A =100

X 2 A + X 2 B + X 2C =200 X 1 B+ X 2 B + X 3 B=80

X 3 A + X 3 B + X 3 C =50 X 1 C + X 2C + X 3 C =220

X ij ≥0

Step 2: Establish the transportation tableau.

Tableau 1

To

From A B C Supply
7 5 9

150

1
10 12 10

200

2
6 3 14

50

Demand 100 80 220 400

Step 3: Set up the initial feasible solution using Northwest Corner Rule (NCR).

Tableau 1

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
30 170
200

2
6 3 14
50
50

Demand 100 80 220 400

Step 4: Compute the stone squares.

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700

X1 B 50 x 5 250

X2 B 30 x 12 360

X2C 170 x 10 1,700

X3C 50 x 14 700
Total Transportation Cost 3,710

Step 5: Compute for closed path and improvement indices for the initial tableau.

Improvement Index is the increase/decrease in a total cost (for minimization problem) and total

profit (for maximization problem) that would result from reallocating one unit to an unused square.
Tableau 1

To

From A B C Supply
7 5 9
100 50 (+)
150

1
10 12 10
30
170 200

2
6 3 14
50
50

Demand 100 80 220 400

Observe the closed path of unused square X 1 C . The X 1 C will contain a positive sign, X 2 C is negative,

X 2 Bis positive and X 1 B is negative.

Tableau 1

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
(+) 30 170
200

2
6 3 14
50
50

Demand 100 80 220 400

For the second unused square X 2 A , X 2 Bwill contain a positive sign, X 1 A is negative, X 1 Bis positive and

X 2 Bis negative.

Tableau 1

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
30 120
200

2
6 3 14
(+) 5
50
0
3

Demand 100 80 220 400

In the third unused square X 3 A , X 3 A will contain a positive sign, X 1 A is negative, X 1 B is positive, X 2 B

Is negative, X 2 C is positive and X 3 C is negative.

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
30 170
200
 2
6 3 14
(+)
50
50

Demand 100 80 220 400

And in our last unused square X 3 B, X 3 B will contain a positive sign, X 2 B is negative, X 2 C is positive,

X 3 C is negative.

We have already established the computation of the improvement index below based from the

previous presentation.

Unused Closed Path Computation of

Square Improvement Indices

X1C + X 1C −X 2 C + X 2 B −X 1 B +9−10+12−5=6

X2 A + X 2 A −X 1 A + X 1 B −X 2 B +10−7+5−12=−4
Lowest
+6−7+5−12+10−14=−12 negative
X3 A + X 3 A − X 1 A + X 1 B −X 2 B + X 2 C −X 3 C

X3 B + X 3 B− X 2 B + X 2 C −X 3 C +3−12+10−14=−13

Step 5: Identify the lowest negative value in the improvement index, which is -13 under cell X 3 B.

Locate the closed path in computing the cell X 3 B that is used as a deduction to cost, which are cells

X 2 Band X 3 C . Choose the lower amount (allocation) between the two cells; X 2 B has 30, while X 3 C has

50, so we will choose 30.

Tableau 1

To

From A B C Supply
7 5 9
100 50
150

1
 10 12 10
30 120
200

2
6 3 14
50
50

Demand 100 80 220 400

Step 6: Allocate the amount of X 2 B to cell X 3 B, and then balance the stone square according to the

total amount needed in each row and column.

Tableau 2

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
200
200

2
6 3 14
30 20
50

Demand 100 80 220 400

Note: Unused squares from the previous tableau will not have any allocation in the
preceding tableau except for the current identified improvement index stone squares.

Step 7: Solve for the stone squares.

Source-Destination Quantity Shipped x Unit Cost Total Cost

 X1 A 100 x 7 700

X1 B 50 x 5 250

X2C 200 x 10 2,000

X3 B 30 x 3 90

X3C 20 x 14 280
Total Transportation Cost 3,320

Step 8: Test the solution for improvement by determining the improvement index as applied in the first

iteration.

Unused Closed Path Computation of

Lowest
Square Improvement Indices negative
X1C + X 1C −X 3 C + X 3 B−X 1 B +9−14+ 3−5=−7

X2 A + X 2 A −X 1 A + X 1 B −X 3 B + X 3 C −X 2 B +10−7+5−3+14−10=9

X2 B + X 2 B− X 3 B + X 3 C −X 2 C +12−3+14−10=13

X3 A + X 3 A − X 1 A + X 1 B −X 3 B +6−7+5−3=1

Since the improvement index still has a negative value, we will continue the computation.

Step 9: The lowest negative value in the improvement index is -7 under cell X 1 C . Locate the closed

path in computing the cell X 1 C that is used as a deduction to cost, which are cells X 1 B and X 3 C .

Choose the lower amount between the two cells; X 1 B has 50, while X 3 C has 20, so we will choose 20.

Step 10: Allocate the amount of X 3 C to cell X 1 C , and then balance the stone square according to the

total amount needed in each row and column.

Tableau 3

To

From A B C Supply
7 5 9
100 30 20
150

1
10 12 10
200
200

2
6 3 14
50
 50

Demand 100 80 220 400

Step 11: Solve for the stone squares.

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700

X1 B 30 x 5 150

X1C 20 x 9 180

X2C 200 x 10 2,000

X3 B 50 x 3 150
Total Transportation Cost 3,180

Step 12: Test the solution for improvement by determining the improvement index.

Unused Closed Path Computation of

Square Improvement Indices

X2 A + X 1 A −X 1 A + X 1 C −X 2C +10−7+9−10=2

X2 B + X 2 B− X 1 B + X 1 C − X 2 C +12−5+ 9−10=6

X3 A + X 3 A − X 1 A + X 1 B −X 3 B +6−7+5−3=1

X3C + X 3C −X 1 C + X 1 B −X 3 B +14−9+5−3=7

Since the improvement index does not contain any negative value the solution is optimum.

The optimal solution and the maximum profit are

X 1 A =100 X 1 B =30 X 1 C =20 X 2 C =200 X 3 B=5

C= 7 ( 100 ) +5 ( 3 ) +9 ( 20 ) +10 ( 0 ) +12 ( 0 ) +10 ( 200 ) +6 ( 0 ) =3 (50 )+14 (0)

= 700+150+180+ 0+0+2,000+0+ 150+0

= 3,180

Modified Distribution (MODI)

 Using the previous solution on the initial feasible solution provided in the Stepping Stone

Method.

Tableau 1 V1 V2 V3

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
30 170
200

2
6 3 14
50
50

Demand 100 80 220 400

Step 1: Compute the stone squares.

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700

X1 B 50 x 5 250

X2 B 30 x 12 360

X2C 170 x 10 1,700

X3C 50 x 14 700
Total Transportation Cost 3,710

Step 2: Represent a new variable for each row and column.

Let U i=¿a value assigned to row i,

V j=¿ a value assigned to column j, and

C ij=¿ a value to cost in square ij.

 Using the formula of cost of the stone square ij.

C ij=U i+ V j

Step 3: Substitute the appropriate value for each C ij , let U i be equal to zero or U i=0 and compute for

the unknown.

X 1 A > U 1+ V 1=7 X 2 B >U 2 +V 2=12 X 3 C >U 3 +V 3 =14

X 1 B >U 1 +V 2=5 X 2 C >U 2 +V 3 =10

U 1 +V 1=7 U 1 +V 2=5 U 2 +V 2=12 U 2 +V 3=10 U 3 +V 3=14

0+V 1 =7 0+V 2 =7 U 2 +5=12 7+V 3 =10 U 3 +3=14

V 1=7 V 2=5 U 2=12−5 V 3=10−7 U 3=14−3

U 2=7 V 3=3 U 3=11

Note: Improvement index solution established in the Modified Distribution

Method is the same with the Stepping Stone Method.

Step 4: Evaluate the stone square in the initial solution by finding the improvement index.

Unused C ij−U i−V j Improvement Indices

Square
Lowest
X1C X 1 C −U 1−V 3 9−0−3=6 negative

X2 A X 2 A −U 2−V 1 10−7−7=−4
 X3 A X 3 A −U 3 −V 1 6−11−7=−12

X 3 BB X 3 B−U 3 −V 2 3−11−5=−13

Step 5: Identify the lowest negative value in the improvement index, which is equal to -13. Find the

closed path into X 3 B just like what we did in the previous method. Allocate 30 the amount of X 2 B , then

balance the stone square according to the total amount needed in each row and column.

Tableau 2
V1 V2 V3

To

From A B C Supply
7 5 9
100 50
150

1
10 12 10
200
200

2
6 3 14
30 20
50

Demand 100 80 220 400

Step 6: Solve for the stone squares.

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1 A 100 x 7 700

X1 B 50 x 5 250

X2C 200 x 10 2,000

 X3 B 30 x 3 90

X3C 20 x 14 028
Total Transportation Cost 3,320
Step 7: Substitute the appropriate value for each C ij, let U i be equal to zero or U i=0 and compute for

the unknown.

X 1 A > U 1 +V 1=7 X 2 C > U 2 +V 3=10 X 3 C > U 3 +V 3=14

X 1 B > U 1 +V 2=5 X 3 B > U 3 +V 2=3

U1 + V1 = 7 U1 + V2 = 5 U3 + V2 = 3 U3 + V3 = 14 U1 + V3 = 10

0+V1 = 7 0 + V2 = 5 U3 + 5 = 3 -2 + V3 = 14 U2 + 16 = 10

V1 = 7 V2 = 5 U3 = 3 – 5 V3 = 14 + 2 U2 = 10 - 16

V3 = 16 U2 = -6

Step 8: Evaluate the stone square in the initial solution by finding the improvement index.

Improvement
Unused Square C1j - U1 – Vj
Indices
X1C X1C - U1 – V3 9 -0 – 16 = -7 Lowest
Step 9: Identify the lowest negative
negative
X2A X2A – U2 – V1 10 - (-6) – 7 = 9
value in the improvement index
X2B X2B – U2 - V2 12 – (-6) – 5 =13
which is equal to -7. Allocate the
X3A X3A – U2 – V1 6 – (-2) – 7 = 1
amount 20 units to X1c from

which X3c which contains the

lesser amount than X1B. Then, balance the stone square according to the total amount needed in

each row and column.

Tableau 3 V1 V2 V3

To
 From A B C Supply
7 5 9
100 30 20
150

1
10 12 10
200
200

2
6 3 14
50
50

Demand 100 80 220 400

Step 10: Solve for the stone squares.

Source-Destination Quantity Shipped x Unit Cost Total Cost

X1A 100 x 7 700

X1B 30 x 5 150

X1C 20 x 9 180

X3C 200 x 10 2,000

X3B 50 x 3 150
Total Transportation Cost 3, 180

Step 11: Substitute the appropriate value for each Cs, let Us be equal to zero or U1 = 0 and compute

for the unknown.

X1A > U1 + V1 = 7 X2C U1 + V3 = 9 X3B > U3 + V2 = 3

X1B > U1 + V2 = 5 X2C > U2 + V3 = 10

U1 + V1 = 7 U1 + V2 = 5 U1 + V3 = 9 U2 + V3 = 10 U3 + V2 = 3

0 + V1 = 7 U1 + V2 =5 0 + V3 = 9 U2 + 9 = 10 U3 + 5 = 3
 V1 = 7 V2 =5 V3 = 9 U2 = 10 – 9 U3 = 3 - 5

U2 = 1 U3 = -2

Step 12: Evaluate the stone square in the initial solution by finding the improvement index.

Unused Square Cs - U1- V1 Improvement Indices

X2A X2A – U2 – V1 10 – 1 – 7 = 2

X2B X2B – U2 – V2 12 – 1 – 5 = 6

X3A X3A – U3 – V1 6 – (-2) – 7 = 1

X3C X3C – U3 – V3 14 – (-2) – 9 = 7

The optimal solution and the maximum profit are

X1A = 100 X1B = 30 X1C = 20 X2C = 200 X3B = 50

C = 7(100) + 5(30) + 9(20) + 10(0) + 12(0) + 10(200) + 6(0) + 3(50) + 14(0)

= 700 + 150 + 180 + 0 + 0 + 2,000 + 0 + 150 + 0

= 3, 180

C. Points to Remember in a Transformation Problem

Only unused square are assigned as part of improvement index computation.

The unused square in a closed path is always the first in the closed path pattern.

Each closed path contains an even number of squares.

After a pattern change has been made, at least some of the used stone squares will have changed,

and all of the stone squares allocation must be re-examined.

The direction around a closed path may be either clockwise or counterclockwise (closed path

Enrichment Exercise 5.4

Establish the optimal feasible solutions of the LP model using SSM and MODI.

Minimize: C = 15X1A + 24X1B + 12X1C + 19X1D + 10X2A + 11X2B + 13X2C + 13X2D

+ 16X3A + 21X3B + 11X3C + 26X3D

Subject to: X1A + X1B + X1C + X1D = 140 X1A + X2A + X3A = 100 Xh > 0

X2A + X2B + X2C + X2D = 160 X1B + X2B + X3B = 100

X3A + X3B + X3C + X3D = 50 X1C + X2C + X3C = 50

X1D + X2D + X3D = 100

pattern).

Maximization Optimal Feasible Solution

In this section we will discussed the optimal solution set of a maximization transportation

model for both of the Stepping Stone Method (SSM) and Modified Distribution Method (MODI).

Example: The GMS Inc. has one-year contract to supply motors for all refrigerators produced by the

SSS Inc. SSS Inc. which manufactures refrigerators at three locations around the metropolis: Makati,

Manila, and Mandaluyong. Plans call for the following number of refrigerators to be the produced at

each location.

Location Supply
Legarda 35

Sta. Cruz 20

Caloocan 60
Total 115
GMS Inc. has three plants that are capable of producing the motors. The plants and production

capacities are:

Location Supply
Makati 40

Manila 45

Mandaluyong 30
Total 115

Because of the varying production and transportation costs, the profit that GMS earns on each motor

depends on which plant produced the motor and which destination it was shipped to. The following

table gives the accounting department estimates of the profit per unit.

Shipped to
Produced at
Makati Manila Mandaluyong
Legarda 700 1,000 900

Sta. Cruz 1,200 500 400

Caloocan 800 300 1,100

With profit maximization as a criterion, GMS Inc. wants to determine how many motors should

be produced from each plant and how many motors should be shipped from each plant to each

destination. Find the initial solution using Northwest Corner Rule (NCR). Determine the optimal

solution using Stepping Stone Method (SSM) and Modified Distribution (MODI).

Solution:

Stepping Stone Method (SSM) Solution:

In order to solve a transportation problem it is necessary to follow the following steps.

Step 1: Establish the Mathematical Model for the problem.

Let 1 = Legarda A = Makati

2 = Sta. Cruz B = Manila

3 = Caloocan C = Mandaluyong
 Notice that the profit values per unit are all multiplies of 100, thus, we will factor out 100 for

computational convenience.

Maximize: P = 7X1A + 10X1B + 9X1C + 12X2A + 5X2B + 4X2C + 8X3A + 3X3B + 11X3C

Subject to: X1A + X1B + X1C = 35

X2A + X2B + X2C = 20

X3A + X3B + X3C = 60

X1A + X2A + X3A = 40

X1B + X2B + X3B = 45

X1C + X2C + X3C = 30

Xn > 0

Step 2: Set up the initial feasible solution using Northwest Corner Rule (NCR).

Tableau 1

Suppl
To A B C
y
From
7
1 35
35 10 9
5

2 12 15 4 20

5
5 3

3 30 11 60

30
Deman
40 45 30 115
d

Step 3: Compute the stone squares.

Source-Destination Quantity shipped x Unit Cost Total Cost

 X1A 35 x 700 24, 500

X2A 5 x 1,300 6,000

X2B 15 x 500 7, 500

X2C 30 x 300 9,000

X3C 30 x 1,100 33,000

Total Transportation Cost 80,000

Step 4: Compute for closed path and improvement indices for the initial tableau.

Unused Computation of Highest

Closed Path positive
Square Improvement Indices
+X1B – X1A+ X2A– X2B
X1B +10 – 7+12–5 = 10
+X1C – X3C+ X3B– X2B+X2A
X1C +9-11+3-5+12–7 =1
-X1A
X2C +4-11+3-5=-9
+X2C – X3C+ X3B– X2B
X3A +8-12+5-3=-2
+X3A – X2A+ X2B– X3B

Step 5: Identify the highest positive value in the improvement index, which is 10 under cell X1B.

Locate the closed path in computing the cell X1B that is used as addition to the profit, which are cell

X1A and X2B. Choose the lower amount between the two cells; X1A has 35, while X2B has 15, so we

will choose 15.

Tableau 2

Suppl
To A B C
y
From
7
1 35
35 10 9
2 5 20

12 15 4
 5
5 3

3 30 11 60

30
Demand 40 45 30 115

Step 6: Allocate 15 for the amount of X2B to cell X1B, and the balance the stone square according to

the total amount needed in each row and column.

We represent the profit per unit by hundreds for computational convenience.

Step 7: Solve for stone squares.

Source-Destination Quantity shipped x Unit Cost Total Cost

X1A 20 x 700 14, 000

X1B 15 x 1,000 15,000

X2A 20 x 1,200 24, 500

X3B 30 x 300 9,000

X3C 30 x 1,100 33,000

Total Transportation Cost 95,000

Step 8: Test the solution for improvement by determining the improvement index.

Unused Computation of Improvement

Closed Path
Square Indices
X1C + X1C- X3C+ X3B- X1B +9 – 11 + 3 – 10 = 9
X2B + X2B- X1B+ X1A- X2A +5 – 10 + 7 – 12 = -10
Highest
+ X2C- X3C+ X3B- X1B+ X1A-
positive
X2C +4 – 11 +3 – 10 + 7 – 12 = -19
X2A Since the
X3A + X3A- X1A+ X1B- X3B +8 – 7 +10 – 3 =8
improvement

index still has a positive value we will continue the computation.

Step 9: The highest positive value in the improvement index is 8 under cell X3A. Locate the closed

path in computing the cell X3A that is used as addition to the profit, which are cells X1A and X3B.

Choose the lower amount between the two cells; X1A has 20, while X3B has 30, so we will

choose 20.
Step 10: Allocate the amount of X1A to cell X3A, and then balance the stone square according to

the total amount needed in each row and column.

Tableau 3

To

From A B C Supply
7 10 9
35
35

1
12 5 4
20
20

2
8 3 11
20 10 30 4
60

Demand 40 45 30 115

Step 11: Solve for the stone squares.

Source-Destination Quantity shipped x Unit Cost Total Cost

X1A 35 x 1,000 35, 000

X2A 20 x 1,200 24,000

X3A 20 x 800 16,000

X3B 10 x 300 3,000

X3C 30 x 1,100 33,000

Total Transportation Cost 111,000
Step 12: Test the solution for improvement by determining the improvement index.

Unused Computation of Improvement

Closed Path
Square Indices
X1A + X1A- X3A+ X3B- X1B 7-8+3-10=8
X1C + X1C- X3C+ X3B- X1B +9-11+3-10=-9
X2B + X2B- X3B+ X3A- X2A +5-3+8-12=-2
X2C + X2C- X3C+ X3A- X2A +4-11+8-12=-11
Since the

improvement index does not contain any positive value the solution is optimum.

The optimal solution and the maximum profit are

 X1B=35 X2A=20 X3A=20 X3B=10 X3C=30

P=700(0)+1,000(35)+900(0)+1,200(20)+500(0)+400(0)+800(20)+300(10)+1,100(30)

=0+35,000+0+24,000+0+0+16,000=3,000+33,000

=111,000

Steps in Establishing the Stepping Stone Method

1. Use the current solution to create a unique stepping-stone path for each unused square. Use
these paths to calculate the marginal cost of introducing each unused square into the solution.
2. A. Minimization Problem. If all the marginal cost is greater than or equal to zero, stop; you have
the optimal solution. If not, select the cell with the most negative marginal cost. (Ties may be
broken arbitrarily).
3. Using the stepping-stone path, determine the maximum number of items that can be allocated to
the route selected in step 2 and adjust the allocation appropriately.
4. Return to step 1.