Professional Documents
Culture Documents
Industrial Engineering (IE) : Dr. Khallel Ibrahim Mahmoud
Industrial Engineering (IE) : Dr. Khallel Ibrahim Mahmoud
N
o(1)
University of Technology
Electro-mechanical Engineering Dept.
Introduction
Production and
Productivity
Break Even Analysis(B.E)
2011
Dr. Khallel
Introduction:
1.1 Concept of Industrial Engineering (IE)
Industrial Engineering is concerned with the design,
Improvement, and installation of integrated system of men,
material, and machines for the benefit of mankind .It draws
upon specialized knowledge and skills in the mathematical
and physical sciences together with the principles and
methods of engineering analysis and design to specify,
predict and evaluate the results to be obtained from such
systems.
1.2 IE Objectives
The basic objectives of Industrial Engineering are:
1- Improving operating methods and controlling costs.
2- Reducing these costs through cost reduction programs.
The aim of IE department is to provide specialized services
to production departments, such as methods improvement,
time study, Job evaluation and merit rating and to head new
projects if required.
1.3 IE Activities
Basic activities of industrial engineering as stated to
American Institute of industrial engineering as follows:
1- Processes (and methods) selection.
2- Selection and design of tools and equipment.
2
Dr. Khallel
Dr. Khallel
Dr. Khallel
2- Productivity
The standard of living of industrialized nations depends upon
the economic efficiency of all its industrial enterprise great or
small.
Dr. Khallel
output
input
+
-
+
c
c
-
++%
+%
-%
--%
Dr. Khallel
output
input
system
Dr. Khallel
Job type
Batch production
Variet
y
Mass
Production
Low
Low
Production volume
High
After this brief introduction to types of production system and
productivity let us discuss how the productivity measured
and improved.
8
Dr. Khallel
3- Total productivity:
Man power
material
Dr. Khallel
P
Machines
5- Factors affecting productivity:
1. Raw material, its nature and quality.
2. Utilization of manpower.
3. Utilization of plant, equipment and machinery.
4. Basic nature of manufacturing processes employed.
5. Efficiency of plant.
6. Volume, capacity, and uniformity of production.
6- The ways in which the productivity can be increased
summarized as under:
1) Increase manpower effectiveness at all levels.
2) Method improvement.
3) Improve basic production processes by research and
development.
4) Use better production equipment.
5) Improve / simplify product design and reduce variety.
6) Better production planning and control.
1) Technology based
a. CAD/CAM/CIMS
b. Robotics.
c. Laser technology.
d. Modern maintenance technology.
e. Energy technology.
f. Flexible manufacturing system (FMS).
2) Employee based
a. Incentives
b. Promotion
c. Job design
d. Quality circle
3) Material based
a. Material planning and control
b. waste elimination
c. Recycling and reuse of waste materials
d. Purchasing Logistics
4) Process based
a. Method engineering and work simplification
b. Process design
11
Dr. Khallel
Dr. Khallel
Questions:
1. Write short notes on:
a) Techniques of Industrial engineering
(Ans. 1.5)
(Ans. 2.4)
12
Dr. Khallel
Cost
Profit
Cost
curve
B.E.
P
Total
v.cos
Total
cost
Loss
F.cos
t
Production
volume
Dr. Khallel
Dr. Khallel
P= R-Tc
Quantity increase in price by making:
- Better product.
- Advertisement.
- New product limited by market price.
Increase in planned quantity: Increase share of market
and increase those products with high profit or increase
share of market with the increase price according to
quantity of market.
15
Lec.N
o(2)
Industrial Engineering
(IE)
Dr. Khallel Ibrahim Mahmoud
University of Technology
Electro-mechanical Engineering Dept.
Linear Programming
Model
2011
Linear Programming
LP is a mathematical modeling technique designed to optimize the usage of limited
resources, such as available materials, labour and machine time.
The LP model includes three basic elements:
- Decision variables that we seek to determine.
- Objective (goal) that we aim to optimize.
- Constraints that we need to satisfy.
Steps in formulating LP problems:
12345-
II
2
X1, X2 are the decision variables, when we know their values the problem will be
solved.
3- The objective function for the LP is:
Maximize Z = 8X1+6X2
This means the profit Z depend on how many units (X1, X2) are manufactured. Z=
c1x1+ c2x2 where c1, c2 are the respective profits for each type of product. c1=8
c2=6 and Z = 8X1+6X2 and we should select values of the decision variables
X1, X2 that result in the maximum value of Z.
4- There are two constraints :
Maximum production capacity for Dep.I 60 hours
Maximum production capacity for Dep.II 48 hours
( Note: because all the constraints in this problem are maximum capacity , all
constraints are the type)
Now we have to write the coefficient for each decision variable in each constraint.
The two constraints can be expressed as:
4X1+2X2 60
constraint Dep.I
2X1+4X2 48
constraint Dep.II
Consider the first constraint ( Dep.I) what is the coefficient of X1 in this constraint?
It is the processing time (Hours) required per unit of X1. In other word, it is the
processing time used in manufacturing each unit, first product, or 2 hours.
Similarly, the coefficient of X2 in this first constraint is 1 hour.
Therefore the two constraints can be expressed as :
4X1+2X2 60
2X1+4X2 48
And the no. negatively restriction is all Xj 0 ,
X1, X2 0
Graphical LP solution:
Steps in the graphical method
1- Formulate the objective and constraint functions.
2- Draw a graph with one variable on the horizontal axis and one on the
vertical axis.
3- Plot each of the constraints as if they were lines or equlities.
4- Outline the feasible solution space.
5- Circle the potential solution points .These are the intersections of the
constraints or axes on the inner (minimization) or outer (maximization)
perimeter of the feasible solution space.
6- Substitute each of the potential solution point values of the two decision
variables into the objective function and solve for Z.
7- Select the solution point that optimizes Z.
Solution of maximization model
To demonstrate the steps of the graphical solution of a maximization problem we
use the previous example:
Z = 8X1+6X2
Subject to:
4X1+2X2 60
2X1+4X2 48
X1 0
X2 0
Solution:
1- Plot the constraints (shown in the following figure) change constraints to
equalities:
4X1+2X2 60
4X1+2X2 = 60
2X1+4X2 48
2X1+4X2 = 48
For each constraint ( set X1=0 and solve for X2 , then set X2=0 and solve for X1)
the graph the constraint as if it were an equality
4X1+2X2 = 60
X1=15 X2=0
X1=0
2X1+4X2 = 48
X2=30
X1=24 X2=0
X1=0
X2=12
2- Outline the feasible solution space the values of X1 and X2 at points M,A,B
and C are four potential solutions to problem.
Note: point B can be determined as follow:
4X1+2X2 = 60
(2X) 2X1+4X2 = 48
4X2+2X2 = 60
4X2+8X2 = 96
6X2=36
X2=36/6 = 6
Then: 4X1+2X2 = 60
4X1+2(6) = 60
4X1 = 48
X1= 48/4 = 12
Points
M
X1
0
X2
0
Z
Z=8(0)+6(0)=0
A
15
0
Z=8(15)+6(0)=120
B
12
16
Z=8(12)+6(6)=132
C
0
12
Z=8(0)+6(12)=72
To maximize Z, the optimal solution is point B, where X1=12 and X2=6 and
Z=132 profit.
X
4X1+2X2 60
C
B
2X1+4X2 48
Feasible
solution space
X1
M
A
Minimization case:
Consider the following LP problem:
Min Z = 3X1+8X2
Subject to:
X1 80
X2 60
X1+ X2=200
X1, X2 0
Solved problem: X1=80 X2=60
X1=0
X2=200
X1=200
X1=0
B/ X1+ X2=200
X1=80
Then X2=200-80
=120
Point
A
B
X2
200
120
X1
0
80
Z
3(0)+8(200)=1600
3(80)+8(120)=1200
Min Z=1200
X1=80
X2=120
X
20
X1+ X2=200
16
X180
B
12
80
X260
40
X
40
80
12
16
20
Special cases in LP
Five special cases and difficulties arise at times when using the graphical approach
to solve LP problems:
1)
8
Region satisfying
3rd constraint
6
Region satisfying
first 2 constraints
2
X
2
As seen in the figure there is no feasible solution region for this problem because
of the presence of conflicting constraints.
2)
Unboundedness:
Sometimes a linear program will not have a finite solution. This means that in a
maximization problem, for example, one or more solution variables, and the profit,
can be made infinitely large without violating any constraints. If we try to solve
such a problem graphically, we will note that the feasible region is open-ended.
Let us consider a simple example to illustrate the situation.
Z = 3X1+5X2
Subject to: X1 5
X2 10
X1+2 X210
X1, X2 0
X1 5
15
X210
10
Feasible Region
5
X1+2 X210
X
5
10
15
20
As you see, because this is a maximization problem and the feasible region extends
infinitely the right, there is unboundedness or unbounded solution.
3)
X1, X2 0
X
35
30
2X1+ X230
25
20
Redundan
t
15
10
X125
Feasibl
e
X1+ X220
X
5
10
15
20
25
30
The third constraint, X125 is redundant and unnecessary in the formulation and
solution of the problem because it has no effect on the feasible region set.
4) Alternate Optimal Solutions:
An LP problem may on occasion, have two or more alternate optimal solutions.
Graphically, this is the case when the objective functions isoprofit or isocost
line runs perfectly parallel to one of the problems constraints.
X
6
5
4
Isoprofit line for
12/line segment AB
3
B
2
1
X
1
Isoprofit
line for 8
As you see any point along the line between A and B provides an optimal X1
and X2 combination Z=12
5) Degenercy:
Degenercy is a condition that arises when one of the decision variable equal zero.
Look to the following example:
Max Z = 3X1+9X2
Subject to: X1+4X2 8
X1+2X2 4
X1, X2 0
Solution: X1=8
X2=2
X1=4
X2=2
Then: at point A
X1=0
X2=2
Max Z=18
5
4
3
A
2
1
Problems:
B
1
Q1/ Montana wood products manufacturers two high-quality products, chairs and
bookshelf units. Its profit is $15 per chair and $21 per bookshelf unit. Next weeks
production will be constrained by two limited resources, labor and wood. The labor
available next week is expected to be 920 labor hours, and the amount of wood
available is expected to be 2400 board feet. Each chair requires 4 labor hours and 8
board feet of wood. Each bookshelf unit requires 3 labor hours and 12 board feet of
wood. Management would like to produce at least 100 units of each product.
a-
Line I
Line II
Colour
Standard
Economy
3
1
2
1
1
6
The daily running costs for two lines average 6000 for line I and 4000 for line II.
It is given that the company must produce at least 24 colour, 16 standard, and 48
economy TV sets for which an order is pending.
You are required to formulate the above problem as (LP) taking the objective
function as minimization of total cost. Also determine the number of days that the
two lines should be run to meet the requirements.
Q6: Suppose two types of television sets are produced with a profit of 6 units from
each television of type II. In addition 2 and 3 units of raw materials are needed to
produce one television of type I and II respectively. And 4 and 2 units of time are
required to produce one television of type I and II respectively. If 100 units of raw
materials and 120 units of time are available. How many units of each type of
television should be produced to maximize profit?
Q7: A wood product firm uses available time at the end of each week to make
goods for stocks. Currently two products on the list of items are produced for
stock: a chopping board and a knife holder. Both items require three operations:
cutting, gluing, and finishing.
The manager of the firm has collected the following data on these products:
Item
Chopping
board
Knife holder
0.8
13
Finishing
12
3
The manager has also determine that during each week, 56 minutes are available
for cutting, 650 minutes are available for gluing and 360 minutes are available for
finishing.
ab-
X2100
X1, X2 0
X1=0 X2=306.6
X2=0
X1=230 .. (1)
X1=0
X2=200
X2=0
X1=300 (2)
Point
A
B
C
X2
100
133.3
100
X1
100
100
150
Z
3600
4299.3
4350
X
350
300
250
20
15
B
100
50
X
50
X235
10
15
20
25
30
X1=0 X2=306.6
X2=0
X1=30
Point
M
A
B
C
D
B: 2X1+X2= 60
X2=35
X1
0
0
12.5
25
25
X1= (60-35)/2 = 12.5
X2
0
35
35
10
0
Z
0
1400
2150
1900
1500
60
X125
2X1+X260
50
40
A
X235
30
20
10
D
M
10
-X1+X2 5
15
20
25
30
X1, X2 0
X1=-5 X2=5
X2
0
5
11
10
0
Point
X1
M
0
A
0
B
6
C
8
D
13
Hence, Z is Maximum (230) at C.
Z
0
75
225
230
130
X
28
2X1+X226
24
-X1+X2 5
20
16
B
12
2X1+4X2 56
A
4
D
-8
-4
12
16
18
20
24
28
X1, X2 0
Point
A
B
C
X2
12
10
6
X1
2
4
6
Z
52
48
36
18
16
14
12
3X1+2X230
A
B
10
8
6
X1+X214
2X1+X218
X
8
10
12
14
16
18
Production Requirement
2X1+6X248
X1, X2 0
Point
A
B
C
D
X1
0
4
12
24
X2
24
12
4
0
Z
96000
72000
88000
144000
Thus, to Minimize cost, line I should be run for 4 days and line II for 12 days.
X
28
24
20
3X1+X224
16
12
8
X1+X216
C
2X1+6X248
D
4
12
16
20
24
80
70
60
50
40
30
C
20
10
D
30
40
50
60
70
80
holder= X2
Max = 2X1+6X2
Subject to: 1.4X1+0.8X256 (1)
5X1+13X2 650 . (2)
12X1+3X2 360 . (3)
a) The solution at point A
X1=0 X2=50
Z= 2(0) +6(50) = 300
140
120
10
80
60
A
40
20
C
X
20
60
80
10
X2=1.6
12
14
16
Lec.N
o(3)
Dr. Khallel
Industrial Engineering
(IE)
Dr. Khallel Ibrahim Mahmoud
University of Technology
Electro-mechanical Engineering Dept.
Linear Programming
Model
Simplex Method
2011
Dr. Khallel
Simplex Method
The linear programming situation with two decision variables
is easy tackled by a graphical method of solution. However
most practical linear programming problem contain more
than two decision variables. Graphic representation is
difficult (three dimensional) and requires more time for the
determination of an optimal solution. The solution method in
such cases is the simplex method. The simplex method is an
iterative procedure that consists of moving from one basic
feasible solution to another in such a way that the value of
the objective function does not decrease (in the
maximization problem). This process continue until an
optimal solution is reached, if one exist.
The steps required to solve the problem maximizing
are:
Step 1: Define the problem in standard form. This indicates
the construction of a linear programming model.
Step 2: convert the inequalities into equation by insert slack
variables.
Step 3: Construct a matrix of the coefficients of these
equations.
Step 4: This step amends the previous feasible solution so
as to improve the profit.
The simplex method will be illustrated with the following
example.
Dr. Khallel
Dr. Khallel
S1
S2
Zj
Cj-Zj
40
80
0
Coefficient of OF
variables
(Cj)
6
X1
2
8
0
6
X1 = 0
Initial solution:
X2 = 0
S1 = 40
S2 = 80
Z=0
Zj:
Zj (q) = 40(0) + 80(0) = 0
Zj (X1) = 2(0) + 8(0) = 0
Zj (X2) = 5(0) + 4(0) = 0
Zj (S1) = 1(0) + 0(0) = 0
Zj (S2) = 0(0) + 1(0) = 0
Cj Zj :
8
X2
5
4
0
8
0
S1
1
0
0
0
0
S2
0
1
0
0
(Cj Zj) S2 = 0
Dr. Khallel
Dr. Khallel
(A)
(q)
X2
8
X2
1
S2
48
Zj
64
Cj-Zj
6
X1
0
S1
0
S2
0
Observe that the new tableau yields the new basic solution
(X2 =8 , S2= 48 ) with the new value of Z = 64 . An
examination of the last tableau shows that it is not optimal
solution because the variable X1 has a positive coefficient in
the (Cj-Zj) row ( )
Dr. Khallel
= 1, 0 ,
new X2 row : 8- ( X
)= 5
- ( X 1) = 0
1- ( X0 ) = 1
-( X
)=
(A)
(q)
X2
S2
Zj
Cj-Zj
85
6
X1
0
8
X2
1
0
S1
0
S2
Dr. Khallel
Optimum value
15/2 = 7.5
5
85
Dr. Khallel
we get
S1= 80
S2 =- 60
A1 , A2 Artificial variables
Dr. Khallel
X2+ 0S1-S2+A1+0A2= 60
X1+ X2+0S1+0S2+0A1+A2= 200
Min Z = 3X1+ 8X2+0S1+0S2+MA1+MA2
( Cj)
(A)
(q)
0
M
S2
A1
A2
Z
CjZj
80
60
200
260M
3
8
X1
X2
1
0
0
1
1
1
M
2M
3-M 8-2M
0
S1
1
0
0
0
0
0
S2
0
-1
0
-M
M
M
A1
0
1
0
M
0
M
A2
0
0
1
M
0
60/1 = 60 ,
Pivot = 1
A1
200/1 = 200
X1
Dr. Khallel
0-(1X-1) = 1
0-(1X1) = -1
1-(1X0) = 1
( Cj)
(A)
(q)
0
8
M
S1
X2
A2
Zj
80
60
140
480+
140M
CjZj
3
X1
1
0
1
M
8
X2
0
1
0
8
0
S1
1
0
0
0
0
M
M
S2 A1 A2
0
0
0
-1
1
0
1
-1
1
M-8 8-M M
3-M
8-M 2M8
Dr. Khallel
(A)
(q)
3
8
M
X1
X2
A2
Zj
80
60
60
720+
60M
CjZj
3
X1
1
0
0
3
8
X2
0
1
0
8
0
0
M
M
S1
S2 A1 A2
1
0
0
0
0
-1
1
0
-1
1
-1
1
3-M M-8 8-M M
60/-1
60/1 = 60
A2
S2: 60 , 0 , 0 , -1 , 1 , -1 , 1
Row X1 : will be remain
Row X2:60- (-1X60) = 120
0- (-1X0) = 0
1- ( -1X0) = 1
0- (-1X-1) = -1
-1 (-1X1) = 0
1- (-1X-1) = 0
S1
Dr. Khallel
0 (-1X1) = 1
( Cj)
(A)
(q)
3
8
0
X1
X2
S2
Zj
CjZj
80
120
60
1200
3
X1
1
0
0
3
0
8
X2
0
1
0
8
0
0
S1
1
-1
-1
-5
5
0
S2
0
0
1
0
0
M
A1
0
0
-1
0
M
X2 = 120
, S2 = 60
, Z= 1200
M
A2
0
1
1
8
M8
Lec.N
o(4)
University of Technology
Electro-mechanical Engineering Dept.
2011
X2 0
Xn 0
y2 0
ym 0
the variables and constraints of the dual problem can be constructed symmetrically
from the primal problem as follows:
1- A dual variable is defined for each of the m primal constraint equations.
2- A dual constraint is defined for each primal of the n primal variables.
3- The left-hand-side coefficients of the dual constraint equal the constraint
(column) coefficient of the associated primal variable. Its right hand side
equals the objective coefficient of the same primal variable.
4- The objective coefficient of the dual equal the right- hand side of the primal
constraint equations.
The following examples demonstrate the implementation of these rules.
Ex/ Consider the primal model:
Max Z =6X1 + 8X2
St.to: 2X1+5X240
8X1+4X280
X1 , X2 0
The dual of this problem as:
Min = 40Y1+80Y2
St. to: 2Y1+8Y26
5Y1+4Y28
Y1 , Y2 0
The solution of the dual problem can be obtained by using simplex minimization
procedures.
Min Z = 40 Y1+80 Y2+oS1+oS2+MA1+MA2
M
M
A1
A2
Zj
Cj-Zj
Y2
A2
Zj
Cj-Zj
Y2
Y1
Zj
Cj-Zj
80
M
80
40
40
80
0
0
M
M
Y1
Y2
S1
S2
A1
A2
6
2
8
-1
0
1
0
8
5
4
0
-1
0
1
14M
7M
12M
-M
-M
M
M
40-7M 80-12M
M
M
0
0
1
-1/8
0
1/8
0
5
4
0
-1
-1/2
1
60+5M 20+4M
80
1/2M-10 -M 10-1/2M
M
20-4M
0
10-1/2M
M
1/2M-10
0
7/16
0
1
-5/32
1/16
5/32
-1/16
5/4
1
0
1/8
-1/4
-1/8
1/4
85
40
80
-7.5
-5
7.5
5
0
0
7.5
5
M-7.5
M-5
:( primal)
Cj
8
6
X2
X1
Zj
Cj-Zj
5
7.5
85
6
X1
0
1
6
0
Dual
S1 = 7.5
S2 = 5
Y1 = 5/4
Y2 = 7/16
= 85
Ex1/ Obtain the dual of the following LP:
Max Z =5X1 + 10X2+8X3
St.to: 3X1 + 5X2+2X360
8
X2
1
0
8
0
0
S1
-1/8
5/4
-5/4
Primal
X1 = 7.5
X2 = 5
S1 = 5/4
S2 = 7/16
Z = 85
0
S2
-1/16
5/32
7/16
-7/16
4X1 + 4X2+4X372
2X1 + 4X2+5X3100
X1 , X2,X3 0
Solution: Min = 60Y1+72Y2+100Y3
St. to: 3Y1+4Y2+2Y35
5Y1+4Y2+4Y310
2Y1+4Y2+5Y38
Y1 , Y2 , Y3 0
Ex2/ Consider the following LP and write the associated dual problem:
Max Z =7X1 + 5X2
St.to: 3X1+X248
2X1+X240
X1 , X2 0
Min = 48Y1+40Y2
St. to: 3Y1+2Y27
Y1+Y25
Y1 , Y2 0
Ex3/ Write the dual for each of the following primal problem:
Max Z =10X1 + 12X2
St.to: X1+X215
X1 = 6
X2 8
Solution: X1+X215
X1 6
-X1 -6
X16
X2 8
-X2-8
X180
X260
3) Max Z = 23X1 + 32X2
Sub. To: 10X1+6X22500
5X1+10X22000
X1+2X2500
4) Write the primal problem for the following:
Min = 60Y1+72Y2+100Y3
St. to: 3Y1+4Y2+2Y35
5Y1+4Y2+4Y310
2Y1+4Y2+5Y38
Y1 , Y2 , Y3 0
The assignment model
Suppose a company has M jobs that must be completed and it has at least n
workers who can perform any of the M jobs but possibly in a different amount of
time. Which worker should be assigned to each job to minimize the overall time to
complete all M jobs, if each worker is assigned to one and only one job. This is the
classical assignment problem.
The general assignment model with n workers and M jobs is represented in the
following matrix:
jobs
workers
1
2
n
ai
bj
C11
C21
Cn1
1
C12
C22
Cn2
1
C1m
C2m
Cnm
1
1
1
1
The element Cij represents the cost of assigning worker i to job j (i= 1,2,..,n)
(j= 1,1,..,m).
If we let:
Xij =
ij
Xij
for j= 1,2,, m
for i= 1,2,, n
Step 4: Select the smallest uncover element, and subtract it from every uncovered
element, then add it to every element at the intersection of two lines. And repeat
step 3 until the feasible assignment found.
For example, suppose that three jobs must be assigned to three machines, each
machine must be assigned to only one job, and each job must be assigned to only
one machine.
The costs are shown below. Solve the problem by using the Hungarian algorithm.
Jobs
M1
25
15
22
A
B
C
Machines
M2
31
20
19
M3
35
24
17
Solution:
A
B
C
M1
10
0
7
M2
12
1
0
M3
18
7
0
A
B
C
M1
0
0
8
M2
1
0
0
M3
7
6
0
, B-M2 20 ,
A
B
C
M1
0
0
7
M2
2
1
0
M3
8
7
0
A
B
C
M1
0
0
7
M2
2
1
0
M3
8
7
0
C-M3
Use the Hungarian method to solve the same problem for the maximum
productivity.
Job
A
B
C
A
B
C
A
B
C
M1
25
15
22
M2
31
20
19
M3
35
24
0
M1
10
20
13
M1
0
0
0
M2
4
15
16
M2
0
1
9
M3
0
11
18
M3
0
1
15
A
B
C
A
B
C
M1
0
10
3
M1
1
0
0
M2
0
11
12
M2
0
0
8
Second
M3
M2
M2 20
M3 24
M1
M1
Unbalanced cases
If MN
Max or Min
Example 1:
M3
0
11
18
M3
0
0
14
Use the Hungarian algorithm to solve the assignment problem having the following
time table.
Job
J1
J2
J3
Solution:
J1
J2
J3
Dummy
J4
J1
J2
J3
J4
M1
80
50
50
M3
120
130
230
M4
100
80
150
M1
80
50
50
0
M2
110
160
100
0
M3
120
130
230
0
M4
100
80
150
0
M1
10
0
0
30
M2
10
80
20
0
M3
20
50
150
0
M4
0
0
70
0
M1
0
0
0
40
J1
J2
J3
J4
M2
110
160
100
M2
0
70
10
0
M3
10
40
140
0
M4
0
0
70
10
Note: M<N
So we added dummy row with
Cij= 0
J1
J2
J3
J4
M1
0
0
0
0
M2
30
110
50
0
M3
40
80
180
0
M4
20
30
100
0
J1
J2
J3
J4
M1
0
0
0
20
M2
10
90
30
0
M3
20
60
160
0
M4
0
10
80
0
M1
20
60
80
M2
60
30
100
M3
50
80
90
M4
55
75
80
T4
T5
65
70
80
65
75
60
70
65
Example 3/ suppose four people can each perform any one of four different jobs
but possibly in different amount of time. The following table gives the
corresponding time to perform the various jobs. Which person should be assigned
to each job to minimize the total time to perform all four jobs?
Person
A
B
C
D
1
2
15
13
4
Cij
Second solution,
4
7
8
11
9
Solution:
Ex 2: first solution,
Cij
T1 M5
T1 M5
T2 M3
80
T2 M4
75
T3 M2
100
T3 M2
100
T4 M4
70
T4 M3
75
T5 M1
70
T5 M1
70
____
____
320
320
Lec.N
o(5)
University of Technology
Electro-mechanical Engineering Dept.
Transportation Model
2011
Transportation Model
Introduction
The transportation model is a special class of the linear programming problem. It
deals with the solution in which an item is shipped from sources (e.g. factories) to
destinations (e.g. warehouses). The objective is to determine the amounts shipped
from each source to each destination that minimize the total shipping cost while
satisfying both the supply limits and the demand requirements.
The Mathematical Model
The general problem can be presented by the network.
- There are M sources.
S1,S2..Sm
- And n destinations.
D1,D2..Dn
- The arcs linking the sources and destinations represent the routes between
the sources and the destinations.
- The transportations cost per unit = Cij and the amount shipped = Xij
- The amount of supply at source i is ai and the amount of demand at
destination j is bj.
The objective of the model is to determine the unknown Xij that will minimize the
total transportation cost, while satisfying all the supply and demand restrictions.
a1
a2 Units of supply ai
S1
S2
am
Si
Sources
c11
x11
cm
c12
c13
x12
xm
x13
Dj
b1
b2
bj
bn
Destination
Objective Function:
Min Z:
Units of demand
Cij Xij
Subject to:
j = 1,2,n
i = 1,2,m
1)
bcA)
The method starts at the North West corner cell of the tableau (variable X11).
Step 1: Allocate as much as possible to the select cell, and adjust the associated
amounts of supply and demand by subtracting the allocated amount.
Step 2: Cross out the row or column with zero supply or demand to indicate that no
further assignments can be made in that row or column. If both the row and
column net to zero simultaneously, cross out one only, and leave a zero supply
(demand) in the uncrossed-out row (column).
Step3: If exactly one row or column is left uncrossed-out, stop. Otherwise, move to
the cell to the right if a column has just been crossed or the one below if a row has
been crossed out. Go to step1.
Example1:
In the following transportation problem use the north west- corner method
(N.W.C) to find the starting solution.
S/D
S1
S2
S3
S4
Demand
Solution:
D2
3
9
10
13
120
D1
5
3
11
6
160
S/D
S1
S2
S3
11
S4
D3
10
8
7
6
60
D2
3
10
ai
150
70
80
D1
D3
150
10
60
10
60
13
20
6
bj
160
120
Z= 5X150 + 3X10 + 9X60 + 10X60 +7X20 + 6X40
= 2300
B)
Supply
150
70
80
40
340/340
40
40
60
M+N-1=6
Least-Cost method:
340/340
S2
S3
11
S4
D2
10
ai
150
70
10
80
D1
D3
3
30
120
70
20
60
13
40
bj
160
120
Z= 5X30 + 3X120 + 3X70 + 11X20 + 7X60 + 6X40
40
60
340/340
= 1600
C)
Vogal Method
Vogal method is an improved of version of the least cost method that generally
produces better starting solutions.
Step 1: For each row (each column) determine a penalty measure by subtracting
the smallest unit cost element in the row (column) from the next smallest unit cost
element in the same row (column).
Step 2: Identify the row or column with the largest penalty. Break ties arbitrarily.
Allocate as much as possible to the variable with the least unit cost in the selected
row or column. Adjust the supply and demand and cross out the satisfied row or
column.
Step 3: a) If exactly one row or column with supply or demand remains uncrossed
out, stop.
b) If one row or column with positive supply (demand) remains uncrossed out,
determine the basic variables in the row (column) by the least-cost method, stop.
c) If all the uncrossed out rows and columns have (remaining) zero supply and
demand, determine the zero basic variables by the least-cost method.
d) Otherwise, go to step 1.
Example 3: Solve the transportation model of example (1) by using vogal method.
S/D
S1
D1
5
S2
S3
11
S4
D2
10
ai
150
V1 V2 V3 V4 V5 V6
2
5
70
10
80
40
3
30
D3
120
70
20
60
13
40
bj
160
120
60
340/340
V1
2
6
1
V2
2
1
V3
3
1
V4
5
V5
Z= 5X30 + 3X120 + 3X70 + 11X20 + 7X60 + 6X40
= 1600
This solution happens to have the same objective value (1600) as in the Least-cost
method (L.C). usually, however vogal method is expected to produce better
starting solutions for the transportation method.
Stepping Stone Method (S.S)
After determining the starting ( using any of the three methods) we use one of the
following method :
1.
2.
D2
ai
70
S2
100
D1
bj
90
By using the N.W.C
S/D
S1
S2
bj
80
170/170
D2
3
ai
70
100
D1
70
20
80
90
80
170/170
6-
= 1140
S1D
70
80
S/D
S1
D1
6
S2
-7
D2
ai
70
3
70
7
100
90
10
bj
90
Z= 1140 2X70 = 1000
80
170/170
<
From/To
or
>
Dummy
Factory
capacity
ai
250
300
300
250
E
8
50
F
Requirement
bj
200
7
300
50
5
200
150
200
150
150
700\850
Z = 3350
From/To
Plant W
Plant
supply
ai
200
175
200
Plant X
10
50
Plant Y
12
100
7
15
6
75
75
Dummy
plant
Requirement
bj
0
250
0
100
50
50
150
500\450
Z= 2850
Problems:
Q1/ Consider the following transportation problem. Find the starting solution by
using the following methods:
123From/To
V1
V2
V3
Demand bj
Solutions:
1-
N.W.C method.
L.C method.
Vogal method.
C1
2
10
7
4
C2
2
8
6
3
C3
2
5
6
4
C4
1
4
8
4
N.W.C = 93
Supply ai
3
7
5
15\15
23-
L.C = 79
Vogal = 68
Q2 / Use the north west-corner method to find the starting solution (the initial
solution) for the following problem:
Factory
From/To
S1
S2
S3
Demand bj
Solution: Z= 65
D1
0
1
5
20
Warehouse
D2
4
2
3
10
D3
0
5
7
15
D4
2
6
9
15
Supply ai
5
10
15
60\30
Q3/ determine the initial solution for the following problem(use the three
methods).
Warehouse
From/To
A
B
C
Demand bj
1
10
14
18
1000
Markets
2
8
17
7
2000
3
6
5
11
500
4
4
2
9
1500
Supply ai
2000
1300
1700
5000\5000
Solutions:
123-
N.W.C Z= 52200
L.C Z= 30700
Vogal Z= 30700
Q4/ Use the northwest corner method to find the starting solution, then determine
the optimum solution by using (S.S) method for the following problem.
S/D
S1
S2
S3
S4
bj
D1
5
3
11
6
160
D2
3
9
10
13
120
D3
10
8
7
6
60
ai
150
70
80
40
340\340
Z0 = 2300
Z1 = 2300 8X60 = 1820
Z2 = 1820 5X40 = 1620
Z3 = 1620 1X20 = 1600
C1
C2
C3
C4
ai
3
3
V2
10
1
V3
3
6
3
6
1
bj
15\15
C3
C4
ai
3
4+3-1 = 6
Z= 2X3 + 1X10 + . +8X4 = 93
L.C.method
From\To
V1
C1
2
C2
2
1
3
V2
10
2
bj
2
V3
15\15
4+3-1 =6
Z = 1X3 + 1X4 + + 6X3 = 79
Vogal method
From\To
V1
C1
C2
C3
C4
ai
3
U1 U2 U3 U4 U5
1
3
V2
10
3
V3
4
8
bj
U1
U2
U3
U4
Z = 2X3 + 5X3 + + 6X1 = 68
15\15
Q2: ai bj
From \To
S1
S2
S3
bj
D1
0
1
5
20
N.W.C method
From\To
S1
D2
4
2
3
10
D3
0
5
7
15
D4
2
6
9
15
ai
5
10
15
60\30
bj - ai = 30
D1
D2
D3
D4
ai
5
10
15
30
5
S2
1
10
S3
5
5
Dummy S4
10
0
0
bj
20
10
15
15
15
15
60\60
Q3:
L.C.M
Market
From\To
A
10
1000
warehouse
14
4
4
300
500
17
ai
2000
200
2
1300
1300
C
bj
18
11
1700
2000
1000
1700
500
1500
5000\5000
N.W.C
Market
From\To
A
1
10
2
8
1000
warehouse
B
C
bj
5
300
1300
11
1700
1000
14
17
18
1000
7
1000
ai
2000
2000
200
500
1500
1500
5000\5000
Vogal method
From\To
A
1
10
2
300
3
6 500
4
4 200
ai
2000
V1
2
V2
2
1300
1700
V3
2
V4
2
1000
B
14
17
1300
C
18
1700
11
bj
V1
V2
V3
V4
V5
15\15
V5