GOAL PROGRAMMING

© 2003 Thomson/South-Western 1
Slide
 Introduction
 Most of the optimization problems
considered to this point have had a
single objective.
 Often, more than one objective can be
identified for a given problem.
Maximize Return or Minimize Risk
Maximize Profit or Minimize Pollution
 These objectives often conflict with one
another.

© 2003 Thomson/South-Western 2
Slide
Sebagian besar masalah optimasi yang dibahas samp
Sebagian besar masalah optimasi yang dibahas sampai
saat ini
mempunyai satu tujuan.
Seringkali, lebih dari satu tujuan dapat diidentifikasi untuk
suatu
masalah tertentu.
Maksimalkan Pengembalian atau Minimalkan Risiko
Memaksimalkan Keuntungan atau Meminimalkan Polusi
Tujuan-tujuan ini sering kali bertentangan satu sama lain.
ai saat ini mempunyai satu tujuan.
Seringkali, lebih dari satu tujuan dapat diidentifikasi
untuk suatu masalah tertentu.
Maksimalkan Pengembalian atau Minimalkan Risiko
Memaksimalkan Keuntungan atau Meminimalkan Polusi
Tujuan-tujuan ini sering kali bertentangan satu sama lain.

© 2003 Thomson/South-Western 3
Slide
 Goal Programming (GP)
 Most LP problems have hard
constraints that cannot be
violated...
There are 1,566 labor hours
available.
There is RM850,00 available for
projects.

© 2003 Thomson/South-Western 4
Slide
 Kebanyakan soal LP memiliki batasan keras yang tidak dapat dilanggar...

Ada 1.566 jam kerja yang tersedia.

Ada RM850,00 tersedia untuk proyek.

© 2003 Thomson/South-Western 5
Slide
 Goal Programming (GP)
o In some cases, hard constraints are too
restrictive...
oYou have a maximum price in mind
when buying a car (this is your “goal”
or target price).
oIf you can’t buy the car for this price
you’ll likely find a way to spend more.
o We use soft constraints to represent
such goals or targets we’d like to
achieve.
© 2003 Thomson/South-Western 6
Slide
 Dalam beberapa kasus, batasan keras terlalu membatasi...
Anda sudah memikirkan harga maksimum saat membeli mobil (ini adalah
“tujuan” atau harga target Anda).
Jika Anda tidak dapat membeli mobil dengan harga ini, kemungkinan besar
Anda akan menemukan cara untuk membelanjakan lebih banyak.
Kami menggunakan batasan lunak untuk mewakili tujuan atau target yang
ingin kami capai

© 2003 Thomson/South-Western 7
Slide
 GP: Example 1
Indah Beach Hotel Expansion
Roslan wants to expand the convention center
at his hotel in Port Dickson.
The types of conference rooms being
considered are:
Size (sq m) Unit Cost
Small 400 RM18,000
Medium 750 RM33,000
Large 1,050 RM45,150

© 2003 Thomson/South-Western 8
Slide
 GP: Contoh 1 Perluasan Hotel Indah Beach

Roslan ingin memperluas pusat konvensi di hotelnya di Port Dickson.

Jenis ruang konferensi yang dipertimbangkan adalah:

Size (sq m) Unit Cost

Small 400 RM18,000
Medium 750 RM33,000
Large 1,050 RM45,150

© 2003 Thomson/South-Western 9
Slide
 Example 1

Roslan would like to add 5 small, 10 medium and 15

large conference rooms.
He would also like the total expansion to be 25,000
square meter and to limit the cost to RM1,000,000.

© 2003 Thomson/South-Western 10
Slide
 Roslan ingin menambah 5 ruang konferensi kecil, 10 ruang sedang, dan 15 ruang
konferensi besar.
Dia juga ingin total perluasan menjadi 25.000 meter persegi dan membatasi
biaya hingga RM1.000.000.

© 2003 Thomson/South-Western 11
Slide
 Defining the Decision
Variables
X1 = number of small rooms to add
X2 = number of medium rooms to
add
X3 = number of large rooms to add

© 2003 Thomson/South-Western 12
Slide
 Defining the Goals
Goal 1: The expansion should include approximately
5 small conference rooms.
Goal 2: The expansion should include approximately
10 medium conference rooms.
Goal 3: The expansion should include approximately
15 large conference rooms.
Goal 4: The expansion should consist of
approximately 25,000 square meter.
Goal 5: The expansion should cost approximately
RM1,000,000.

© 2003 Thomson/South-Western 13
Slide
 Defining the Goals

Sasaran 1: Perluasan harus mencakup sekitar 5 ruang konferensi kecil.

Sasaran 2: Perluasan ini harus mencakup sekitar 10 ruang konferensi berukuran
sedang.
Sasaran 3: Perluasan ini harus mencakup sekitar 15 ruang konferensi besar.
Sasaran 4: Perluasan harus mencakup sekitar 25.000 meter persegi.
Sasaran 5: Perluasan ini memerlukan biaya sekitar RM1.000.000.

© 2003 Thomson/South-Western 14
Slide
 Defining the Goal Constraints
Small Rooms
 
X1  d  d 5
1 1

Medium Rooms
 
X 2  d  d 10
2 2

Large Rooms
 
X 3  d  d 15
3 3
where
 
d , d 0
i i

© 2003 Thomson/South-Western 15
Slide
 Defining the Goal Constraints
(cont’d)
Total Expansion
 
400X 1  750 X 2  1,050 X 3  d  d 25,000
4 4

Total Cost (in RM1,000s)

 
18X 1  33X 2  4515
. X 3  d  d 1,000
5 5

where
 
d , d 0
i i

© 2003 Thomson/South-Western 16
Slide
 GP Objective Functions
There are numerous objective
functions we could formulate for a
GP problem.
Minimize the sum of the deviations:
 d
i
i

 d i  (choice 1)
MIN
Problem: The deviations measure different things, so what
does this objective represent?

© 2003 Thomson/South-Western 17
Slide
 GP Objective Functions
(cont’d)
Minimize the sum of percentage deviations

MIN

1
i t di  di  (choice 2)
 value of goal i .
where t represents the target
i

© 2003 Thomson/South-Western 18
Slide
 GP Objective Functions
(cont’d)
Problem:
Suppose the first goal is underachieved
by 1 small room and the fifth goal is
overachieved by RM20,000.

We underachieve goal 1 by 1/5=20%

We overachieve goal 5 by 20,000/1,000,000=
2%

© 2003 Thomson/South-Western 19
Slide
 GP Objective Functions
(cont’d)
This implies being RM20,000 over
budget is just as undesirable as
having one too few small rooms.

Is this true? Only the decision

maker can say for sure.

© 2003 Thomson/South-Western 20
Slide
 GP Objective Functions
(cont’d)
Weights can be used in the previous
objectives to allow the decision maker
indicate desirable vs. undesirable deviations
the relative importance of various goals

© 2003 Thomson/South-Western 21
Slide
 GP Objective Functions
(cont’d)
Minimize the weighted sum of deviations

MIN
 w d
i
i

i

w d 
i i  (choice 3)

Minimize the weighted sum of % deviations

1  
MIN
i t  i i i di 
w d  w  
(choice 4)
i

© 2003 Thomson/South-Western 22
Slide
 Defining the Objective
 Assume
It is undesirable to underachieve any of the first
three room goals
It is undesirable to overachieve or underachieve
the 25,000 sq meter expansion goal
It is undesirable to overachieve the RM1,000,000
total cost goal

© 2003 Thomson/South-Western 23
Slide
 Defining the Objective

w 1  w 2  w 3  w 4  w 4  w 5
MIN : d1  d 2  d 3  d4  d4  d 5
5 10 15 25,000 25,000 1,000,000

Initially, we will assume all the above

weights equal 1.

© 2003 Thomson/South-Western 24
Slide
 GP: EXAMPLE 2
Kreatif Company employs skilled artisans to
produce clay bowls and mugs with
authentic Malaysian designs and colors.
The two primary resources used by the
company are special pottery clay and
skilled labor.

© 2003 Thomson/South-Western 25
Slide
 EXAMPLE 2
The company desires to know how many
bowls and mugs to produce each day to
maximize profit. The two products have
the following resource requirements for
production and profit per item produced:

© 2003 Thomson/South-Western 26
Slide
 EXAMPLE 2
Resource requirement
Products Labor Clay Profit
(hr/unit) (kg/unit (RM/unit)
)
Bowl 1 4 40
Mug 2 3 50

There are 40 hours of labor and 120 kg of clay available

each day for production. Formulate a LP model.

© 2003 Thomson/South-Western 27
Slide
 EXAMPLE 2
Decision variables:
x1 = number of bowls to produce
x2 = number of mugs to produce

Objective function: to maximize total profit

Maximize Z = 40x1 + 50x2
Constraints:
Labor: x1  2 x 2 40

Clay: 4 x1  3 x2 120

© 2003 Thomson/South-Western 28
Slide
 The Complete LP Model
Maximize Z = 40x1 + 50x2

Subject to
x1  2 x 2 40
4 x1  3 x2 120
x1 , x2 0
1. Solve using graph
2. Solve using Excel Solver
3. Solve using DS For Windows
© 2003 Thomson/South-Western 29
Slide
 Goal Programming
Goal programming may be used to
solve linear programs with multiple
objectives, with each objective
viewed as a "goal".
In goal programming, di+ and di- ,
deviation variables, are the
amounts a targeted goal i is
overachieved or underachieved,
respectively.

© 2003 Thomson/South-Western 30
Slide
 Goal Programming
The goals themselves are added to
the constraint set with di+ and di-
acting as the surplus and slack
variables.
One approach to goal programming
is to satisfy goals in a priority
sequence. Second-priority goals
are pursued without reducing the
first-priority goals, etc.

© 2003 Thomson/South-Western 31
Slide
 Goal Programming
For each priority level, the objective
function is to minimize the
(weighted) sum of the goal
deviations.
Previous "optimal" achievements of
goals are added to the constraint set
so that they are not degraded while
trying to achieve lesser priority
goals.

© 2003 Thomson/South-Western 32
Slide
 Goal Programming Approach
Step 1:
Decide the priority level of each goal.

Step 2:
Decide the weight on each goal. If a
priority level has more than one goal,
for each goal i decide the weight, wi , to
be placed on the deviation(s), di+
and/or di-, from the goal.

© 2003 Thomson/South-Western 33
Slide
 Goal Programming Approach
Step 3:
Set up the initial linear program.
Min w1d1+ + w2d2-
s.t. Functional Constraints,
and Goal Constraints

Step 4: Solve the current linear program.

If there is a lower priority level, go to

Step 5. Otherwise, a final solution has been

reached.

© 2003 Thomson/South-Western 34
Slide
 Goal Programming Approach
Step 5: Set up the new linear program.

Consider the next-lower priority level

goals and formulate a new objective
function based on these goals. Add a
constraint requiring the achievement of
the next-higher priority level goals to be
maintained.

© 2003 Thomson/South-Western 35
Slide
 Goal Programming Approach
Step 5: cont…
The new linear program might be:

Min w3d3+ + w4d4-

s.t. Functional Constraints,

Goal Constraints, and
w1d1+ + w2d2- = k

Go to step 4. (Repeat steps 4 and 5 until

all priority levels have been examined.)
© 2003 Thomson/South-Western 36
Slide
 Recall Kreatif Company
Problem
Maximize Z = 40x1 + 50x2

Subject to
x1  2 x 2 40
4 x1  3 x2 120
x1 , x2 0

Now, let us suppose that instead of having one objective, the company
has several objectives that are listed below in order of importance.

© 2003 Thomson/South-Western 37
Slide
 Alter the problem:
To avoid layoffs, the company do not
want to use fewer than 40 hours of
labor per day (priority 1).
The company would like to achieve a
satisfactory profit of RM 1,600 per
day (priority 2).

© 2003 Thomson/South-Western 38
Slide
 Alter the problem:
Because the clay must be stored in a
special place so that it does not dry
out, the company prefers not to
keep more than 120 kg on hand
each day (priority 3).
Because high overhead costs results
when the plant is kept open past
normal hours, the company would
like to minimize the amount of
overtime (priority 4).
© 2003 Thomson/South-Western 39
Slide
 Alter the problem:
To avoid layoffs, the company do not want to
use fewer than 40 hours of labor per day
(priority 1).
x1  2 x2 40
 
x1  2 x2  d  d 40 1 1
 
d ,d
1 1 are deviational variables.

will minimize the underachievement variable: d
1

© 2003 Thomson/South-Western 40
Slide
 Alter the problem:
The company would like to achieve a satisfactory
profit of RM 1,600 per day (priority 2).

Maximize Z = 40x1 + 50x2

40 x1  50 x 2 1600
 
40 x1  50 x2  d  d 1600 2 2


will minimize the underachievement variable: d 2

© 2003 Thomson/South-Western 41
Slide
 Alter the problem:
Because the clay must be stored in a special place
so that it does not dry out, the company prefers
not to keep more than 120 kg on hand each day
(priority 3).
4 x1  3 x2 120
 
4 x1  3 x2  d  d 120 3 3


will minimize the overachievement variable: 3 d

© 2003 Thomson/South-Western 42
Slide
 Alter the problem:
Because high overhead costs results when the plant is
kept open past normal hours, the company would like
to minimize the amount of overtime (priority 4).

x1  2 x 2 40
 
x1  2 x2  d  d 401 1


will minimize the overachievement variable: 1 d

© 2003 Thomson/South-Western 43
Slide
 The Complete GP Model
   
Minimize ( d1 , d 2 , d 3 , d1 )
Subject to
 
x1  2 x2  d  d 40 1 1
 
40 x1  50 x2  d  d 1600 2 2
 
4 x1  3 x2  d  d 120 3 3
 
x1  2 x2  d  d 40 1 1

© 2003 Thomson/South-Western 44
Slide
 The Complete GP Model
   
Minimize ( d1 , d 2 , d 3 , d1 )

Subject to
 
x1  2 x2  d  d 40 1 1
 
40 x1  50 x2  d  d 1600 2 2
 
4 x1  3 x2  d  d 120 3 3

     
x1 , x 2 , d , d , d , d , d , d 0
1 1 2 2 3 3

© 2003 Thomson/South-Western 45
Slide
Demonstration Using Excel Solver

and DS for Windows

© 2003 Thomson/South-Western 46
Slide
 Comments About GP

GP involves making trade-offs

among the goals until the most
satisfying solution is found
GP objective function values should
not be compared because the
weights are changed in each
iteration. Compare the solutions!

© 2003 Thomson/South-Western 47
Slide
 Comments About GP

An arbitrarily large weight will

effectively change a soft constraint
to a hard constraint.
Hard constraints can be place on
deviational variables.

© 2003 Thomson/South-Western 48
Slide
 Summary of Goal
Programming
1. Identify the decision variables
in the problem.
2. Identify any hard constraints
in the problem and formulate
them in the usual way.
3. State the goals of the
problem along with their target
values.

© 2003 Thomson/South-Western 49
Slide
 Summary of Goal
Programming
4. Create constraints using the
decision variables that would achieve
the goals exactly.
5. Transform the above constraints
into goal constraints by including
deviational variables.
6. Determine which deviational
variables represent undesirable
deviations from the goals.

© 2003 Thomson/South-Western 50
Slide
 Summary of Goal
Programming
7. Formulate an objective that
penalizes the undesirable deviations.
8. Identify appropriate weights for
the objective.
9. Solve the problem.
10. Inspect the solution to the
problem. If the solution is
unacceptable, return to step 8 and
revise the weights as needed.

© 2003 Thomson/South-Western 51
Slide
 Food of thought…

One cannot escape the feeling that

these mathematical formulas have
an independent existence and an
intelligence of their own, that they
are wiser than we are, wiser even
than their discoverers.

© 2003 Thomson/South-Western 52
Slide