Goal Programming

J a m e s P. I g n i z i o C arlos Romero
Resource
Resource Manag
Managemen
ementt Associate
Associatess Technical
echnical Unive
Universit
rsityy of Madrid 
Madrid 

I. INTROD
INTRODUC
UCTIO
TION
N VI. THE MULTIDI
MULTIDIMEN
MENSION
SIONAL
AL DUAL
II. HISTORIC
HISTORICAL
AL SKETCH
SKETCH VII. ALGORIT
ALGORITHMS
HMS FOR
FOR SOLUTI
SOLUTION
ON
III. THE MULTIPLEX
TIPLEX MODEL
MODEL VIII. GOAL PROGRAMMIN
PROGRAMMING
G AND UTILITY
UTILITY OPTIMIZA
OPTIMIZATION
TION
IV.
IV. FORMS OF
OF THE ACHIEVEM
ACHIEVEMENT
ENT FUNCTION
FUNCTION IX.
IX. EXTE
EXTENS
NSIO
IONS
NS
V. GENERAL FORM
FORM OF THE MULTIPLEX
MULTIPLEX MODEL
MODEL X. THE
THE FUTU
FUTURE
RE

GLOSSARY and objectives, has often been cited as being the

achievement function The function that serves to
measure the achievement of the minimization of  goals and objectives) as based on its extensive list of
unwanted goal deviation variables in the goal pro-
gramming model.
goal function  A mathematical
mathematical function that is to be
achieved at a specified level (i.e., at a prespecified
“aspiration”
“aspiration” level).
goal program  A mathematical model, consisting of  telligence (particularly genetic algorithms and neural
linear or nonlinear functions and continuous or
discrete variables, in which all functions have been
transformed into goals.
multiplex Originally this referred to the multiphase
simplex algorithm
algorithm employed to to solve linear
linear goal
programs. More recently it defines certain specific
models and methods employed in multiple- or
single-objective
single-objective optimization in general.
negative deviation The amount of deviation for a
given goal by which it is less than the aspiration
level.
 positive deviation The amount of deviation for a given
goal by which it exceeds the aspiration level.
satisfice  An old Scottish word referring to the desire,
in the real world, to find a practical solution to a
given problem, rather than some utopian result for
an oversimplified model of the problem.
GOAL PROGRAMMING, a powerful and effective
methodology for the modeling, solution, and analysis
of problems having multiple and conflicting goals
“workhorse” of multiple objective optimization (i.e.,
the solution to problems having multiple, conflicting
 
successful applications in actual practice. Here we de-
scribe the method and its history, cite its mathemati-
cal models and algorithms, and chronicle its evolu-
tion from its original form into a potent methodology 
that now incorporates techniques from artificial in-
networks). The article concludes with a discussion of 
recent extensions and a prediction of the role of goal
programming in real-world problem solving in the
21st century.
I . I NT
NT RO
RO DU
DU CT
CT IO
IO N
A. Defini
Definitio
tions
ns and Origin
Origin
Real-world decision problems—unlike
problems—unlike those found in
textbooks—involve multiple, conflicting objectives and
goals, subject to the satisfaction of various hard and 
soft constraints.
constraints. In short, and as the experienced prac-
titioner is well aware, problems that one encounters
outside  the classroom are invariably massive, messy,
changeable, complex, and resist treatment via con-
 ventional approaches. Yet the vast majority of tradi-
tional approaches to such problems utilize conven-
tional models and methods that idealistically and
unrealistically (in most cases) presume the optimiza-
tion of a single-objective subject to a set of rigid

Encyclopedia of Information Systems, Volume 2

Copyright 2003, Elsevier Science (USA). All rights reserved. 489
490 Goal Programming

constraints. Goal programming was introduced in an and comfortably, into the concept of optimization or
attempt to eliminate or, at the least, mitigate this dis- efficiency (i.e., nondominated solutions) as used by 
quieting disconnect. Conceived and developed by  more conventional forms of mathematical modeling.
 Abraham Charnes and William Cooper, goal pro- This is because, in goal programming, we seek a use-
gramming was originally dubbed “constrained regres- ful, practical, implementable, and attainable solution
sion.” Constrained regression, in turn, was and is a rather than one satisfying the mathematician’s desire
powerful nonparametric method for the development  for global optimality. (However, if one wishes, it is rel-
of regression functions (e.g., curve-fitting) subject to atively trivial to develop efficient, or nondominated
side constraints. solutions for any goal programming problem. That 
Charnes and Cooper first applied constrained re- matter is briefly described in a section to follow.)
gression in the 1950s to the analysis of executive com-
pensation. Recognizing that the method could be ex-
tended to a more general class of problems—that is, C. A Brief List of Applications
any quantifiable problem having multiple objectives
and soft, as well as rigid constraints—Charnes and Goal programming’s label as the “workhorse” of 
Cooper later renamed the method goal programming multiple-objective optimization has been achieved by 
 when describing it within their classic 1961 two-volume its successful solutions of important real-world prob-
text  Management Models and Industrial Applications of  lems over a period of more than 50 years. Included
Linear Programming. among these applications are:

• The analysis of executive compensation for

B. Philosophical Basis
General Electric during the 1950s
• The design and deployment of the antennas for
The two philosophical concepts that serve to best dis-
the Saturn II launch vehicle as employed in the
tinguish goal programming from conventional (i.e.,
single-objective) methods of optimization are the in-  Apollo manned moon-landing program
corporation of flexibility in constraint functions (as • The determination of a siting scheme for the
opposed to the rigid constraints of single-objective Patriot Air Defense System
optimization) and the adherence to the philosophy of  • Decisions within fisheries in the United Kingdom
“satisficing” as opposed to optimization. Satisficing, in • A means to audit transactions within the financial
turn, is an old Scottish word that defines the desire to sector (e.g., for the Commercial Bank of Greece)
find a practical, real-world solution to a problem— • The design of acoustic arrays for U.S. Navy 
rather than a utopian, optimal solution to a highly  torpedos
simplified (and very possibly oversimplified) model of  • As well as a host of problems in the general areas
that problem. The concept of satisficing, as opposed of agriculture, finance, engineering, energy, and
to optimization, was introduced by Herbert Simon resource allocation.
in 1956.
 As a consequence of the principle of satisficing,
the “goodness” of any solution to a goal programming D. Overview of Material to Follow
problem is represented by an achievement  function,
rather than the objective function of conventional op- In this article, the topic of goal programming is cov-
timization. The goal programming achievement func- ered in a brief but comprehensive manner. Sections
tion measures the degree of  nonachievement  of the to follow discuss the past, present, and future of goal
problem goals. The specific way in which this programming as well as the models and algorithms
nonachievement is measured characterizes the partic- for implementation. The reader having previous ex-
ular subtype of goal programming approach that is posure to the original goal programming approach
being employed, and may be defined so as to include  will (or should) immediately notice the many signifi-
the achievement of commensurable as well as non- cant changes and extensions that occurred during the
commensurable goals. 1990s. As just one example, powerful and practical hy-
It should be emphasized that, because a goal pro- brid goal programming and genetic algorithm mod-
gramming problem is to be satisficed, it is possible eling and solution methods will be discussed. Readers
that the solution derived may not fit, conveniently  seeking more detailed explanations of any of the ma-
Goal Programming 491

terial covered herein are referred to the Bibliography  III. THE MULTIPLEX MODEL
at the end of this article.
A. Numerical Illustrations
II. HISTORICAL SKETCH  Any single-objective problem, and most multiple-
objective ones, can be placed into a model format 
 As mentioned, goal programming was conceived by  that has been designated as the multiplex model, and
 Abraham Charnes and William Cooper nearly a half  then solved via the most appropriate version of a mul-
century ago. The tool was extended and enhanced by  tiplex (or sequential goal programming) algorithm.
their students and, later, by other investigators, most  For example, consider a conventional (albeit simple)
notably Ijiri, Jääskeläinen, Huss, Ignizio, Gass, linear programming problem taking on the following
Romero, Tamiz, and Jones. traditional form:
In its original form, goal programming was strictly 
limited to linear multiple-objective problems. Ignizio, in Maximize z   10x 1  4x 2 (1)
the 1960s, extended the method to both nonlinear and Subject to: x 1  x 2  100 (2)
integer models, developed the associated algorithms for
these extensions, and successfully applied them to a x 2  4 (3)
number of important real-world problems, including, as x  0 (4)
previously mentioned, the design of the antenna systems
for the Saturn II launch vehicle as employed in the Ignoring the fact that this undemanding single-
 Apollo manned moon-landing program. During that  objective model can be solved by inspection, let us
same period, and in conjunction with Paul Huss, Ignizio transform it into the multiplex form for the sake of il-
developed a sequential algorithm that permits one to lustration. To do so, we add a negative deviation vari-
extend—with minimal modification—any single-objec- able to, and subtract a positive deviation variable from,
tive optimization software package to the solution of any  each constraint. In addition, we transform the maxi-
class of goal programming models (the approach was mizing objective function into a minimizing form by 
also developed, independently, by Dauer and Kruger). simply multiplying the original objective function by 
Later in that same decade, Ignizio developed the a negative one. The resultant model, in multiplex
concept of the multidimensional dual, providing goal form, can be written:
programming with an effective economic interpreta-
Lexicographically minimize U
tion of its results as well as a means to support sensi-
tivity and postoptimality analysis. Huss and Ignizio’s  {(1   2), ( 10x 1  4x 2)} (5)
contributions in engineering, coupled with the work
Satisfy: x 1  x 2   1  1  100 (6)
of Charnes, Cooper, Ijiri, Jääskeläinen, Gass, Romero,
Tamiz, Jones, Lee, Olson, and others in management  x 2  2   2  4 (7)
science served to motivate the interest in multiple ob-
x  , ,   0 (8)
 jective optimization that continues today.
Goal programming is the most widely  applied  tool The new variables (i.e., the negative and positive de-
of multiple-objective optimization/multicriteria deci-  viation variables, that have been added to the con-
sion making. However, today’s goal programming straints) indicate that a solution to the problem may 
models, methods, and algorithms differ significantly  result, for a given constraint  i , in a negative deviation
from those employed even in the early 1990s. Goal (i   0), or a positive deviation ( i   0), or no devi-
programming, as discussed later, may be combined ation (i    i   0). That is to say that we can under-
 with various tools from the artificial intelligence sec- achieve a goal (be it a hard or soft constraint), over-
tor (most notably genetic algorithms and neural net- achieve it, or precisely satisfy it. In the multiplex
 works) so as to provide an exceptionally robust and formulation, the deviation variables that are to be
powerful means to model, solve, and analyze a host of  minimized have been shown in boldface, as well as ap-
real-world problems. In other words, today’s goal pro- pearing in the first (highest priority) term of the
gramming—while maintaining its role as the “work- achievement function of function (5).
horse” of multiple-objective decision analysis—is a  While this new formulation may appear unusual
much different tool than that described in most text- (at least to those schooled in traditional, single-
books, even those published relatively recently. objective optimization), it provides an accurate
492 Goal Programming

representation of the linear programming problem orig- Maximize z 1  3x 1  x 2
inally posed. To appreciate this, examine the achieve- (profit per time period) (9)
ment function, as represented by formula (5).
The multiplex achievement function is a vector, Maximize z 2  2x 1  3x 2
rather than a scalar as in conventional single-objective (market shares captured per time period) (10)
optimization (e.g., linear programming). The terms Satisfy: 2x 1  x 2  50
in this vector are ordered according to priority. The (raw material limitations) (11)
first term [i.e., 1  2 in function (5)] is reserved for
the unwanted  deviation variables for all rigid  con- x 1  20
straints, or hard goals—restrictions that supposedly  (market saturation level, product 1) (12)
must  be satisfied for the solution to be deemed feasi- x 2  30
ble. Any solution in which this first term takes on a (market saturation level, product 2) (13)
 value of zero is thus—in math programming terms—
a feasible solution. In goal programming, such a so- x  0
lution is deemed “implementable,” inferring that it  (nonnegativity conditions) (14)
could actually be implemented in the real-world prob-
lem under consideration. If the reader cares to graph this problem, he or she
Once the first term has been minimized, the next   will see that there is no way in which to optimize both
term (the second term, or 10x 1  4x 2 in this case) objectives simultaneously—as is the case in virtually 
can be dealt with. The algorithm will seek a solution any nontrivial, real-world problem. However, the pur-
that minimizes the value of this second term, but this pose of goal programming is to find a solution, or so-
must be accomplished without degrading the value al-  lutions, that simultaneously  satisfice  all objectives. But 
ready achieved in the higher priority term.  And this is the first these objectives must be transformed into goals.
manner in which one seeks the lexicographic minimum To transform an objective into a goal, one must assign
of an ordered vector. some estimate  (usually the decision maker’s prelimi-
 Again, this formulation may appear unusual but it  nary estimate) of the aspired level for that goal. Let’s
not only accurately represents the linear programming assume that the aspiration level for profit [i.e., func-
(LP) problem, it also indicates the way in which most  tion (9)] is 50 units while that of market shares [i.e.,
commercial software actually solves LP models. Specifi- function (10)] is 80 units. Consequently the multi-
cally, LP problems are generally solved by the two-phase plex model for the goal programming problem is
simplex algorithm, wherein the first phase attempts to shown below, wherein the two transformed objectives
find a feasible solution and the second seeks an optimal now appear as (soft) goals (19) and (20), respectively:
solution that does not degrade the feasibility achieved
Lexicographically minimize U
in phase 1. Multiplex algorithms simply extend this no-
 {(1   2   3), ( 4   5)} (15)
tion to any number of phases, according to the formu-
lation employed to represent the given problem. Satisfy: 2x 1  x 2   1  1  50
(raw material limitations) (16)

B. Multiplex Form of the x 1   2  2  20

(market saturation level, product 1) (17)
Goal Programming Problem
x 2   3  3  30
 While any single-objective optimization problem can (market saturation level, product 2) (18)
be represented in a manner similar to that described
3x 1  x 2  4   4  50
above, our interest lies in multiple-objective opti-
(profit goal) (19)
mization and, more specifically, in goal programming
(GP). Consequently, let us examine the multiplex 2x 1  3x 2  5   5  80
model for a specific GP problem. (market shares goal) (20)
Consider the illustrative problem represented be-
, ,   0
x 
low, wherein there are two objective functions to be
(nonnegativity conditions) (21)
optimized, functions (9) and (10), and a set of con-
straints to be satisfied. For purposes of discussion, we The multiplex model for the problem indicates—via
assume that the constraints of (11) through (13) are the achievement function—that the first priority is to
hard, or rigid. satisfy the hard goals of (16), (17), and (18). Note
Goal Programming
50
40
A B
30
20 C
D
10
0
0 10 20 30 40
Figure 1 The satisficing region for the example.
that the nonnegativity conditions of (21) will be im-
plicitly satisfied by the algorithm.
Once the deviation variables (i.e., 1  2  3) as-
sociated with those hard goals have been minimized
(albeit not necessarily to a value of zero), the associ-
ated multiplex (or sequential GP) algorithm proceeds
to minimize the unwanted deviations (i.e., 4  5)
associated with the profit and market share goals,
 while not degrading the values of any higher ordered
achievement function terms.
Figure 1 serves to indicate the nature of the goal pro-
gramming problem that now exists. Note that the solid
lines represent the constraints, or hard goals, while the
dashed lines indicate the original objectives—now trans-
formed into soft goals. It is particularly important to
note that those solutions satisficing this problem form a
region, bounded by points A, B, C, and D, and including
all points within and on the edges of the bounded re-
gion. This contrasts with conventional optimization in
 which the optimal solution is most often a single point.
The achievement functions for the linear pro-
gramming and goal programming illustrations posed
previously represent but two possibilities from a large
and growing number of choices. We describe a few of  noncommensurable and thus other forms of the
the more common achievement functions in the next  achievement function are more realistic. One of these
section.
IV. FORMS OF THE ACHIEVEMENT FUNCTION
The three earliest, and still most common forms of 
the Multiplex achievement function are listed here
and discussed in turn:
1. Archimedean (also known as weighted goal
programming)
493
2. Non-Archimedean (also known as lexicographic,
or preemptive goal programming)
3. Chebyshev (also known as fuzzy programming).
A. Archimedean Goal Programming
The achievement function for an Archimedean GP
model consists of exactly two terms. The first term al-
 ways contains all the unwanted deviation variables asso-
ciated with the hard goals (rigid constraints) of the
problem. The second term lists the unwanted deviation
 variables for all soft goals (flexible constraints), each
50  weighted according to importance. Returning to our
previous goal programming formulation, assume that 
the market shares goal [i.e., function (20)] is consid-
ered, by the decision maker, to be twice as important as
the profit goal. Consequently, the Archimedean form
of the achievement function could be written as follows.
Notice carefully that  5 has now been weighted by 2:
Lexicographically minimize U
 {(1   2   3), ( 4  25)} (22)
Note that, as long as the unwanted deviations are min-
imized, we have achieved a satisficing solution. This
may mean that we reach a satisficing solution, for our
example, whether the profit achieved is 50 or more 
units. This infers a one-sided measure of achievement.
If we wish to reward an overachievement of the profit,
then either the model should be modified or we
should employ one of the more recent developments
in achievement function formatting. The latter mat-
ter is discussed in a forthcoming section. For the mo-
ment, however, our assumption is that we simply seek
a solution to the achievement function given in Eq.
(22), one that provides a satisficing result.
Realize that Archimedean, or weighted goal pro-
gramming, makes sense only if you believe that nu-
merical weights can be assigned to the nonachieve-
ment of each soft goal. In many instances, goals are
is the non-Archimedean, or lexicographic achieve-
ment function.
B. Non-Archimedean Goal Programming
The achievement function for a non-Archimedean
GP model consists of two or more terms. As in the case
of the Archimedean form, the first term always con-
tains the unwanted deviation variables for all the hard
goals. After that, the deviation variables for all soft 
494
goals are arranged according to priority—more specif-
ically, a nonpreemptive priority.
To demonstrate, consider the problem previously 
posed as Archimedean GP. Assume that we are unable,
or unwilling, to assign weights to the profit or market  resulting Chebyshev formulation is simply:
share goals. But we are convinced that the capture of 
market share is essential  to the survival of the firm. It  Minimize 
might then make sense to assign a higher priority to
market share than to profit, resulting in the non-
 Archimedean achievement function given here:
Lexicographically minimize U
 {(1   2   3), ( 5), ( 4)} (23)
 While the achievement function of Eq. (23) consists
of but a single deviation variable for the second and
third terms, the reader should understand that sev-
eral deviation variables may appear in a given term—
if you are able to weight each according to its per-
ceived importance.
C. Chebyshev (Fuzzy) Goal Programming
There are numerous forms of Chebyshev, or fuzzy,
goal programming but we restrict our coverage to just  mization problem in which we seek to minimize a sin-
one in this subsection. The notion of Chebyshev GP
is that the solution sought is the one that minimizes
the maximum deviation from any single soft goal. Re-
turning to our profit and market shares model, one
possible transformation is as follows:
Minimize  (24)
Satisfy: 2x 1  x 2  50
(raw material limitations) (25)
x 1  20
(market saturation level, product 1) (26)
x 2  30
(market saturation level, product 2) (27)
  (U 1  z 1)/(U 1  L 1) (28)
  (U 2  z 2)/(U 2  L 2) (29)
, x  0 (30)
 where:
U k   the best possible value for objective k  (e.g., op-
timize the problem without regard to any other ob-
 jectives but objective k )
L k   the worst possible value for objective k  (e.g., op-
timize the problem without regard to objective k )
  a dummy variable representing the worst devia-
tion level
Goal Programming
z k   the value of the function representing the k th ob-
 jective (e.g., z 1  3x 1  x 2 and z 2  2x 1  3x 2).
Given the specific model of (24) through (30), the
Satisfy: 2x 1  x 2  50
(raw material limitations)
x 1  20
(market saturation level, product 1)
x 2  30
(market saturation level, product 2)
  (70  3x 1  x 2)/(70  60)
  (110  2x 1  3x 2)/(110  70)
, x  0
This model may be easily transformed into the multi-
plex form by means of adding the necessary deviation
 variables and forming the associated achievement 
function. However, it is clear that the Chebyshev 
model, as shown, is simply a single objective opti-
gle variable, . In other words, we seek to minimize
the single worst deviation from any one of the prob-
lem goals/constraints.
 While the Archimedean, non-Archimedean, and
Chebyshev forms of the achievement function are the
most common, other, newer versions may offer cer-
tain advantages. As mentioned, these newer forms of 
the achievement function are briefly described in a
later section on extensions of GP.
V. GENERAL FORM OF THE MULTIPLEX MODEL
 Whatever the form of the achievement function, the
multiplex model takes on the following general form:
Lexicographically minimize U
 {c(1)T  v , c(2)T  v , ..., c(K )T  v } (31)
Satisfy: F( v )  b (32)
 v   0 (33)
 where:
K   the total number of terms (e.g., priority levels)
in the achievement function
F( v )  the problem goals, of either linear or nonlin-
ear form, and in which negative and positive devi-
ation variables have been augmented
Goal Programming
b  the right-hand side vector
 v  the vector of all structural (e.g., x  j ) and deviation
(i.e., i  and i ) variables
(k )
c  the vector of coefficients, or weights, of  v in the
k th term of the achievement function
(K )T 
c  the transpose of  c(k ).
 We designate functions (31) through (33) as the
primal form of the muliplex model. Starting with this
primal form, we may easily derive the dual of any goal
programming, or multiplex, model.
VI. THE MULTIDIMENSIONAL DUAL
The real power underlying conventional single-
objective mathematical programming, particularly lin-
ear programming, lies in the fact that there exists a
dual for any conventional mathematical programming
model. For example, the dual of a maximizing LP
model, subject to type I () constraints, is a mini-
mizing LP model, subject to type II () constraints.
The property of duality allows one to exploit, for ex-
ample, LP models so as to develop additional theories
and algorithms, as well as provide a useful economic
interpretation of the dual variables.
One of the alleged drawbacks of goal program-
ming has been the “lack of a dual formulation.” It is
difficult to understand why this myth has endured as
the dual of goal programming problems was devel-
oped, by Ignizio, in the early 1970s. It was later ex-
tended to the more general multiplex model in the
1980s. However, space does not permit any exhaustive
summary of the multidimensional dual (MDD) and
thus we present only a brief description.
 We listed the general form of the multiplex model in
(31) through (33). Simply for sake of discussion, let’s
examine the dual formulation of a strictly  linear  multi-
plex model, taking on the primal form listed below:
PRIMAL: Lexicographically minimize U
 {c(1)T  v , c(2)T  v , ..., c(K )T  v } (34)
Satisfy: Av   b (35)
 v   0 (36)
 where:
K   the total number of terms (e.g., priority levels)
in the achievement function
 v  the vector of all structural (e.g., x  j ) and deviation
(i.e., i  and i ) variables
(k )
c  the vector of coefficients, or weights, of  v in the
k th term of the achievement function
495
 Av   b are the linear constraints and goals of the
problem, as transformed via the introduction of 
negative and positive deviation variables.
If you are familiar with single-objective optimization,
 you may recall that the dual of a linear programming
model is still a linear programming model. However, in
the case of a GP, or multiplex model, its dual—the mul-
tidimensional dual—takes on the form of a model in
 which (1) the “constraints” have multiple, prioritized
right-hand sides and (2) the “objective function” is in
the form of a vector. More specifically, the general form
of the multidimensional dual is given as:
DUAL:
Find  Y  so as to lexicographically maximize  w   bT  Y 
(37)
Subject to: A T  Y  ⇐ c(1), c(2), ..., c(K ) (38)
 Y , the dual variables, are unrestricted
and  multidimensional (39)
Note that the symbol ⇐ indicates the lexicographic na-
ture of the inequalities involved (i.e., the left-hand side
of each function is lexicographically less than or equal to
the multiple right-hand sides). Physically, this means that 
 we first seek a solution subject to the first column of 
right-hand side elements. Next, we find a solution sub-
 ject to the second column of right-hand side elements—
but one that cannot degrade the solution achieved for
the previous column of right-hand side values. We con-
tinue in this manner until a complete set of solutions has
been obtained for all right-hand side values.
The transformation from primal to dual may be
summarized as follows:
• A lexicographically minimized achievement 
function for the primal translates into a set of 
lexicographically ordered right-hand sides for the
dual.
• If the primal is to be lexicographically minimized,
the dual is to be lexicographically maximized.
• For every priority level (achievement function
term) in the primal, there is an associated vector
of dual variables in the dual.
• Each element of the primal achievement function
corresponds to an element in the right-hand side
of the dual.
• The technological coefficients of the dual are the
transpose of the technological coefficients of the
primal.
The development of the MDD for goal programming
(or general multiplex) models leads immediately to
496
both a means for economic interpretation of the dual
 variable matrix as well as supporting algorithms for
solution. A comprehensive summary of both aspects,
as well as numerous illustrative numerical examples,
are provided in the references.
VII. ALGORITHMS FOR SOLUTION
A. Original Approach
 As noted, algorithms exist for the solution of goal
programming (as well as any problem that can be
placed into the multiplex format) problems in either
the primal or dual form. The original emphasis of  portant advantages to the employment of parallel
goal programming was, as discussed, on linear goal
programs, and the GP algorithms derived then (by  the algorithm is supported by parallel processors, the
Charnes and Cooper, and their students) were “mul-
tiphase” simplex algorithms. That is, they were based
on a straightforward extension of the two-phase sim-
plex algorithm.
B. Serial Algorithms: Basic Approach
 Assuming that a serial  algorithm (one in which only a
single solution exists at a given time, as is the case
 with the well-known simplex algorithm for linear pro-
gramming) is used to solve a goal programming prob-
lem, the fundamental steps are as follows:
Step 1. Transform the problem into the multiplex
format.
Step 2. Select a starting solution. [In the case of a lin-
ear model, the starting solution is often the one in
 which all the structural variables (i.e., the x  j ’s) are
set to zero.]
Step 3. Evaluate the present solution (i.e., determine
the achievement function vector).
Step 4. Determine if a termination criterion has been
satisfied. If so, stop the search process. If not, go to
step 5. [Note that, in the case of linear models, the
shadow prices (dual variable values) are used to de-
termine if optimality has been reached.]
Step 5. Explore the local region about the present best  of the present population).
solution to determine the best direction of move-
ment. (For linear models, we move in the direction
indicated by the best single shadow price.)
Step 6. Determine how far to move in the direction of  Step 5. Utilize the genetic algorithm operations of se-
best improvement, and then do so. (In linear mod-
els, this is determined by the so-called “Theta” or
“blocking variable” rule.)
Step 7. Repeat steps 3 through 6 until a termination
rule is satisfied.
Goal Programming
C. Parallel Algorithms: Basic Approach
The references provide details and illustrations of the
use of such serial multiplex algorithms, for linear,
nonlinear, and integer goal programming problems.
However, with the advent of hybrid multiplex algo-
rithms, the use of  parallel  algorithms is both possible
and, in the minds of some, preferable.
 A parallel algorithm follows somewhat the same
steps as listed previously for serial algorithms. The pri-
mary difference is that rather than employing a single
solution point at each step, multiple points—or pop-
ulations of solutions—exist at each iteration (or “gen-
eration”) of the algorithm. There are two very im-
algorithms in goal programming. The first is that, if 
speed of convergence to the final solution is signifi-
cantly increased. The second advantage is less well
known, but quite likely even more significant. Specif-
ically, it would appear from the evidence so far that 
solutions to certain types  (more specifically, those em-
ploying evolutionary operations) of parallel algo-
rithms are far more stable, and less risky, than those
derived by conventional means.
D. Hybrid Algorithm Employing
Genetic Algorithms
The basics of a hybrid goal programming/genetic al-
gorithm for solving multiplex models are listed below.
Such an approach has been found, in actual practice,
to achieve exceptionally stable solutions—and do so
rapidly. Furthermore, it is particularly amenable to
parallel processing.
Step 1. Transform the problem into the multiplex
format.
Step 2. Randomly select a population of trial solutions
(typically 20 to a few hundred initial solutions will
compose the first generation).
Step 3. Evaluate the present solutions (i.e., determine
the achievement function vector for each member
Step 4. Determine if a termination criterion has been
satisfied. If so, stop the search process. If not, go to
step 5.
lection, mating, reproduction, and mutation to de-
 velop the next generation of solutions (see the ref-
erences for details on genetic algorithms).
Step 6. Repeat steps 3 through 5 until a termination
rule is satisfied.
Goal Programming
The parallel algorithm developed via combining goal
programming with genetic algorithms offers a conve-
nient way in which to address any single or multiob-
 jective optimization problem, be they linear, nonlin-
ear, or discrete in nature. The single disadvantage is
that global optimality cannot be ensured.
VIII. GOAL PROGRAMMING
AND UTILITY OPTIMIZATION
 Although goal programming is developed within a
satisficing framework, its different variants can be in-
terpreted from the point of view of utility theory as
highlighted in this section. This type of analysis helps
to clarify goal programming variant as well as to pro-
 vide the foundations of certain extensions of the
achievement function.
Let us start with lexicographic goal programming,
 where the noncompatibility between lexicographic or-
derings and utility functions is well known. To properly  Solutions into Efficient Solutions
assess the effect of this property on the pragmatic value
of this approach, we must understand that the reason for
this noncompatibility is exclusively due to the nonconti-
nuity of preferences underlying lexicographic orderings.
Therefore, a worthwhile matter of discussion would
not be to argue against lexicographic goal program-
ming because it implicitly assumes a noncontinuous
system of preferences but to determine if the charac-
teristics of the problem situation justify or not a system
of continuous preferences. Hence, the possible prob-
lem associated with the use of the lexicographic variant  achievement function of the multiplex formulation
does not lie in its noncompatibility with utility func-
tions, but in the careless use of this approach. In fact,
in contexts where the decision maker’s preferences are
clearly continuous, a model based on nonpreemptive
 weights should be used. Moreover, it is also important  solutions and the edge BD the set of satisficing and ef-
to note that a large number of priority levels can lead
to a solution where every goal, except those situated in
the first two or three priority levels, is redundant. In
this situation, the possible poor performance of the
model is not due to the lack of utility meaning of the
achievement function but to an excessive number of  at the same time. Note that there are more refined
priority levels or to overly optimistic aspiration levels
(i.e., close to the ideal values of the goals).
Regarding weighted (Archimedean) goal pro-
gramming, we know that this option underlies the
maximization of a separable and additive utility func-
tion in the goals considered. Thus, the Archimedean
solution provides the maximum aggregate achieve-
ment among the goals considered. Consequently, it  B. Extensions of the Achievement Function
seems advisable to test the separability between at-
tributes before the decision problem is modeled with
the help of this variant.
497
Regarding Chebyshev goal programming, it is rec-
ognized that to this variant underlies a utility function
 where the maximum (worst) deviation level is mini-
mized. In other words, the Chebyshev option under-
lies the optimization of a MINMAX utility function,
for which the most balanced solution between the
achievement of the different goals is achieved.
These insights are important for the appropriate
selection of the goal programming variant. In fact,
the appropriate variant should not be chosen in a
mechanistic way but in accordance with the decision
maker’s structure of preferences. These results also
give theoretical support to the extensions of the
achievement function to be commented on in the
next section.
I X. E XT EN SI ON S
A. Transforming Satisficing
The satisficing logic underlying goal programming
implies that its formulations may produce solutions
that do not fit classic optimality requirements like ef-
ficiency; that is, solutions for which at least the achieve-
ment of one of the goals can be improved without de-
grading the achievement of the others. However, if 
one wishes, it is very simple to force the goal pro-
gramming approach to produce efficient solutions.
To secure efficiency, it is enough to augment the
 with an additional priority level where the sum of the
 wanted deviation variables is maximized.
For instance, in the example plotted in Fig. 1 the
closed domain ABCD represents the set of satisficing
ficient solutions. Thus, if the achievement function of 
the multiplex model (15) through (21) is augmented
 with the term (4  5) placed in a third priority 
level, then the lexicographic process will produce so-
lution point B, a point that is satisficing and efficient 
methods capable of distinguishing the efficient goals
from the inefficient ones, as well as simple techniques
to restore the efficiency of the goals previously classi-
fied as inefficient. Technical details about this type of 
procedure can be found in the Bibliography.
 According to the arguments developed in Section
 VIII, from a preferential point of view the weighted
498
and the Chebyshev goal programming solutions rep-
resent two opposite poles. Thus, since the weighted
option maximizes the aggregate achievement among
the goals considered, then the results obtained with
this option can be biased against the performance
achieved by one particular goal. On the other hand,
because of the preponderance of just one of the
goals, the Chebyshev model can provide results with
poor aggregate performance between different   words, any marginal change is of equal importance
goals. The extreme character of both solutions can
lead in some cases to possibly unacceptable solutions
by the decision maker. A possible modeling solution
for this type of problem consists of compromising
the aggregate achievement of the Archimedean
model with the MINMAX (balanced) character of   weighted goal programming model or to each prior-
the Chebyshev model. Thus, the example of the ear-
lier section can be reformulated with the help of the
following multiplex extended goal programming
model:
Lexicographically minimize U
 {(1  2   3), [(1  Z )  Z (4   5)]} (40)
Satisfy: 2x 1  x 2   1  1  50
(raw material limitations) (41)
x 1   2  2  20
(market saturation level, product 1) (42)
x 2   3  3  30
(market saturation level, product 2) (43)
3x 1  x 2  4   4  50
(profit goal) (44)
2x 1  3x 2  5   5  80
(market shares goal) (45)
(1  Z ) 4    0 (46)
(1  Z ) 5    0 (47)
, x , ,   0 (48)
 where the parameter Z  weights the importance at-
tached to the minimization of the sum of unwanted
deviation variables. For Z   0, we have a Chebyshev  However, this is not the case. In fact, significant simi-
goal programming model. For Z   1 the result is a
 weighted goal programming model, and for other val-
ues of parameter Z belonging to the interval (0, 1) in-
termediate solutions between the solutions provided
by the two goal programming options are considered.
Hence, through variations in the value of parame-
ter Z , compromises between the solution of the max-
imum aggregate achievement and the MINMAX so-
lution can be obtained. In this sense, this extended
formulation allows for a combination of goal pro-
gramming variants that, in some cases, can reflect a
Goal Programming
decision maker’s actual preferences with more accu-
racy than any single variant.
Other extensions of the achievement function have
been derived taking into account that in all tradi-
tional goal programming formulations there is the
underlying assumption that any unwanted deviation
 with respect to its aspiration level is penalized ac-
cording to a constant marginal penalty. In other
no matter how distant it is from the aspiration level.
This type of formulation only allows for a linear rela-
tionship between the value of the unwanted deviation
and the penalty contribution. This corresponds to the
case of the achievement functions underlying a
ity level of a lexicographic model.
This type of function has been termed a one-sided
penalty function, when only one deviation variable is
unwanted, or V-shaped penalty function when both
deviation variables are unwanted. However, other
penalty function structures have been used. Thus, we
have the two-sided penalty functions when the deci-
sion maker feels satisfied when the achievement of a
given goal lies within a certain aspiration level inter-
 val, or the U-shaped penalty function if the marginal
penalties increase monotonically with respect to the
aspiration levels. Several authors have proposed im-
provements and refinements to the goal program-
ming model with penalty functions. Details about this
type of modeling are described in the Bibliography.
C. Goal Programming and MCDM
Goal programming is but one of several approaches
possible within the broader field of multiple criteria
decision making (MCDM). It is a common practice
 within MCDM to present its different approaches in
an independent, disjoint manner, giving the impres-
sion that each approach is completely autonomous.
larities exist among most of the MCDM methods. In
this sense, the multiplex approach presented in Sec-
tion V is a good example of a goal programming
structure encompassing several single- and multiple-
objective optimization methods.
Furthermore, goal programming can provide a uni-
fying basis for most MCDM models and methods. With
this purpose, extended lexicographic goal program-
ming has recently been proposed. To illustrate this
concept, consider the representation provided in Eqs.
(49) through (52):
Goal Programming
Lexicographically minimize U
 {1 1   1  (i i    i i ) p , ...,  j   j 
i h 1
 j   (i i    i i ) p , ...,
i h  j  tential flexibility inherent to goal programming.
Q  Q    Q   (i  i    i i ) p } (49)
i h Q 
Satisfy:  j  (i i    i i )   j   0
i   h  j   j   {1, ..., Q } (50)
 f i  (x)   i    i   t i  i   {1, ..., q } (51)
, ,   0 x  F (52)
 where  p  is a real number belonging to the interval
[1, ∞) or ∞. Parameters i  and i  are the weights re-
flecting preferential and normalizing purposes attached
to the negative and positive variables of the i th goal, re-
spectively;  j  and  j  are control parameters; and h  j  rep-
resents the index set of goals placed in the  j th priority  cant advances, from a strictly pragmatic perspective,
level. The block of rigid constraints x  F, can be trans-
ferred to an additional first priority level in order to for-
mulate the model within a multiplex format.
If the above structure is considered the  primary  ceptionally stable solutions—the type of solutions
model, then it is easy to demonstrate that an important  needed in most real-world situations.
number of multiple-criteria methods are just secondary 
models  of the extended lexicographic goal program-
ming model. Thus, the following multicriteria meth-
ods can be straightforwardly deduced just by applying
different parameter specifications to the above model:
1. Conventional single-objective mathematical
programming model
2. Nonlinear and linear weighted goal programming
3. Lexicographic linear goal programming
4. Chebyshev goal programming
5. Reference point method
6. Compromise programming (L1 bound) and (L ∞
bound or fuzzy programming with a linear
membership function)
7. Interactive weighted Tchebycheff procedure.
The use of goal programming as a unifying frame-
 work seems interesting at least for the following rea-
sons. The extended lexicographic goal programming
model stresses similarities between MCDM methods
that can help reduce gaps between advocates of differ-
ent approaches. Moreover, this unifying approach can
become a useful teaching tool in the introduction of  placement, Report SID-63. Downey, CA: North American
MCDM, thus avoiding the common presentation based
upon a disjoint system of methods. Finally, the extended
lexicographic goal programming approach allows us to
499
model decision-making problems for which a good rep-
resentation of a decision maker’s preferences requires
a mix of goal programming variants. In short, this type
of general formulation can increase the enormous po-
X . T HE F UT UR E
In the nearly half century since its development, goal
programming has achieved and maintained its repu-
tation as the workhorse of the multiple-objective op-
timization field. This is due to a combination of sim-
plicity of form and practicality of approach. The more
recent extensions to the approach have, thankfully,
not negatively impacted on these attributes.
 We envision a future in which goal programming
 will continue to utilize methods from other sectors,
particularly artificial intelligence. The most signifi-
involve combining goal programming with evolution-
ary search methods, specifically genetic algorithms.
Such extensions permit the rapid development of ex-
SEE ALSO THE FOLLOWING ARTICLES
Decision Support Systems • Evolutionary Algorithms • Exec-
utive Information Systems • Game Theory • Industry, Artifi-
cial Intelligence in • Model Building Process • Neural Net-
 works • Object-Oriented Programming • Strategic Planning
for/of Information Systems • Uncertainty 
BIBLIOGRAPHY
Charnes, A., and Cooper, W. W. (1961). Management models and 
industrial applications of linear programming,  Vols. 1 and 2.
New York: John Wiley.
Charnes, A., and Cooper, W. W. (1977). Goal programming
and multiple objective optimization. Part I.  European Journal 
of Operational Research, Vol. 1, 39–54.
Dauer, J. P., and Kruger, R. J. (1977). An iterative approach to
goal programming. Operational Research Quarterly,  Vol. 28,
671–681.
Gass, S. I. (1986). A process for determining priorities and
 weights for large- scale linear goal programming models.
 Journal of the Operational Research Society, Vol. 37, 779–784.
Ignizio, J. P. (1963). S-II trajectory study and optimum antenna
 Aviation Corporation.
Ignizio, J. P. (1976). Goal programming and extensions. Lexington
Series. Lexington, MA: D. C. Heath & Company.
500 Goal Programming

Ignizio, J. P. (1985). Introduction to linear goal programming. Bev- Simon, H. A. (1956). Rational choice and the structure of the
erly Hills, CA: Sage Publishing. environment. Psychological Review,  Vol. 63, 129–138.
Ignizio, J. P., and Cavalier, T. M. (1994). Linear programming. Tamiz, M., and Jones, D. (1996). Goal programming and Pareto
Upper Saddle River, NJ: Prentice Hall. efficiency.  Journal of Information and Optimization Sciences, Vol.
Markowski, C. A., and Ignizio, J. P. (1983). Theory and prop- 17, 291–307.
erties of the lexicographic LGP dual. Large Scale Systems, Vol. Tamiz, M., Jones, D. F., and Romero, C. (1998). Goal pro-
5, 115–121. gramming for decision making: An Overview of the current 
Romero, C. (1991). Handbook of critical issues in goal program-  state-of-the-art.  European Journal of Operational Research, Vol.
ming. Oxford, UK: Pergamon Press. 111, 569–581.
Romero, C. (2001). Extended lexicographic goal program-  Vitoriano, B., and Romero, C. (1999). Extended interval goal
ming: A unifying approach. OMEGA, International Journal of  programming.  Journal of the Operational Research Society, Vol.
Management Science,  Vol. 29, 63–71. 50, 1280–1283.