Advances in Environmental Science and Energy Planning
Developing Expert Systems in Decision Making by Applying the Fuzzy
Set Theory
OTILIJA SEDLAK, ZORAN CIRIC, SOFIJA ADZIC
Faculty of Economics Subotica
University of Novi Sad,
Subotica, Segedinski put 9-11
SERBIA
[email protected], [email protected], [email protected]
http://www.ef.uns.ac.rs
Abstract: - The main goal of this paper was to present the knowledge based fuzzy model that will provide more
successful and efficient decision making in the area of marketing, in relation with dilemma “to produce or to
buy”. Namely, authors tried to develop a model that will be suitable for making marketing decisions in
production systems. Methodology in the paper obtained analysis of the theory of marketing and knowledge
based systems, the development of the specific model for decision making, so as the application of developed
model on one case from domestic production system. The research problem in this paper was proposed in terms
of model development. Authors developed model for decision making, based on successful integration of
marketing and knowledge based and fuzzy theories. Also, this was proposed as universal model which can be
implemented in each production system. They implemented the model in decision making problem related to
the debate “to produce or to buy” on one real decision problem in domestic production system. The paper was
divided in three parts. In the first part authors analyzed the main theoretical developments in the area of
marketing and decision making, fuzzy and knowledge based systems. The second part was dedicated to the
methodology and thus the development of the model for marketing decision making. The third part presented
the model and specific decision making problem that has been solved by developed model. At the end, authors
pointed the advantages of the model but also some limitations and possibilities for the future researches.
Key-Words: - knowledge-based systems, fuzzy logic, decision making process, uncertainty
decision making on global or local marketing
engagement, choice of target markets, methods of
entering target markets, combining marketing mix
elements and implementing, i.e. controlling
marketing activities.
Marketing decisions in modern-day business are
made under conditions of growing uncertainty,
which cannot be measured, and business risks,
which are measurable. High uncertainty and risk
levels result from disruptive innovations and great
unexpected shocks [14]. Marketing decisions have
always been made under the conditions of
uncertainty and risk, but modern-day pace of change
and intensity of shock are more extreme than ever.
For this reason, market decision makers should be
qualified for appropriate assessment of acceptable
risk level, so as to secure the best effects and control
the damage from made marketing decision.
Applying the fuzzy set theory is a good example of
such approach to marketing decision making.
In most cases, marketing decisions are made
under conditions of uncertainty and high risk.
1 Introduction
In order to expound on the application of the fuzzy
set theory in marketing decision making properly, it
is necessary to start from the essence of marketing
decision making. Marketing decision making is an
integral part of marketing management, which is, in
a sense, the art and science of applying core
marketing concepts to choose target markets and
get, keep, and grow customers through creating,
delivering, and communicating superior customer
value [13].
Marketing decision makers include marketers,
marketing managers, buyers, shopping centres,
influential individuals, users, analysts, and also
formal or informal authorities featuring in the
process of choosing marketing alternatives. They
feature both on the supply and demand side, given
that a social and managerial process by which
individuals and groups obtain what they need and
want through creating and exchanging products and
values with others [12]. Besdies, marketing decision
making process is more often analysed from the
supply aspect in professional literature, covering
ISBN: 978-1-61804-280-4
41
Advances in Environmental Science and Energy Planning
According to some authors [15]; [26], this is most
contributed to by:
• a relatively high number of relevant variables;
• impossibility
of
controlling
relevant
variables;
• their instability and nonlinearity, stochasticity
of relevant variables;
• difficult quantification and measurement of
effects of relevant variables;
• shortage of marketing information.
The decision maker’s expertise and appropriate
assessment of tolerable risk levels (i.e. the
subjective factor) is therefore of extreme importance
for the final effects of decision made.
The main goal of this paper was to present the
knowledge based fuzzy model that will provide more
successful and efficient decision making in the area
of marketing, in relation with dilemma “to produce
or to buy”, because the make-or-buy methodology is
one of the most critical strategic decisions within
logistics outsourcing and should be taken in a
structured and consistent manner [2]; [4]. Namely,
authors tried to develop a model that will be suitable
for making marketing decisions in production
systems. Methodology in the paper obtained
analysis of the theory of marketing and knowledge
based systems, the development of the specific
model for decision making, so as the application of
developed model on one case from domestic
production system. The research problem in this
paper was proposed in terms of model development.
Authors developed model for decision making,
based on successful integration of marketing and
knowledge based and fuzzy theories. Also, this was
proposed as universal model which can be
implemented in each production system. They
implemented the model in decision making problem
related to the debate “to produce or to buy” on one
real decision problem in domestic production
system.
The paper was divided in three parts. In the first
part authors analyzed the main theoretical
developments in the area of marketing and decision
making, fuzzy and knowledge based systems. The
second part was dedicated to the methodology and
thus the development of the model for marketing
decision making. The third part presented the model
and specific decision making problem that has been
solved by developed model. At the end, authors
pointed the advantages of the model but also some
limitations and possibilities for the future
researches.
ISBN: 978-1-61804-280-4
2 Problem Formulation
Making marketing decisions is aimed in two basic
directions, i.e. towards solving problems on the one
hand and using the benefits of business
opportunities on the other. At any rate, marketing
decision making enables a transition from the
current state into a new, i.e. desired state, which
marketing decision makers deem to be more
favourable for the company.
The well-known opinion is that different
problems “call” for different types of decision
making. To a great extent, types of marketing
decisions also determine methods applied in making
those decisions. Furthermore, marketing decision
makers must be qualified for precise problem
definition, primarily in terms of certainty or
uncertainty of its outcome.
Theoretically, marketing decision making
process is structured differently, but the key phases
of this process that can be singled out are the
diagnostic phase or problem identification, action
plan which implies defining the alternative avenues
of undertaking marketing activities and choosing
marketing aleternatives, and also implementation,
including marketing control and revision. Marketing
theory displays a great similarity of the decision
making proces to the problem solving process,
which nas even led some authors to equate their
meaning.
When viewing the areas of decision making in
marketing, decision making in the business
community and practice is often related to the issues
of creating an optimum marketing program, i.e.
making decisions on the product, developing and
launching a new product, pricing, marketing
channels and marketing communication. In all
instances of marketing decision making, companies
that apply marketing orientation in business
operations produce effects manifested in the
operation of all business functions.
The classical decision making operates with a set
of alternatives over the space of decision, a set of
states off affairs over the space of the state of
affairs, relation pointing to states or outcomes
expected from each of the alternative actions, and
finally, the utility or the goal function, which
arranges these outcomes in accordance with the
desired outcome. Decision making is said to be
performed under conditions of certainty, when the
outcomes of any action can be precisely determined
and arranged. In such cases, alternatives are chosen
that lead to outcomes with maximum utility. On the
other hand, a decision is made under conditions of
risk, when the only knowledge available regarding
42
Advances in Environmental Science and Energy Planning
from food production sector [25]; [24],
pharmaceutical production [33], bicycle derailleur
and freewheel production [23], temporary agency
work [1], etc. Solving this problem demands
meticulous approach and engaging experts of
various profiles, where the marketing expert figures
as the coordinator of all activities.
Resolving the given problem can be approached
along three avenues:
(a) by the classical method;
(b) by applying the expert system and
(c) by applying the fuzzy set theory.
The classical method is usually related to the
transaction costs. The transaction cost theory of the
firm introduced by [7] has been made a standard
framework for the study of institutional
arrangements by [30]; [31]. The Coasian framework
helps explain not only the existence of the firm, but
also its size and scope. Some firms are highly
integrated. Others are much more specialized. Why
do some firms choose a vertically integrated
structure, while others specialize in one stage of
production and outsource the remaining stages to
other firms? In other words, should firm make its
own inputs, should it buy them on the spot market,
or should in maintain an ongoing relationship with a
particular supplier? Traditionally, economists
viewed vertical integration or vertical control as an
attempt to earn monopoly rents by gaining control
of input markets or distribution channels. The
transaction cost approach, by contrast, emphasizes
that vertical coordination can be an efficient means
of protecting relationship – specific investments or
mitigating other potential conflicts under incomplete
contracting. As transaction cost economics was
developed in the 1970’s and 1980’s, a stream of
empirical literature emerged explaining the “makeor-buy decision” using transaction cost reasoning,
where the influence of transaction costs on decisions
to make or buy components was assessed indirectly
through the effects of supplier market competition
and two types of uncertainty, volume and
technological [29].
The traditional dichotomous make and buy
sourcing strategies have been established in the
transaction cost economics (TCE) literature [23].
The basic characteristic of the classical approach to
resolving a business dilemma is almost exclusive
reliance on comparing the economic effects of both
decisions. In other words, the total costs related to
producing the given part/subset/product are
compared with the total cost of purchasing the same
item (provided that that there are suppliers where
the given part/subset/product can be purchased), and
then the decision maker opts for the more cost-
the state of outcomes is their probability
distribution.
In any real-life situation, including the decision
making and management fields – whether it is about
setting goals and formulating strategies, or selecting,
implementing and monitoring the selected strategy –
many processes are unfit for mathematical
modelling. In normal real-life situations in various
areas of human activity, very often we do not know
the equations for most non-linear processes and
systems and therefore resort to approximation.
Fuzzy systems approximate those equations. Fuzzy
systems enable us to make optimum approximations
of the non-linear universe. If it is possible to build a
mathematical model, we shall use it. There is always
a question: how can we know that mathematical
approximation corresponds to a process in reality?
Fuzzy systems enable us to model the universe in
linguistic terms, rather than forcing us to write a
mathematical model of the universe. The technical
term for it is model-free function approximation
[26]. Here is important to emphasize that for many
knowledge-intensive applications, it is important to
develop an environment that permits flexible
modelling and fuzzy querying of complex data and
knowledge including uncertainty [16].
A very common problem encountered by
marketing decision makers in companies, also
related to production, finance and purchase, is the
dilemma whether a certain product, part or semimanufactured product should be produced by
engaging their own resources, at their own plants, or
obtained from suppliers. This problem requires
coordinated approach to solving, and can manifest
itself in practice in three forms:
• produce or purchase a product previously not
used by the company;
• independently produce the product previously
purchased from suppliers or
• purchase the product currently produced by
engaging their own resources at their own
plants.
The “make-or-buy decision” is about the choice
of weather to carry out a particular process or
activity within a business or to buy it in from a
supplier. In recent decades, the “make-or-buy
decision” shifted from the level of reactive clerical
function to the center of business strategy. It was
realized that sourcing decision have a great effect on
business strategy and that sourcing decision has a
great effect on a business future survival. The make
or buy decision can often be a major determinant of
profitability making a significant contribution to the
financial health of the company [19]. This dilemma
was explored in past years in different systems,
ISBN: 978-1-61804-280-4
43
Advances in Environmental Science and Energy Planning
economic, political development, science and
technological progress”.
Since the make or buy decision is presented in
wide range of researches in the world, there have
been made several attempts for creation the
framework for “make or buy” decision making
process [6]; [9]; [5]. Based on these researches
authors proposed the methodology for development
of one knowledge based framework for decision
making in the mentioned area. Fig. 1 (Appendix 1)
effective option. Only in rare cases (depending on
the decision maker’s expertise) are additional
criteria taken into account as well [27].
Outsourcing is purchasing from someone else a
product or service that have been previously
manufactured internally. It is becoming an
increasingly important element in strategic decisionmaking and an important way to increase efficiency
and often quality. The sourcing decision can cover a
wide range of goods and services ranging from
manufactured
components,
semi-processed
materials [8] and constancy services. The outcome
of these decisions is influenced by a considerable
number of factors and the soundness of
assumptions. Traditionally, cost had been viewed as
the main consideration in sourcing decision [26];
[17]. Also, one more factor which is important in
the process of sourcing decision making is the level
of technology development. For example, on the
basis of 1990–2002 panel data set on Spanish
companies and an exogenous proxy for
technological change it have been provided thee
causal evidence that technological change increases
the likelihood of outsourcing [3].
Almost unknown to the academic world, and to
the general public, the application of intelligent
knowledge-based systems is rapidly and effectively
changing the future of the human species. Today,
human well-being is, as it has been for all of history,
fundamentally limited by the size of the world
economic product. Thus, if human economic wellbeing (which I personally define as the bottom
centile annual per capita income) is ever soon to
reach an acceptable level (e.g., the equivalent of
$20,000 per capita per annum in 2004), then
intelligent knowledge-based systems must be
employed in vast quantities. This is primarily
because of the reality that few humans live in
efficient societies (such as the United States,
Canada, Japan, the UK, France, and Germany, for
example) and that inefficient societies, many of
which are already large, and growing larger, may
require many decades to become efficient. In the
meantime, billions of people will continue to suffer
economic impoverishment-an impoverishment that
inefficient human labour cannot remedy. To create
the extra economic output so urgently needed, we
have only one choice: to employ intelligent
knowledge-based systems in great numbers, which
will produce economic output prodigiously, but will
consume hardly at all. According [20] “the creation
and development of the knowledge based society
and knowledge economy are perceived as one of the
most important priorities of the modern society and
its lifestyle development, as well as of social,
ISBN: 978-1-61804-280-4
3 Problem Solution
It is due to the importance of additional criteria that
a prototype of the expert system was developed
(including software solutions). As the economic
model in Fig. 2 (Appendix 2) shows, the number of
attributes taken into account is far greater compared
to the classical decision making method.
The decision maker communicates with the
expert system by choosing the domains of values for
the attribute of the decision making leaf nodes.
Using the built-in logarithm, the expert system
proposes a solution with explanation based on these
data. The risk of making a wrong decision is thus
multiply reduced, which was the basic objective of
developing the expert system [10].
Fuzzy systems enable us to make optimum
approximations of the non-linear universe. If it is
possible to build a mathematical model, we shall use
it. Fuzzy systems enable us to model the universe in
linguistic terms, rather than forcing us to write a
mathematical model of the universe. The technical
term for it is model-free function approximation.
The Fuzzy Approximation Theorem claims that a
graph can always be covered with a finite number of
fuzzy patches. The more uncertain the rule, the
larger the fuzzy patch. According to the Fuzzy
Approximation Theorem, a fuzzy system can
approximate a continuous system to a sufficient
degree of accuracy. This includes almost all systems
studied by science. Fuzzy systems can model
dynamic systems changing over time [22].
Viewed geometrically, every portion of human
knowledge. A fuzzy system is a large set of fuzzy
“if … then” rules, representing “a large set of
patches”. The more knowledge, the more rules. The
more rules, the more sets. If the rules are more
indefinite, i.e. uncertain, the patches are larger. If
the rules are more definite, the patches are smaller.
If the rules are so precise that they are not fuzzy,
then patches are reduced to points.
The Fuzzy Approximation Theorem says more
than that. Theoretically, all equations can be
translated into rule patches. Fuzzy systems
44
Advances in Environmental Science and Energy Planning
was approached. Input and output variables were
defined as different fuzzy sets, both continuous and
discontinuous. Defining fuzzy sets intrinsically
implies that the appropriate degree of belonging is
ascribed to all possible values, followed by defining
inference rules, which mandatorily includes the
operation of logical implication with input
conjunction and/or disjunction. Having determined
the degree of belonging, values of output variables,
with minimum but certain and guaranteed degrees
of belonging, are chosen for every rule, considering
all the input variables. After this, the observed
output variable is still not unequivocally defined, for
it has different values with different degrees of
belonging at all points, depending on the number of
used rules. The next step is choosing the output
variable values with maximum degrees of
belonging. An output variable is thus obtained and
subsequently defuzzified, i.e. a specific numeric
value is determined, which is either the input
variable on the next, higher level, or the decision,
i.e. solution to the problem, on the last level. Thus,
the final decision is made by using multiple steps,
taking into account all possible circumstances and
all the possible variable values occurring in it. Such
complete consideration and problem analysis would
be impossible to do without fuzzy logic, relying
only on experts’ knowledge, experience and
intuition.
The following is the overview of fuzzy variables
with possible value domains:
approximate systems in physics, communication,
physiology, etc. Fuzzy systems can be applied
wherever the brain is used.
It is hard to deny that modern-day knowledge is
fuzzy. Meanings of statements are undoubtedly
fuzzy. Knowledge has always been regarded in
terms of rules. If knowledge is fuzzy, then rules are
fuzzy as well. Fuzzy rules connect fuzzy sets. Fuzzy
patches cover the system graph. It is the Fuzzy
Approximation Theorem and fuzzy patches that
explain the functioning of fuzzy systems.
A solution to the given problem by applying
fuzzy set theory, the economic model (Fig. 2;
Appendix 2) can be supplemented with an oriented
graph. Input variables occur at the first and second
levels. The second level already contains four
functions of two arguments; the third level also
contains two functions of two arguments, but some
of the arguments at the previous are level defined
functions, i.e. complex functions. The fourth level is
a function of two arguments defined at the third
level, i.e. the obtained function at the fourth level is
multiply complex, and its value is the solution to the
problem. (If required, this model can be extended to
more or reduced to fewer levels.) However, these
are not functions and their arguments in the classical
mathematical sense. The input variables (arguments)
are linguistic variables (“high raw material costs”,
“medium quantity of product”, “aggressive
suppliers’ approach”, “status quo in the forthcoming
3-5 years, but development over the subsequent ten
plus years”, etc. The same applies to output
variables (functions), representing input functions
for complex functions up to the last level, such as
production costs, purchase price, foreseen
technological change, production limitations etc.
Such nature of input and output variables, i.e. value
domains that cannot be accurately quantified, make
the fuzzy set theory extremely convenient for
problem solving in marketing decision making.
From a standpoint of computational theory of
perception of "computing with words", this method
begins by converting the natural perceptions of how
usual or unusual the possible states are into
probability estimates on a ratio scale, and converting
the natural perceptions of how acceptable or
unacceptable the possible outcomes of state-action
pairs are into utility numbers on an interval scale
using Von Neuman-Morgenstern (1947) utility
theory or some similar method.

















Fuzzy logic is very often successfully applied for
modelling problems where interdependences
between individual variables are highly complex.
This is how resolving the above described problem
ISBN: 978-1-61804-280-4

45
DECISION: produce, purchase;
PRODUCTION LIMITATIONS: enormous,
substantial, negligible, none;
ECONOMIC EFFECTS: significant, favourable,
unfavourable, devastating;
EXTERNAL FACTORS: favourable, neutral,
unfavourable;
EXPERTISE: high, average, low;
CAPACITY UTILISATION: high, average, low;
PRODUCTION COSTS: high, average, low;
PURCHASE PRICE: high, moderate, low;
TECHNOLOGICAL CHANGES: confining, tolerable,
negligible;
SUPPLIERS’ APPROACH: aggressive, moderate,
passive;
QUANTITY OF PRODUCT: large, medium, small;
PRODUCTION SCHEDULE: weekly, monthly,
quarterly, semi-annual;
RAW MATERIAL COSTS: high, moderate, low;
LABOUR COSTS: high, moderate, low;
SUPPLIER RELATIONS: excellent, good, poor
TRANSPORT COSTS: high, moderate, low;
LONG-TERM CHANGE: development, status quo,
elimination;
MID-TERM CHANGE: development, status quo,
elimination.
Advances in Environmental Science and Energy Planning
EXPERTISE Λ
According to their belonging functions, the above
fuzzy variables in the observed problem can be
observed as follows:
 Group I: raw material costs, labour costs,
production costs, transport costs, purchase
price, quantity of product, capacity utilisation,
external factors (Fig. 3). The value interval
(a,b) is determined in each specific case. The
observed variables are measured in monetary
units, excluding capacity utilisation, foreseen
technological change and external factors
expressed in percentage.
high
high
high
average
average
<high
low
low
CAPACITY
UTILISATION
=>
high
average
low
high
average
low
high
<high
PRODUCTION
LIMITATIONS
PURCHASE
PRICE =>
high
<high
high
moderate
low
high
moderate
low
ECONOMIC EFFECTS
SUPPLIERS’
APPROACH =>
>passive
passive
aggressive
moderate
passive
aggressive
moderate
EXTERNAL
FACTORS
unfavourable
neutral
unfavourable
neutral
favourable
neutral
favourable
none
negligible
substantial
negligible
substantial
enormous
substantial
enormous
b)
PRODUCTION
COSTS Λ
high
high
average
average
average
low
low
low
unfavourable
devastating
favourable
unfavourable
devastating
significant
favourable
unfavourable
c)
TECHNOLOGICAL
CHANGES Λ
confining
confining
tolerable
tolerable
confining
negligible
negligible
Fig. 3 Group I
Table 1
Step 1
Step 3
(a)
QUANTITY
PRODUCT Λ
large
large
large
medium
< large
< large
low
OF
PRODUCTION
SCHEDULE =>
< quarterly
quarterly
semi-annual
weekly
monthly
> monthly
weekly
CAPACITY
UTILISATION
high
average
low
high
average
low
average
LABOUR
COSTS =>
> low
low
high
average
low
high
average
PRODUCTION COSTS
TRANSPORT
COSTS =>
high
<high
high
average
low
average
low
PURCHASE PRICE
MID-TERM
CHANGE =>
development
status quo
status quo
development
<development
TECHNOLOGICAL
CHANGES
negligible
negligible
tolerable
tolerable
confining
PRODUCTION
LIMITATIONS Λ
EXTERNAL
FACTORS
=>
DECISION
enormous
significant
favourable
produce
enormous
significant
<favourable purchase
enormous
<significant
*
purchase
substantial
significant
>unfavourable produce
substantial
significant
unfavourable purchase
substantial
favourable
favourable
produce
substantial
favourable
<favourable purchase
substantial
<favourable
*
purchase
negligible
>unfavourable
*
produce
negligible
unfavourable
favourable
produce
negligible
unfavourable
<favourable purchase
negligible
devastating
*
purchase
none
>unfavourable
*
produce
none
unfavourable
>unfavourable produce
none
unfavourable
unfavourable purchase
none
devastating
favourable
produce
none
devastating
<favourable purchase
The asterisk (*) marks that the attribute can take on any value from
the domain in this rule.
(b)
RAW MATERIAL
COSTS Λ
high
high
average
average
< high
low
low
high
average
high
average
low
average
low
(c)
SUPPLIER
RELATIONS Λ
excellent
excellent
<excellent
good
good
poor
poor
moderate
low
high
moderate
low
high
moderate

ECONOMIC
EFFECTS =>
Group II: long term change, mid-term change
(Fig. 4). The measurement unit is the degree of
expected change.
(d)
LONG-TERM
CHANGE Λ
>elimination
development
status quo
elimination
elimination
Fig. 4 Group II

Step 2
a)
ISBN: 978-1-61804-280-4
46
Group III: production schedule (Fig. 5). The
measurement unit is time expressed in months.
Advances in Environmental Science and Energy Planning
Fig. 9 Group VII
Fig. 5 Group III

Group IV: relations with suppliers, suppliers’
strategy (Fig. 6). The measurement unit is
assessment grade for suppliers.
Fig. 6 Group IV

Group V: expertise (Fig. 7). Presence of higher
qualification is expressed in percentage.
Fig. 7 Group V

Group VI: production limitations, economic
effects (Fig. 8); measured in percentage, from
the “produce” aspect.
Fig. 8 Group VI

Group VII: decision (Fig. 9); expressed in
percentage, from the “produce” aspect.
Fig. 10 Decision making with defuzzification
ISBN: 978-1-61804-280-4
47
Advances in Environmental Science and Energy Planning
can be considered more fully and exactly applying
fuzzy sets.
The fuzzy set theory provides a solution to this.
Individual steps of the algorithm are based on fuzzy
logic rules. The algorithm was verified on multiple
numeric examples, and development of an
appropriate software product is in progress. The
essential feature of this algorithm is that there is no
possibility of “setting up” a desirable decision by
tailoring input data.
The approximate reasoning algorithm for making
the business decision whether to produce or
purchase consists of three basic steps. Each basic
step is divided into several interconnected steps.
Each of these substeps, or steps, contains a logical
inference rule defined by experts. As specified
numeric values directly entered at the given moment
of making the “purchase or produce” decision, the
input data are as follows: expertise, quantity of
product, production schedule, raw material costs,
labour costs, supplier relations, transport costs, and
long- and mid-term changes. As illustrated in
Figures 3-9 above, fuzzy sets are already prepared at
all levels. Direct entry of the above listed first- and
second-level data activated the above defined
algorithm. The algorithm is illustrated in graph 1(b)
of Fig. 10. Production costs are defuzzified by
determining the centre of gravity using the known
formula. Thus obtained numerical data is the input
value for step 2(b).
As specified numeric values directly entered at
the given moment of making the “purchase or
produce” decision, the input data are as follows:
expertise, quantity of product, production schedule,
raw material costs, labour costs, supplier relations,
transport costs, and long- and mid-term changes. As
illustrated in Figures 3-9 above, fuzzy sets are
already prepared at all levels. Direct entry of the
above listed first- and second-level data activated
the above defined algorithm. The algorithm is
illustrated in graph 1(b) of Fig. 10. Production costs
are defuzzified by determining the centre of gravity
using the known formula. Thus obtained numerical
data is the input value for step 2(b).
References:
[1] Alewell, D., & Hauff, S. Make-or-buy
decisions regarding temporary agency work–
an empirical analysis of the decision process
and expected effects, The International
Journal of Human Resource Management,
22(15), 2011, pp. 3127-3145.
[2] Bajec, P., Jaomin, I., A Make-or-Buy Decision
Process for Outsourcing, Promet-Traffic &
Transportation, 22 (4), 2010, pp. 285-291.
[3] Bartel, A. P., Lach, S., & Sicherman, N.,
Technological change and the make-or-buy
decision, Journal of Law, Economics, and
Organization,
1-28,
2012,
doi:10.1093/jleo/ews035
[4] Bartkus, E.V., Jurevičius, V., Production
Outsourcing in the International Market,
Engineering Economics, ISSN 1392-278,
2007, No 1, pp. 59-68.
[5] Cánez, L. E., Platts, K. W., & Probert, D. R.,
Developing a framework for make-or-buy
decisions, International Journal of Operations
& Production Management, 20(11), 2000, pp.
1313-1330.
[6] Cheshmberah,
M.,
Makui,
A.,
&
Seyedhoseini, S., A new fuzzy MCDA
framework for make-or-buy decisions: A case
study of aerospace industry, Management
Science Letters, 1(3), 2011, pp. 323-330.
[7] Coase, R. H., The Nature of the Firm,
Economica, 4(16), 1937, pp 386-405.
[8] Deshpande, V., Schwarz, L. B., Atallah, M. J.,
Blanton, M., & Frikken, K. B., Outsourcing
Manufacturing:
Secure
Price‐Masking
Mechanisms for Purchasing Component Parts,
Production and Operations Management,
20(2), 2011, pp. 165-180.
[9] Hsiao, H. I., Van Der Vorst, J. G. A. J., Kemp,
R. G. M., & Omta, S. O., Developing a
decision-making framework for levels of
logistics outsourcing in food supply chain
networks, International Journal of Physical
4 Conclusion
This article has developed a heuristic diagram for
making the “purchase or produce” business
decision. The problem per se has all the
characteristics of uncertainty. A serious problem,
therefore, is determining input variables and
forecasting all possible circumstances, which is only
possible based on subjective estimate. An estimate
with a degree of accuracy is much easier to obtain
for a range of data than for an individual value.
Fuzzy sets can be introduced into the existing
decision making models in several ways. As an
economic institution, a company bases its existence
on the environment, both from the aspect of
providing input and from the aspect of achieving
and valorising input. Miscellaneous knowledge and
experience, and also decision making in the areas of
investment, market operations, financial function,
production function or research and development,
ISBN: 978-1-61804-280-4
48
Advances in Environmental Science and Energy Planning
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
Distribution & Logistics Management, 40(5),
2010, pp. 395-414.
Humphreys, P., McIvor, R., Huang, G., An
expert system for evaluating the make or buy
decision,
Computers
&
Industrial
Engineering, 42 (2-4), 2002, pp. 567-585.
Kosko, B., Neural Networks and Fuzzy
Systems – A Dynamical Systems Approach to
Machine Intelligence, Englewood Cliffs:
Prentice – Hall, New Jersey, 1993.
Kosko, B., Neural Networks and Fuzzy
Systems, Englewood Cliffs,: Prentice – Hall,
New Jersey, 1992.
Kotler, P., Marketing management: The
millennium edition. Upper Saddle River, NJ:
Prentice-Hall, 2000.
Kotler, P., & Caslione, J. A., Chaotics: the
business of managing and marketing in the
age of turbulence. New York: AMACOM Div
American Mgmt Assn, 2009.
Kotler, P., Vong, V., Sonders, J., &
Armstrong, G., Principi marketinga (4th
European ed.), Beograd: Mate, 2007.
Koyuncu, M., & Yazici, A., A fuzzy
knowledge-based system for intelligent
retrieval, IEEE Transactions on Fuzzy
Systems, 13(3), 2005, pp. 317-330.
Maltz, A., Carter, J. R., & Maltz, E., How
managers make sourcing decisions about low
cost regions: Insights from perceptual
mapping, Industrial Marketing Management,
40(5), 2011, pp. 796-804.
McIvor, R.T., The influence of capability
considerations on the outsourcing decision: the
case
of
a
manufacturing
company,
International Journal of Production Research,
48(17), 2010, pp. 5031-5052.
McIvor, R. T., & Humphreys, P. K., A casebased reasoning approach to the make or buy
decision, Integrated Manufacturing Systems,
11(5), 2000, pp. 295-310.
Melnikas, B., Knowledge Economy: Synergy
Effects, Interinstitutional Interaction and
Internationalization Processes, Iniznerine
Ekonomika – Engineering Economics, 22(4),
2011, pp. 367-379.
Moses, A., Cross-functional make or buy
decision process ownership, Management
Research Review, 34(9), 2011, pp. 1042-1060.
Munakata, T. & Jany Y., Fuzzy Systems: an
overview, Comm. of The ACM, 37(3), 1994,
pp. 69-76.
ISBN: 978-1-61804-280-4
[23] Park, J. K., & Ro, Y. K., The impact of a
firm's make, pseudo-make, or buy strategy on
product performance, Journal of Operations
Management, 29(4), 2011, pp. 289 -304.
[24] Pascucci, S., Royer, A., & Bijman, J., Should I
make or should I buy? Innovation strategies
and governance structures in the Italian food
sector, International Journal on Food System
Dynamics, 2(2), 2011, pp. 167-180.
[25] Pascucci, S., Royer, A., & Bijman, J. “To
make or to buy”, is this the question? Testing
making or buying decisions to explain
innovation-sourcing strategies in the food
sector, International Food and Agribusiness
Management Review, 15(3), 2012, pp. 99-118.
[26] Sedlak, O., Ćirić, Z., & Ćirić, I., Strategic
Management Under the Conditions of
Uncertainty and Indefiniteness, Strategic
Management, 18(1), 2013, pp. 62-68.
[27] Skowron-Grabowska, B., Selection criteria for
make or buy decision process in enterprises,
E-Manufacturing: Business Paradigms and
Supporting Technologies, 18th International
Conference on CAD/CAM, Robotics, and
Factories of the Future (CARs & FOF), Porto,
Portugal, 2002, pp. 195-202.
[28] Virvilaite, R., Seinauskiene, B., Sestokiene,
G., The Link between Standardization/
Adaptation of International Marketing
Strategy
and
Company
Performance,
Inzinerine
Ekonomika
Engineering
Economics, 22(1), 2011, pp. 106-117.
[29] Walker, G., & Weber, D., A transaction cost
approach
to
make-or-buy
decisions,
Administrative Science Quarterly, 29, 1984,
pp. 373-391.
[30] Williamson, O. E., Markets and Hierarchies:
Analysis and Antitrust Implications, New
York: Free Press, 1975.
[31] Williamson,
O.
E.,
Transaction-cost
economics: the governance of contractual
relations, Journal of Law and Economics,
22(2), 1979, pp. 233-261.
[32] Williamson, O. E. & Ghani, T., Transaction
cost economics and its uses in marketing,
Journal of the Academy of Marketing Science,
40(1), 2012, pp. 74-85.
[33] Zhao, G., & Chen, W., A stochastic model for
pharmaceutical R&D project management in a
make-or-buy
decision
setting,
IEEE
Transactions on Engineering Management,
57(3), 2010, pp. 488-501.
49
Advances in Environmental Science and Energy Planning
Appendix 1
Fig. 1 Framework for Make or Buy decision making
Source: [5] p.1322
Appendix 2
DECISION
PRODUCE
PURCHASE
•
•
ECONOMIC
EFFECTS
PRODUCTION
LIMITATIONS
HIGH
AVERAGE
LOW
•
•
•
HIGH
AVERAGE
LOW
•
•
•
QUANTITY OF
PRODUCT
•
•
•
SUBSTANTIAL
FAVOURABLE
UNFAVOURABLE
DEVASTATING
PRODUCTION
COSTS
CAPACITY
UTILISATION
EXPERTISE &
WORKMANSHIP
QUALITY
•
•
•
•
•
•
•
ENORMOUS
SUBSTANTIAL
NEGLIGIBLE
NONE
•
•
•
•
•
•
•
•
WEEKLY
MONTNLY
QUARTERLY
SEMI-ANNUAL
RAW MATERIAL
COSTS
•
•
•
HIGH
AVERAGE
LOW
•
•
•
PURCHASE PRICE
HIGH
AVERAGE
LOW
•
•
•
HIGH
MODERATE
LOW
•
•
•
EXCELLENT
GOOD
POOR
HIGH
AVERAGE
LOW
SUPPLIERS'
APPROACH
•
•
•
CONFINING
TOLERABLE
NEGLIGIBLE
•
•
•
HIGH
AVERAGE
LOW
LONG TERM
CHANGE
•
•
•
DEVELOPMENT
STATUS QUO
ELIMINATION
MID-TERM CHANGE
•
•
•
DEVELOPMENT
STATUS QUO
ELIMINATION
Fig. 2 Decision tree for problem solving: "Produce or purchase"
ISBN: 978-1-61804-280-4
50
AGGRESSIVE
MODERATE
PASSIVE
TRANSPORT
COSTS
LABOUR COSTS
•
•
•
FAVOURABLE
NEUTRAL
UNFAVOURABLE
FORESEEN
TECHNOLOGICAL
CHANGES
•
•
•
SUPPLIER
RELATIONS
PRODUCTION
SCHEDULE
LARGE
MEDIUM
SMALL
EXTERNAL
FACTORS