THINK FUZZY SYSTEM: DEVELOPING NEW PRICING STRATEGY METHODS FOR CONSUMER GOODS USING FUZZY LOGIC

International Journal of Fuzzy Logic Systems (IJFLS) Vol.7, No.1, January 2017
DOI : 10.5121/ijfls.2017.7101 1
THINK FUZZY SYSTEM: DEVELOPING NEW
PRICING STRATEGY METHODS FOR
CONSUMER GOODS USING FUZZY LOGIC
Antonio Morim*
, Eduardo Sá Fortes*
, Paulo Reis+
, Carlos Cosenza+
, Francisco
Doria+
, Armando Gonçalves+
*
Production Engineering Master’s Program, Federal University of Rio de Janeiro,
Instituto Alberto Luiz Coimbra de Pós–Graduação e Pesquisa de Engenharia ,COPPE,
Bloco G, Ilha do Fundão, 21945-970, Rio de Janeiro - Brasil
ABSTRACT
The main purpose of this article is to present and explore potential applications in marketing administration
related to pricingstrategyusingfuzzylogic. Considering the new trends in consumer behavior in Brazil’s
economy and the consistent growth of C and D social classes an application was developed by the authors
to better understand and adjust pricing strategies: The Think Fuzzy System that combines fuzzy logic
(COPPE Cosenza Model), and some other related strategic concepts, supported by mathematical
microeconomic modeling, utility factor, indifference curves and an experiential hierarchic clustering
model.
KEYWORDS:
Consumer behavior, think fuzzy system, fuzzy logic, pricing methods, microeconomic mathematical
models
1 INTRODUCTION
The recent scenario in Brazil’s economy, with no stability and a continuous economic recession,
brought a new challenge related to C and D social classes: consumers with lessmoney to spend,
adopting new habits.
The biggest challenge is to understand if past consumer behaviors and habits can be used to
position and sell new consumer goods to these social classes in Brazil (DÓRIA, 2009).
The utility conceptdefines that, by observing a consumer’s choices whether he or she was better
off after a change in prices. The consumer revealed by choices what his or her preferences were-
hence the term “revealed preferences”. (SAMUELSON 1997)
Marketing administration more and more try to improve new methods of analyzing consumer
behavior, and new experiential aspects are taken into consideration. Researchers today’s
emphasis, are in studying the importance of consumer emotional answers as main aspects of their
attitudes.(SOLOMON, 2002)

2
The macro consumer behavior is created by social issues but to reach the factors of micro
consumer behavior, so called individual factors, the following indicators have to be considered:
cultural anthropology, history, demography, semiotic/literary criticism, macroeconomics,
sociology, social psychology, microeconomics, clinical psychology and experimental psychology
( SOLOMON et al.2006).
Also in a marketer’s viewpoint, the psychology means that how, when and where people spend
their own time and money, and also, it can show the effects of purchase on the identity,
personality and lifestyle of customer as psychological variables. So, market segmentation should
consider psychological variables (SARLI and TAT, 2011).
Also at one big challenge for the companies is to understand price elasticity of demand, because
is crucial to figure out how changes in prices and demand can be related. Considering these new
emerging social classes, should the companies adjust their actual products to these new customers
in terms of kind of promotions and offers or should these companies develop specific solutions
even creating new products supported by their strong brands.
Also in this context, is important to evaluate, if we can use the Utility Function, as a reference to
consumer purchase understanding.
Consumers make purchase seven in times of economic recession, but depending on the size of the
crisis as we have been facing in Brazil nowadays, for sure some factors have to be taken into
consideration. For example, customers will not by products at higher prices if they understand
that these products have not quality enough.
Changes of the social or economic situation can make changes in the consumer’s purchase
behavior. These purchases could be more purposeful, responsible and taking price into the most
important aspect to consider.
In addition, marketing strategy should be able to be sensitive enough to understand the
importance of capturing consumers’ preferences that could affect products depending on their
origin (ABIDIN,2016).
We could also into consideration, aspects related to local products that in certain moments can
create images that could support consumer decision connecting some country specific values
(SARLI,2011).
Technology products can have different patterns regarding purchase decision including for
example lifestyle (CHANG,2015).
As a great deal of consumers has new budget lines, we can understand that new price elasticity
demand curves can be created.
Also, considering that these new budget lines and differentdemand numbers can be generated.
But what is going to happen if we increase production to answer to the market needs and do not
keep prices in the same levels or change price strategy?(NAGLE, 1987), is there a possibility of
raising the prices and the demand keep on growing among consumers that prefer this soft drink
brand to others? How to prove or assure insome degree of certainty a new price elasticity demand
curve?

International Journal of Fuzzy Logic Systems (IJFLS)
II - PRICING DECISIONS USING FUZZY LOGIC
As marketing actually intensifies its studies related to
influence purchase decisions by customers, some different studies
looking for behavioral reactions of consumers appears (V
1999) refers to the emotional decisi
2015) explores a fuzzy logic approach to product quality and purchase intention of consumers in
two wheelers. (ALAMGIR, 20
(GANIDEH,2011) express his ideas questioning if fuzzy logic can predict consumer ethnocentric
tendencies. (NAEINI,2016) propose a specific research to understand
prioritizing lifestyles in shopping centers
of intentionsthat considers a social
III – Fuzzy Decision System
As we show in Figure 1, we describe our fuzzy decision system.
IV – Fuzzy Logic
Our every-day language is notably imprecise. We avoid absolutes and communicate by giving our
audience a direction, a sense of what we mean. But it works! When we say: "This bald man is a
frequent smoker," everyone understands: "Bald" refers to a quantity
"balding" and "completely hairless"; "frequent smoker" is less than "chain smoker" and more than
"occasional smoker." To better communicate our thoughts, we often confuse things a bit further:
Saying a man is "bald" or a "frequent
most likely, be "quite bald" and a "rather frequent smoker."
PRICING DECISIONS USING FUZZY LOGIC
As marketing actually intensifies its studies related to understand how attitudes and culture can
influence purchase decisions by customers, some different studies also using fuzzy logic
looking for behavioral reactions of consumers appears (VALASKOVÁ, 2015) and
1999) refers to the emotional decisions consumers can take when deciding purchases.
2010), analyses brand names influence in the consumerchoice
express his ideas questioning if fuzzy logic can predict consumer ethnocentric
propose a specific research to understand if there is a form of
prioritizing lifestyles in shopping centers, using fuzzy logic. (SLATEVA, 2011) proposes a model
sthat considers a social CRM system.
Fuzzy Decision System
As we show in Figure 1, we describe our fuzzy decision system.
Figure 1 –Fuzzy Decision System
day language is notably imprecise. We avoid absolutes and communicate by giving our
frequent smoker," everyone understands: "Bald" refers to a quantity of hair somewhere between
Saying a man is "bald" or a "frequent smoker" seems too absolute a characterization. He will,
most likely, be "quite bald" and a "rather frequent smoker."
Vol.7, No.1, January 2017
3
understand how attitudes and culture can
also using fuzzy logic and
and (BAGOOZI,
ons consumers can take when deciding purchases. (YOGI,
consumerchoice.
express his ideas questioning if fuzzy logic can predict consumer ethnocentric
if there is a form of
(SLATEVA, 2011) proposes a model
day language is notably imprecise. We avoid absolutes and communicate by giving our
of hair somewhere between
smoker" seems too absolute a characterization. He will,

4
The vagueness of qualifiers "quite" and "rather "does not hinder communication. Imprecise
language can best reflect human reality. Yet, when marketing began using computers, we were
forced to precisely define the terms we use in order to make them acceptable to the computer.
Thus a "frequent cigarette smoker" is defined, let's say, as one who smokes one pack or more per
day, and a "bald man," as one having 5,000 hairs or less. In this precise world, the poor man who
loses just one of his 5,001 hairs becomes instantly bald.
The one who regularly smokes 19 cigarettes’ every day is not a frequent smoker.
A new mathematical tool is now available, which could help change all that: Introducing FUZZY
LOGIC. Created by Professor LoftiZadeha former chairman of the electrical engineering and
computer sciences department at the University of California at Berkeley.
To start with, a Fuzzy Logic definition of a frequent smoker would be "smokes about one pack or
more a day." Then, rather than categorizing a given smoker as "frequent," the Fuzzy Logician
would allocate it a coefficient (between 0 and 1) to indicate how well the smoker can be
identified with the "frequent smokers" set. With 19 cigarettes per day, the coefficient could be
.75; with 25 cigarettes, .95, and so on. (ZADEH, 1975)
In other words, Fuzzy Logic reflects the way most of us speak, allowing for all the qualifiers one
may use when talking about someone as a frequent smoker. When saying, "he is a rather frequent
smoker," we do not give "rather" a meaning related to a certain number of cigarettes. We give our
impression of how well the subject belongs to the group of "frequent smokers." We use "rather"
as a fuzzy coefficient.
This theory has applicationsto marketing and advertising. It can enable the analysis of consumer
answers, as imprecise as they naturally come. It should put an end to the "women 18 to 49" target
audience definition, which has no reason for being in fuzzy language (they'd probably become
"young and middle-aged women"). It should improve our analysis of consumer emotions, where
the language is often purposefully rendered vague with a plethora of qualifiers: In Fuzzy Logic,
"extremely" can multiply "very" and one can take the square root of "few" or solve a problem like
"if most student are rather frequent smokers, how many smoke frequently?"
The answer is "most" multiplied by "rather", of course.
So we can understand that in instead of using just probabilistic aspects related to qualitative and
quantitative surveys to these customers and markets, we can also usea possibilistic logic
approach, reorganizing these consumers in different types of clusters, considering behaviors and
attitudes.
Triangular Fuzzy Numbers are a better way to demonstrate that fuzzy/possibility is about the
range within the theoretical range but beyond observations.
Geometric figures like a triangle can demonstrate intervals in a better way. Intervals are numbers
whose values are not know with certainty but about which bounds can be established. (Figures
2,3 e 4.(Klir,1995).

5
Figure 2 – Triangular fuzzy figure
Figure 3 – interval demonstrations
In applications it is often convenient to work with triangular fuzzy numbers because of their
computational simplicity, and they are useful in promoting representation and information
processing in a fuzzy environment.
V – DEFUZZIFICATION
Among the many methods that have been proposed in the literature in recent years, seven are
described here for defuzzifying fuzzy output functions.
1. Max membership principle:Also known as the heightmethod, this scheme is limited to peaked
output functions. This method is given by the algebraic expression
Where z* is the defuzzified value, and is show graphically in Figure 4
0 1 2 3 4
0
0.5
1
Robustness
Triangular fuzzy
numbers are robust
characterizations
d = [0.3, 1.7, 3]
e = [ 0.4, 1, 1.5]
f = [ 0.8, 6, 10]
g = [ 0.2, 2, 5]
h = [ 0.6, 3, 6]
0
0 1 2 3 4 5 6 7
1
X
Possibility
X
-20-10 0 10 20 30 40
0
1
Xde/(h+g)-fe
Possibility
de
h + g
X - fe

6
Figure 4 Max membership defuzzification method
2. Centroid method:This procedure (also called center of area, center of gravity) is the most
prevalent and physically appealing of all the defuzzification methods; it is given by the algebraic
expression
Where ∫ denotes an algebraic integration. This method is shown in Figure 5.
Figure 5 Centroid defuzzification method
3. Weighted average method:The weighted average method is the most frequently used in fuzzy
applications since it is one of the more computationally efficient methods. It is given by the
algebraic expression
Where, ∑denotes the algebraic sum and where is the centroid of each symmetric membership
function. This method is shown in Figure 6. The weighted average method,is formed by
weighting each membership function in the output by its respective maximum membership value.
As an example, the two functions shown in Figure 6 would result in the following general form
for the defuzzified value:
z =
µC(z) · zdz
µC(z) dz
1
zz*
µ
z =
µC(z) · z
µC(z)
z
z =
a(0.5) + b(0.9)
0.5 + 0.9

7
Figure 6 Weighted Average method of defuzzification
Since the method is limited to symmetrical membership functions, the values a and b are the
means (centroids) of their respective shapes.
4. Mean max membership:This method (also called middle-of-maxima) is closely related to the
first method, except that the locations of the maximum membership can be no unique (i.e., the
maximum membership can be a plateau rather than a single point). This method is given by the
expression;
Where, points a, and b are as defined in Figure7.
Figure 7 Mean max membership defuzzification method
5. Center of sums:This is faster than many defuzzification methods that are presently in use, and
the method is not restricted to symmetric membership functions. This process involves the
algebraic sum of individual output fuzzy sets, say , , instead of their union. Two drawbacks
to this method are that the intersecting areas are added twice, and the method also involves
finding the centroids of the individual membership functions. The defuzzified value is given by
the following equation:
where the symbol is the distance to the centroid of each of the respective membership
functions.
µ
0 a b z
0.5
0.9
1
z =
a + b
2
0 a z* zb
1
µ
C1 C2
z
z =
Z
z
n
k= 1
µCk (z) dz
z
n
k= 1
µCk
(z) dz
z

8
This method is similar to the weighted average method, except in the center of sums method the
weights are the areas of the respective membership functions whereas in the weighted average
method the weights are individual membership values. Figure 8 is an illustration of the center of
sums method.
FIGURE 8 Center of sums method: (a) first membership function; (b) second membership function; and (c)
defuzzification step
6. Center of largest area:If the output fuzzy set has atleast two convex subregions, then the center
of gravity (i.e., is calculated using the centroid method, Eq. of the convex fuzzy subregion with
the largest area is used to obtain the defuzzified value of the output. This is shown graphically
in Figure 8 and given algebraically here:
Where, is the convex subregion that has the largest area marking up as . This condition
applies in the case when the overall output is nonconvex; and in the case when is convex,
µ
0 2 4 6 8 10 z
C1
(a)
0.5
1.0
~
µ
0 2 4 6 8 10 z
(b)
0.5
1.0
C2~
µ
0 2 4 6 8 10 z
z*
(c)
0.5
1.0
z
z
z =
µCm
(z)zdz
µCm
(z) dz
Cm
applies in
Ck
Ck Ck z

9
is the same quantity as determined by the centroid method or the center of largest area method
(because then there is only one convex region).
Figure 9 Center for largest area method (outlined with bold lines), shown for a nonconvex
7. First (or last) of maxima:This method uses the overall output or union of all individual output
fuzzy sets to determine the smallest value of the domain with maximized membership degree in.
The equations for are as follows.
First, the largest height in the union (denote hgt) is determined,
Then the first of the maxima is found,
An alternative to this method is called the last of maxima, and it is given by
In the above Equations the supremum (sup) is the least upper bound and the infimum (inf) is the
greatest lower bound. Graphically, this method is shown in Figure 10, where, in the case
illustrated in the figure, the first max is also the last max and, because it is a distinct max, is also
the mean max.
Figure 10 First of Max(and last of Max) method
Ck
hgt(Ck) = sup
z∈Z
µC∼
k (z)

10
VI –CONSUMER CHOICE AND DECISION MAKING: PROBLEM
RECOGNITION.
One model of consumer decision-making involves several steps. First, the problem recognition—
the way a person realizes that something is not as it should be. One example, your car is getting
more difficult to start and is not accelerating well. The second step is information search—what
are some alternative ways of solving the problem?The third step involves evaluation
ofalternatives Finally, we have the purchase stage, and sometimes a post-purchase stage
Consumers may go back and forth between the stages.
(Kotler, 2012), consumer involvementwill tend to vary dramatically depending on the type of
product. In general, consumer involvement will be higher for products that are very expensive
(e.g., a home, a car) or are highly significant in the consumer’s life in some other way (e.g., a
word processing program or acne medication).
It is important to consider the consumer’s motivation for buying products. To achieve this goal,
we can use the Means-End chain, wherein we consider a logical progression of consequences of
product use that eventually lead to desired end benefit. Thus, for example, a consumer may see
that a car has a large engine, leading to fast acceleration, leading to a feeling of performance,
leading to a feeling of power, which ultimately improves the consumer’s self-esteem.
Information search and decision-making
Consumers engage in both internal and externalinformation search. Internalsearch involves the
consumer identifying alternatives from his or her memory. For certain low involvement products,
it is very important that marketing programs achieve “top of mind” awareness. For example, few
people will search webfor fast food restaurants; thus, the consumer must be able to retrieve one’s
restaurant from memory before it will be considered.
For high involvement products, consumers are more likely to use an externalsearch. Before
buying a car, for example, the consumer may ask friends’ opinions, read reviews in Consumer
Reports, consult several web sites, and visit several dealerships. Thus, firms that make products
that are selected predominantly through external search must invest in having information
available to the consumer in need—e.g., through brochures, web sites, or news coverage.
A compensatory decision involves the consumer “trading off” good and bad attributes of a
product. Occasionally, a decision will involve a non-compensatorystrategy. For example, a
parent may reject all soft drinks that contain artificial sweeteners. Here, other good features such
as taste and low calories cannotovercome this one “non-negotiable” attribute.
The amount of effort a consumer puts into searching depends on a number of factors such as the
market (how many competitors are there, and how great are differences between brands expected
to be?), product characteristics (how important is this product? How complex is the product?
How obvious are indications of quality?), consumer characteristics (how interested is a consumer,
generally, in analyzing product characteristics and making the best possible deal?), and
situationalcharacteristics (as previously discussed).
A number of factors involve consumer choices. In some cases, consumers will be more
motivated. Some consumers are also more motivated to comparison shop for the best prices,

11
while others are more convenience oriented. Personality impacts decisions. Some like variety
more than others, and some are more receptive to stimulation and excitement in trying new
stores. Perception influences decisions. Some people, for example, can taste the difference
between generic and name brand foods while many cannot. Some consumers are put off by
perceived risk, others will tend to change their behavior through learning, they will avoid for
example products that are not sustainable.
VII- MARKET DEMAND ELASTICITY
The profit maximization (P) implies that marginal revenue should equal marginal cost (MC),
which in turn implies:
P − MC1 P = −E (1)
Ed, is the firm’s price elasticity of demand. Note that this equation can be rewritten as:
P= MC (2) 1 + (1/Ed)
If the firm is a monopolist, then the relevant elasticity is the market elasticity of demand, which I
will denote by Ed. Obtaining an estimate of this market elasticity of demand may or may not be
difficult, but for the moment, let’s assume that we have such an estimate. Given this estimate of
Ed, how can we obtain an estimate of the firm’s price elasticity of demand, Ed.
If we assume that the firms in the market compete using Cournot model (which may or may not
be a reasonable assumption), obtaining the elasticity of demand for an individual firm is fairly
straightforward. Suppose that there are n equal-sized firms in the market, and that they all have
the same marginal cost, c. In the Cournot model, each firm chooses its profit- maximizing output
taking the outputs of its competitors as fixed.
We know that in many industries firms use price as a strategic variable, rather than quantity. The
use of price rather than quantity as a strategic variable intensifies competition and reduces profits.
This means that the effective elasticity of demand for a particular brand is likely to be larger in
magnitude than the Cournot model would predict. Howmuch larger? That depends on the cross-
price elasticities among the different brands. (Pindyck, 2009)
If the cross-price elasticities are large (for example the brands are fairly close substitutes), the
Cournot analysis won’t work: Ed is likely to be much larger than nEd. If, on the other hand,
substitutability among brands is limited, the Cournot result will likely still hold.
Remember that the Cournot model is essentially static; it assumes that each firm takes the output
of its competitors as fixed. As explained above, this assumption is quite reasonable for the large
number of markets where competition is stable and dynamic price wars are generally avoided.
This kind of stable pricing could arise if the same firms having been competing for a long time
(and there is little prospect of entry by new firms), production cost and market demand conditions
are relatively stable, and the firms face a repeated Prisoners’ Dilemmawith no end point (and thus
little or no likelihood of “unravelling”). But the Cournot assumption will not be reasonable for
markets where pricing and output decisions are dominated by dynamic gaming considerations.
Airlines. One example is airlines, where very low short-run marginal costs result in intense price
competition, particularly for certain fare categories. As you have seen, airline pricing is also

12
complicated by the fact that prices are linked to yield management (i.e., the allocation, which
changes from day to day, of the number of seats for each fare category). What can an airline do to
avoid the intense price competition that is so prevalent in the industry? As we have discussed, one
strategy is seek and maintain “monopoly routes,” where it is the only airline to offer non-stop
service.
Price competition becomes intense well before that date as stores face the unhappy prospect of
being stuck with large amounts of unsold inventories that they will have to dump, often at below
cost. Over the years, managers of large retail chains and department stores have become
increasingly sophisticated in their understanding of game theory. What has this done for them?
The result is that they are more aware of the unravelling problem, and they are aware that their
competitors are also aware of the problem. Thus they anticipate that the unravelling will start
earlier.
Not wanting to be the “sucker” who gets undercut and winds up with lots of unsold inventories,
each storetries to get a jump on the unravelling. And as you would expect, the unravelling starts
earlier. Ten years ago the unravelling generally began on or just after Thanksgiving, five years
ago in mid-November, and in more recent years unravelling has started by the beginning of
November. If you ran a retail store chain, how should you go about setting prices? First of all,
you are stuck in a bad place and deserve a lot of sympathy! Beyond that, the best you can do try
to predict when and how the unravelling is likely to occur, make sure you are not on vacation or
asleep when it starts, and take it into account when you determine the quantities that you will
purchase at wholesale.
Soft Drink Industry
A nice thing about industries such as beer, soft drinks, breakfast cereals, airlines, auto-mobiles,
etc., is that they have been around for a long time, and a good deal of data are available so that we
can estimate market demand curves. Often, however, firms must set the prices of new products
for which there is little or no history of demand or consumer response to price changes.
VIII – A PRACTICAL APPLICATION
We took as an example to study the initial aspects related to attitudes and behaviors of
consumers when consuming five different types of, soft drinks, juices and water produced by a
well-known brand in Brazil.
We applied the two methods:
The normal statistic approach conducting first a qualitative survey and after that a quantitative
survey taking all the parameters that are recommended in terms of amostral size, trust factor,
standard deviation, but also with the same consumers we conducted a new type of survey that
considers fuzzy aspects and a experimental hierarchic clustering model.
For the fuzzy aspects we applied the COPPE Cosenza Model, developed at Centre for Higher
Studies in Fuzzy Logic (LABFUZZY) at Universidade Federal do Rio de Janeiro (UFRJ)
(COSENZA, 1981).

13
The hierarchic clustering model applied is a technique that serves to combine objects to create
new groups, it can be used to group variables, objects or people. The inputs are any valid
measures of correlations between these objects, also the numbers of clusters or the level of
clustering can be the inputwe can start with all objects in one cluster and divide and subdivide
then until all objects are in their own single-object cluster applying the Think Fuzzy System to
this of clustering model we can achieve different sub clusters (a cluster inside a cluster) based on
the distance between the group centroid (the centroid is the point whose coordinates are the icons
of all the observation in the specific cluster) (AAKER, 2009).
The consumers where asked to say their opinions, related to 9 important attributes of this soft
drink.
The attributes were: price, quality, brand, health habit, tasty, practical, package, promotion
campaigns andlife of the product concerns.
The explanations below will be related only to the price aspects studied, but nevertheless the
figures 10 and 11 will demonstrate the opportunities found when comparing crispy and fuzzy
results in all attributes that we had contemplated in the survey.
In the first survey, that we called crispy research, we asked to the consumers to give from 1 to 5,
considering 5 and the biggest importance and considering 1 as the lowest importance to each
attribute detailed above.
In the second survey, that we called fuzzy research, we asked to the consumers, to give 5
concepts as detailed below, related to the importance of each attribute:
Very important, important, indifferent,less relevant and irrelevant.
We have done 660 individual surveys crispy and fuzzy with the same persons. The trust factor
utilized was 95%, with 4% error margin. The actual Rio de Janeiro population, according to
IBGE (Instituto Brasileiro de Geografia e Estatística), official Institute to this subject in Brazil, is
almost 16 million people and 53% are in social classes C and D, primary target of this survey.
Survey Aspects – Price
We asked consumers that have the soft drink brand as their first preference what would be their
opinion if there had to be a 10% increase in final prices to them, if this increase was or was not
relevant enough to force them to reduce their purchase of the soft drink.
We collected data in the crispy form, with possible answers from 1 to 5, being 1 the lowest one or
and 5 the highest score.
With the same customers we collected data in the fuzzy form with linguistic expressions from
irrelevant to very important as described above.
Initially we organize them in 5 customer behavioral clusters depending on their evaluation to this
question, from 1 to 5. Taking a deep look in the fuzzy survey final results we have concluded that
part of the customers that had answered the crispy survey with a grade 3, had to be considered as
significant to the decision of raising prices because with linguistic answers they demonstrated
some relevancy to the topic.

14
So from now on to next analysis we will consider six behavioral clusters, considering the cluster
with the answer indifferent, in two definitions: indifferent and less indifferent.
VIII – Crispy Survey Results - Price
PRICE NOT
RELV
LESS
RELV
INDIF IMP VERY
IMP
TOTAL 10 60 290 193 117
Table 1 – Amount of survey answers by type
IX- Fuzzification,Defuzzification
INDIF IMPORT VERY
IMP
1 1 0 0
2 1 0 0
3 1 1 0
4 0 1 1
5 0 0 1
Table 2 – Pertinence for a classic set
INDIF IMPORT VERY
IMP
1 1 0 0
2 0,6 0 0
3 0,1 1 0,5
4 0 0,5 1
5 0 0,3 1
Table 3 - Pertinence for a fuzzy set Fuzzy Triangular Numbers
µ= min(µa,µb,µc); µ= max(µa,µb,µc)
a=indifferent; b=important; c=very important
µtri(x;a,b,c) = max ( min (x-a/b-a, c-x/c-b ),0),for a < b < c.
Indifferent=(1,0),(2,0.6),(3,0.1),
(4,0),(5,0)
Important=(1,0),(2,0),(3,1),(4,0.5),(5,0.3)
Very Important=(1,0),(2,0),(3,0.5)
(4,1),(5,1)
Min=(1,0),(2,0),(3,0.5),(4,0.5),(5,0.3)
Max=(1,0),(3,0),(3,1),(4,1),(5,1)
DeffuzificationChoice – Weighted Average Method
µ(output(u)=µ(a)∇µ(b) (u)
(3x0+3,5X0,3+4x0,5+4,5X0)/(0,3+0,5)= 3,66

15
So, 3,66 is the crisp number after deffuzification.
In figures 10 and 11 you can see and compare price results as well the other attributes.
Figure 10 – crispy survey
Figure 11 – fuzzy survey
In figure 10 we can see the fuzzy results, compared to the results in figure 11 related to the crispy
survey. Notice thatprice had a significant difference. This difference shows in this particular
analysis, that more consumers really understands that an increase of 10% or more, is very
important, also just looking for the other attributes we can understand that there is plenty of space
to a more detailed analysis.
X – CONCLUSION
As a preliminary analysis we can see that the use of fuzzy logic supporting marketing decisions
related to consumer habits and attitudes is important and can amplify the way the consumer can
be understood. We entitled this method as Think Fuzzy System. As mentioned before, we applied
it to all attributes referred in the survey.
Also looking for the micro economic aspects, like price strategy, and decisions related to
elasticity price on demand, fuzzy logic can be used to look for some opportunities.
0
1
2
3
4
5
price
Practical
Life of the
product…
Brand
Health
Habit
Tasty
Quality
Promotio
n…
Package
0
1
2
3
4
5
price
Practical
Life of the
product…
Brand
Health
Habit
Tasty
Quality
Promotio
n…
Package

16
The example of the positive or negative impact of price elasticity in strategic decisions was to
turn tangible the density in which this hybrid tool can be worked. Through the same approach
(COPPE Cosenza Model + Think Fuzzy) it would be possible to monitor, identify and delineate
critical issues, fundamental to the understanding of how community needs are being met or how
certain infrastructural issues are being neglected, such as sustainability for example.
The most import is for sure demonstrate that nowadays when customer decisions are been taken
at web speed and the moments of truth multiply their strength,corporations need more than just
probabilistic tools to understand in depth how consumer attitudes can support better branding
strategy and also improve consumer loyalty to their products and services depending on how they
demonstrate credibility concerning their actions.
Looking for possible future research using our proposed Think Fuzzy System, we will conduct a
research related to electronic commerce and predictive pricing strategy.Alsoa research related to
purchases that keep its traditional aspects regarding been done by new consumer generations.
X – BIBLIOGRAPHY
[1] ALAMGIR, M., et al.Influence of brand name on consumer decision making process - an empirical
study on car buyers (2010). Ann. “Ştefancel Mare” Univ. Suceava. Fascicle Fac. Econ. Public
Admin., 10 (2) (2010), pp. 142–153;
[2] BAGOOZI, et al. (1999). The role of emotions in marketing. Journal of Academic Marketing Science,
27 (2) (1999), pp. 184–206
[3] BIDIN, NUZARARHIA et al (2016);7th. International Economics & Business Management
Conference;
[4] AAKER, David A.(2009), Marketing Research; Wiley & Sons, New York;
[5] CHANG .C.,(2015). A hybrid decision-making model for factors influencing the purchase intentions
of technology products: the moderating effect of lifestyle. Behaviour & Information Technology .
Volume 34, Issue 12;
[6] COSENZA, Carlos A. (2011), Notas de aula disciplina - Introdução à Lógica Fuzzy – COPPE –
UFRJ; Rio de Janeiro;
[7] COSENZA, Carlos A. (1981), - "A Industrial Location Model"- Working paper, Martin Centre for
Architectural and Urban Studies, Cambridge University
[8] DORIA, Francisco, A. (2011) – Notas de aula disciplina- Limites Computacionais e Modelagem
Matemática – COPPE – UFRJ; Rio de Janeiro;
[9] DORIA, Francisco A. & COSENZA, Carlos A. (2009), Crise na Economia. Editora Revan, Rio de
janeiro;
[10] GANIDEH, S. et al, 2011. Can Fuzzy Logic Predict Consumer Ethnocentric Tendencies? An
Empirical Analysis in Jordan. Journal of Physical Science and Application, 100-106;
[11] HENDERSON, James M. & QUANDT, Richard E. (1968), Teoria Microeconomica – Uma
Abordagem Matemática. Biblioteca Pioneira de Ciências Sociais, São Paulo;
[12] KOTLER P. (2012), Marketing Management, New Jersey, Simon & Schuster Co.
[13] LAISoon, W., et al, (2013). Hybrid vehicle adoption - a conceptual study. J. Educ. Vocat. Res. 4 (6),
165e168.
[14] KLIR, George J.(1995)Fuzzy Sets and Fuzzy Logic: theory and applications. Prentice Hall, New
Jersey;
[15] NAEINI, A. et al,(2016), Prioritizing Lifestyles in Shopping Centers, Using Fuzzy Logic Inference
System ( Case Study: Shopping Centers in Zanjan). Intal Management Journal 67-76;
[16] PINDYCK, Robert S. (2009), Microeconomics. Prentice Hall,New Jersey;
[17] SAMUELSON, Paul A. (1997), Fundamentos da Análise, Editora Nova Cultural. São Paulo;
[18] SARLI, A. &Tat, H(2011). Attracting Consumers by Finding out Their Psychographic Traits,
International Journal of Fundamental Psychology & Social Sciences, Vol 1, No.1, pp.6-10

17
[19] SOLOMON, Michael R. (2002), Consumer Behaviour: buying, having, and being. Prentice Hall,
New Jersey;
[20] SOLOMON, Michael R.. et al. (2006). Consumer behavior: A Europeanperspective. England:
Pearson Education Limited. 731 p.
[21] NAGLE, T., (1987). The Strategy & Tactics of Pricing : A Guide to Profitable Decision Making, New
Jersey. Prentice Hall.
[22] VALÁSKOVÁ, K., & KLIESTIK, T., (2015). Behavioral Reactions of Consumers To Economic
Recession. Journal Business: Theory and Practice, Slovakia;
[23] YOGI K.,(2015). An Empirical and Fuzzy Logic Approach to Product Quality and Purchase Intention
of Customers in two Whelers, Pacific Science Review B: Humanities and Social Sciences, 57 -69;
[24] ZADEH, L.A. (1975), "The concept of a linguistic variable and its application to approximate
reasoning", Parts 1 and 2, Information Sciences 8,199-249; 301-357;
[25] ZLATEVA, P. et al., A Model of Intention to Purchase as a Component of Social CRM System, 2011
International Conference on E-business, Management and Economics IPEDR Vol.25 (2011) © (2011)
IACSIT Press, Singapore;

THINK FUZZY SYSTEM: DEVELOPING NEW PRICING STRATEGY METHODS FOR CONSUMER GOODS USING FUZZY LOGIC

More Related Content

Similar to THINK FUZZY SYSTEM: DEVELOPING NEW PRICING STRATEGY METHODS FOR CONSUMER GOODS USING FUZZY LOGIC (20)

Recently uploaded (20)

THINK FUZZY SYSTEM: DEVELOPING NEW PRICING STRATEGY METHODS FOR CONSUMER GOODS USING FUZZY LOGIC