You are on page 1of 15

Int. J. Modelling in Operations Management, Vol. X, No.

Y, xxxx

Measuring brand preferences of smartphone using


TOPSIS
Sushma Kumari, Ashruti Bajpai and
Rachita Dewan
Indian Institute of Information Technology,
Devghat, Jhalwa, Allahabad,
Uttar Pradesh 211012, India
E-mail: sushma945@gmail.com
E-mail: ashruti.bajpai25@gmail.com
E-mail: rachitadewan21@gmail.com

Pravin Kumar*
Delhi Technological University,
Bawana Road, Delhi-110042, India
E-mail: pravin.papers@gmail.com
*Corresponding author
Abstract: The purpose of this research paper is to demonstrate the method
for observing the customer preference of selected smartphone in Indian
market using technique for order preference by similarity to an ideal solution
(TOPSIS). Customers preference is important for product design and
development and also for market positioning of the competitive product.
Although the accuracy of research of smartphone is based on a limited number
of respondents from a particular section, i.e., elite class only because
smartphones are mostly used by the high income group and educated people
(computer and internet friendly). This study shows the market trend of
smartphones in developing countries like India. The main limitation of this
research is involvement of only a particular section of the customer. This
method is limited to give the relative ranking of smartphones not the absolute
value of the rating. The originality of the paper is incorporation of customer
feedback on the ranking analysis of the phones using TOPSIS.
Keywords: multi-criteria decision-making; MCDM; technique of order
preference by similarity to ideal solution; TOPSIS; smartphone.
Reference to this paper should be made as follows: Kumari, S., Bajpai, A.,
Dewan, R. and Kumar, P. (xxxx) Measuring brand preferences of smartphone
using TOPSIS, Int. J. Modelling in Operations Management, Vol. X, No. Y,
pp.xxxxxx.
Biographical notes: Sushma Kumari received her MBA degree from IIIT
Allahabad. She has an interest in supply chain and marketing. She is self
employed and doing consultancy in marketing.
Ashruti Bajpai received her MBA degree from IIIT Allahabad. She has an
interest in operations management and marketing. She is self employed and
doing consultancy in marketing.

Copyright 20XX Inderscience Enterprises Ltd.

S. Kumari et al.
Rachita Dewan received her MBA degree from IIIT Allahabad. She has an
interest in marketing management and working as a Management Trainee in a
private company.
Pravin Kumar is a PhD from IIT Delhi and MTech from IT, BHU, Varanasi.
He has worked as an Associate Professor in the MBA (IT) division, IIIT
Allahabad for the years 20132014 (On lien). Presently, he is working as
an Assistant Professor, in the Mechanical Engineering Department, Delhi
Technological University, Delhi. He authored books on engineering economics
and mechanical engineering. He has also published several research papers in
the areas of supply chain management, operations management and logistics
management in reputed international journals. He has a teaching experience
of over 13 years. His present research works are in the field of supply chain
management, IT management, quality management and fuzzy models etc.

Introduction

The smartphone has become a basic need for customers due to requirements of faster and
more personalised social and business interaction. Cellular phone technology and
additional hardware were integrated into personal digital assistants and they evolved in
the form of smartphone (Baumgart, 2011). A smartphone is technologically advanced
mobile phone which is capable of performing many such functions that are required a
computer or laptop for performing. It can be used for accessing e-mails, browsing the
web and other basic applications. Most of the smartphones are equipped with high
resolution cameras and loaded with added features like games and eBook readers, social
networking apps and hardware inclusions such as GPS maps, Bluetooth, Wi-Fi and
navigations to make it an all simplified and inclusive operational tool. In a competitive
market, brands like Nokia, Samsung, Apple, Blackberry, HTC, Sony, etc., are expanding
their product lines/mix and differentiating their products with a wide variety to increase
the sales turnover and generate more revenue. For last one decade Nokia dominated the
Indian mobile handset market, but it has lost its market share, mainly due to the
competition created by Samsung, Micromax, Apple, HTC and BlackBerry. In general, it
has been observed that brands like Apple, HTC and Blackberry are costlier than the
brands like Nokia, Samsung, Sony, Micromax and Lava and the cost of a smartphone is
one of the important factors in the lower income group of the customers.
Indian smartphone markets are full of low-priced smartphone models from domestic
to foreign handset makers. Smartphone sales in India have grown rapidly (12.8 million
units in the third quarter of 2013). According to International Data Corporation (IDC,
2013), the Indian smartphone market grew by 229% year-on-year to 12.8 million
smartphones in the third quarter of 2013 compared to 3.8 million units in the third quarter
of 2012. Smartphones have become a status symbol in society. The first problem faced by
the consumers and product offering firms is to know the preference order of customers
choice. The second problem is to interpret and to categorise the features of the
smartphones that will fulfil the maximum requirements of customers with minimal cost.
Smartphone technology (iPhone/iPad) offers highly progressive and large resolution with
the built-in millisecond timing of stimulus display and touch screen responses.

Measuring brand preferences of smartphone using TOPSIS

Regarding smartphones, only internet friendly users are aware of their basic needs
and configurations offered by the products manufacturer. Consumers, who have prior
experience with a smartphone, make them to learn quickly from their experience.
Whether the experience is positive or negative, experienced consumers will quickly
fiddle with their subsequent evaluations in purchasing their next Smartphone. Future
smartphones designs should be based on user-friendly operation and research to make
sure that actual user needs and preferences are perceived and incorporated into the
designs (Glasscock and Wogalter, 2006).
In this paper, the relative rankings of some of the popular smartphones in Indian
Markets have been prepared to know the customers preferences. These rankings and
inclusion of various parameters to rank the phone may be helpful in improving the
product design and to know the market positions. Technique of order preference by
similarity to ideal solution (TOPSIS) has been used as a tool for the ranking analysis.
Hwang and Yoon (1981) originally proposed this technique to select the best alternative
with a finite number of criteria (Hwang et al., 1994). TOPSIS is a suitable method to
determine the order preference. In this method a normalised matrix is used to avoid a
biased result due to large dissimilarities in the scaling of the criteria ratings. Based on
literature survey 12 basic features of smartphones have been identified that a common
smartphones possess like battery life, screen resolution, dimension, operating system,
design and appearance, camera and zoom, customer support, web access and
connectivity, audio and video compatibility and document reader etc. These features are
rated by the customers in view of importance, level and satisfaction level. Some
smartphones were identified that are commonly used by the customers in Indian markets.
Ranks of preference of these phones are determined using the TOPSIS. But there are
limitations of TOPSIS is that it shows the value of the rating of the DMUs in relative
terms only not in the absolute term like efficiency etc.
This paper has five sections: first section introduces the research topics; second
section presents the literature review; the third section explains the methodology; the
fourth section discusses the case study and fifth section concludes the research.

Literature review

In the present market scenario, customer expectations are higher quality and broad
features products and services; and therefore producers are more concerned about target
marketing and designing their products according to the needs of customers.
smartphone has become a versatile tool with fast running processors facilitating various
applications, such as finger touch internet and calls replacing the desktops, personal
computers and other phones. A unique brand defines the main features of a product and
service and latest technology that support such features, which may override on corporate
brand after a certain span of time.
Consumers have become highly dependent on smartphones to retrieve useful
information with a simple browse and click to access their smartphones as it is with them
when they often commute, relax at home in peace, take a trip overseas and so on (Genova
et al., 2010). The dependency on smartphones has led to decrease in thinking capability
in humans. Ting et al. (2011) investigated the effect of convenience, social needs and
social influences which highlights the dependency on smartphones and the impact of
purchase behaviour.

S. Kumari et al.

Decision-making problem is the process of finding the best option from the
alternatives. In most of such problems, the multiplicity of criteria for judging the
alternatives is pervasive. That is, for many such problems, the decision maker solves such
problems as multi-decision-making (MCDM) problem (Jahanshahloo et al., 2006).
Multi-criteria decision model has been extensively used by many researchers in the field
of operational research in decision-making environment for ranking the given alternatives
by specifying their respective weights and criteria. A number of MCDMs have been used
for the purpose of decision-making. For example, an analytical hierarchy process (AHP)
decision model was used for facility selection from the view of organisations which
contemplate locations of a new facility or a relocation of existing facilities (Yang and
Lee, 1997). But this method is unfair as it is based on self-assessment bias thus affecting
internal validity.
TOPSIS has been used for solving the various types of problems in management.
Kumar (2008) used TOPSIS integrated with AHP to evaluate third party logistics service
providers in a supply chain. Kumar and Singh (2012) investigated fuzzy AHP and
TOPSIS for third party logistics service providers evaluation. Gangurde and Akarte
(2013) proposed a MCDM approach to evaluate product design alternatives with respect
to the customer requirement. The study deals with the attributes of the most preferred
brands of smartphones by using AHP-modified TOPSIS approach to facilitate the most
suitable product configuration selection based on the customer requirements. This
method facilitated the design of the best product configuration that would maximise
customer satisfaction with minimum cost. Besides these techniques modified AHP was
used to analyse the productivity at the production site (Lee et al., 2012). AHP has
problems due to interdependencies between criteria in the form of pairwise comparison
that can lead to inconsistencies between judgment and ranking criteria.
Mishra et al. (2013) used fuzzy TOPSIS in decision-making for agility in mass
customisation systems, but the fuzzy set theory is difficult to develop and understand and
require numerous simulations before use. Ramanathan (2007) used DEA to select
appropriate suppliers for a firm in one of his research work, but this method does not deal
with imprecise data and assumes that all input and output are exactly known. Few
researchers have been using the fuzzy DEA to incorporate the impreciseness in outputs
and inputs. Simple multi-attribute rating technique (SMART) is also used for weight
assessment, but procedure is not so much convenient (Velasquez and Hester, 2013).
Therefore, this study, taking into consideration the benefits and limitations of other
MCDM techniques, uses TOPSIS method to find positive and negative ideal solutions.
TOPSIS method is a technique for order preference by similarity to ideal solution
which is also called positive ideal solution that maximises the benefit criteria
and minimises the cost criteria, whereas on the other hand the negative ideal solution
which is also called anti-ideal solution maximises the cost criteria or attributes and
minimises the benefit criteria or attributes (Amiri-Aref et al., 2012). In general, the
benefit criteria/attributes are those for maximisation, while the cost criteria/attributes are
those for minimisation. TOPSIS assumes that each criterion takes either increasing or
decreasing utility. The best alternative method is the one, which is closest to the positive
ideal solution and far from the negative ideal solution. TOPSIS uses information for
attribute implementation and provides a ranking of alternatives. There are several
applications of TOPSIS and it is used in many fields of research. Due to logical
reasoning, TOPSIS has been used to solve many real-world problems in the recent
years. This advantage will accommodate many applications in the future. The

Measuring brand preferences of smartphone using TOPSIS

primary data for the TOPSIS is collected by a survey conducted by the authors during
November-December of 2013. TOPSIS helps the decision makers to solve the problems
and finding proper solutions to the problems by comparing and ranking the set of
alternatives or decision-making units (DMUs). One of the major advantages of using
TOPSIS is that it helps in performance measurement from a set of alternatives and that is
why it is mainly known as the utility based method. In the past few years the TOPSIS
method has been applied in many fields like manufacturing, human resources, quality
control, product design and formation, location analysis, etc. A relative advantage of
TOPSIS is the ability to identify the best alternative and it has been comparatively tested
with a number of other multi attribute methods. One major limitation of this method is
that it gives only relative performance measures of the alternatives not the absolute one.
In this section, a number of MCDM tools have been highlighted that are used for
comparing the DMUs, but there are some limitations in each of the MCDM methods. For
example, AHP is based on pairwise comparison, but the importance of one parameter in
the pair with other parameter may vary from user to user and the final rating may be
different for different users. In the case of TOPSIS, there is no such type of problems.
DMUs are rated against the criteria and the weights of the criteria have been determined
by the market survey. However, these days researchers are using the integrated
framework of AHP and TOPSIS in the case of performance measurement.

Methodology

Presently, there are many brands that are providing smartphones. A customer always tries
to get the best return for his investments. A smartphone provides a lot of features; this is
the reason that a smartphone provide has better demand than a mobile phone. Also, the
customers of smartphones are increasing day-by-day. This study explores that which
brand of smartphones has a largest preference in Indian markets. The data were collected
by the customer survey. It was an online survey. There were 526 respondents who filled
the form through Facebook, Google+, e-mails, etc.
Twelve features as battery life (F1), screen resolution (F2), dimension (F3), operating
system (F4) design and appearance (F5), camera and zoom (F6), customer support (F7),
web access and connectivity (F8), audio compatibility (F9), video compatibility (F10),
document reader (F11) and price (F12) has been incorporated in the analysis.
The survey was conducted among the youth (student communities) at Allahabad in
India. Allahabad is one of the important cities in India. There are a number of central and
state Government and private institutions. Online survey was conducted among the users
of smartphones. In India, most of the users of smartphones are university students and
officers. There were 61% male respondents and 39% female respondents. 72% (approx)
of the respondents lie in the age group of 1834 years, the reason can be, smartphones are
the advanced phones, so the people within this age are more comfortable than the people
of other age-groups. Old aged people prefer less featured mobile phone 60% of the
respondents were graduates. 58% of the respondents were students with no income, only
6% of the respondents lie within the Rs. 0.51.0 million income range and 2% of
respondents lie in the income range of 1.0 million and above.
Five points Likert scale is used to measure the importance level on the basis of ratings
given by the customers to each feature, which denotes the importance of one feature over
the other while using a smartphone. The ratings ranging from:

S. Kumari et al.
1
Not at all
important

Slightly
important

Moderately
important

Somewhat
important

Extremely
important

Similarly the satisfaction level is also measured by five point Likert scale where the
customer rated each feature on the basis of satisfaction gained while using a particular
feature of the smartphone. The ratings ranging from:
1
Very
dissatisfied

Dissatisfied

Neutral

Satisfied

Very satisfied

3.1 Calculating customer satisfaction index


It is the index that provides information regarding the degree of satisfaction for
the products or services. CSI is a measure of the degree to which a product or service
meets the customers requirement. This index helps dealers or the customers to know
how the product is meeting to the customers expectations and perceptions. Hence, CSI is
considered as one of the important tools in the decision-making process as it affects the
decision process and ranking of the product alternatives.
In order to calculate this index, the questionnaire included the question like how do
you rate the following features with respect to the importance level on five point rating
scale? The customers were asked to rate all the 12 features based on their importance
level as how much the features are important to the customers. Also, we rated the most
six preferred phones based on the satisfaction level they arrived on usage of their
respective phones. The CSI has given by equation below is based on customer weight age
to various features as suggested by Hong et al. (2008).

CSI =

F
f =1
F

w j xij

f =1

wj

where
CSI

customer satisfaction index

total number of features considered

wj

weights of features based on importance level

xij

average of features based on satisfaction level of smartphones.

The sample calculation of customer satisfaction index (CSI) for model S1 is given below:
4.14 4.762712 + 4.13 4.423729 + 3.97 4.508475 + 4.21 4.610169
+4.27 4.542373 + 4.39 4.491525 + 4.15 4.728814 + 4.40 4.779661
+4.25 4.474576 + 4.18 4.745763 + 3.64 4.847458 + 4.62 25,500
CSI =
4.14 + 4.13 + 3.97 + 4.21 + 4.27 + 4.39 + 4.15 + 4.40 + 4.25 + 4.18 + 3.64 + 4.62

CSI for S1 = 4.6226

Measuring brand preferences of smartphone using TOPSIS

Other CSIs are S1 = 4.6226, S2 = 406216, S3 = 4.622, S4 = 4.6219, S5= 4.6214,


S6 = 4.6188. So we can observe that most trust smartphone, according to CSI is S1.

3.2 Calculation through TOPSIS method


TOPSIS method considers two types of attributes or criteria:

benefit attributes/criteria

cost attributes/criteria.

In this method, two artificial alternatives are hypothesised as:


a

positive ideal alternative: the one which has the best level of all attributes considered

negative ideal alternative: the one which has the worst attribute values.

TOPSIS selects the alternative that is the closest to the positive ideal solution and far
from negative ideal alternative. It assumes that we have m alternatives and n
attributes/criteria and the result is the score of each option with respect to each criterion.
Following steps have been used in TOPSIS:

3.2.1 Step 1
Constructing a decision matrix refers to m alternatives and n criteria.
The given numbers xij and their matrix;
x11
x
21
D=

xm1

x12 x1n
x22 x2 n

xm 2 xmn

where xij shows the rating of ith DMU with respect to jth criteria.
i =1, 2, 3,, m is the number of DMU and j = 1, 2, 3,, n is the number of criteria.

3.2.2 Step 2
Calculating the normalised decision matrix as:
r11 r12
r
r22
21
R=

rn1 rn 2

r1m
r2 m

rnm

This step transforms various attribute dimensions in non-dimensional attributes, which


allows comparisons across criteria.
Normalise scores or data can be calculated as:

S. Kumari et al.
xij

rij =

for i = 1, , n; j = 1, , m

xij2

i =1

3.2.3 Step 3
Constructing the weighted normalised decision matrix.
To get the weighted normalised elements of the matrix multiply each column of the
normalised decision matrix by its associated weight of the criteria.
An element of the new matrix is:
v11 v12
v
v22
21
v=

v n1 vn 2

v1m
v2 m

vmn

vij = w j rij

where wj = the weights of different attributes.

3.2.4 Step 4
Determining the positive and negative ideal solutions as v +j and v j respectively by
finding the maximum and minimum values of weighted normalised elements in each
column in the case of benefit criteria and just reverse for cost criteria.
Determine the positive ideal and negative ideal solutions.
Positive ideal solution,
A = {v1 , , vn }

vij = max ( vij ) if

j J ; min ( vij ) if
i

j J

where J is associated with benefit criteria and J is associated with cost criteria.
Negative ideal solution,
A = {v1, , vn }

vij = min ( vij ) if

j J ; max ( vij ) if

j J

where J is associated with benefit criteria and J is associated with cost criteria.

3.2.5 Step 5
Calculate the separation measures for each alternative.
The separation from the positive ideal alternative is represented by Si*
Si* =

(v

*
ij

1/ 2

2
v ij )

where i = 1, , m and

j = 1, 2, , n

Measuring brand preferences of smartphone using TOPSIS

Similarly, the separation from the negative ideal alternative is represented by Si


Si =

( v v
ij

1/2

ij

)2

where i = 1, 2, , m and

j = 1, 2, , n

3.2.6 Step 6
Calculate the relative closeness to the ideal solution Ci* closest to 1 depicts the best
alternative.
Ci* = Si / ( Si* + Si ) , where, 0 < Ci* < 1

where

Si*

represents a positive ideal solution

Si

represents a negative ideal solution.

3.2.7 Step 7
Rank the preference order as Ci* . The one that has the shortest distance to ideal solution
is the best alternative. The relationship of alternatives reveals that, any alternative which
has the longest distance negative ideal solution is guaranteed to have the shortest distance
to positive ideal solutions.

Case study

TOPSIS is the appropriate tool to measure the brand preference of smartphones. It


calculates the ideal solutions for given alternatives. Smartphones ranges from Rs. 5,000
to Rs. 50,000 with varied features and numerous applications. Nowadays people are
unable to take a decision about purchasing the appropriate phone because it needs so
much surfing and research based on new market trends and most purchased phones.
To know the opinion of the customers about the smartphones online survey was
conducted and the data collected were compiled. In this survey, only those customers
were involved who are internet users, i.e., friendly with Facebook, e-mail, etc. We can
observe that most of the respondents were the youth of the country. Twelve features as
battery life (F1), screen resolution (F2), dimension (F3), Operating System (F4) design
and appearance (F5), camera and zoom (F6), customer support (F7), web access and
connectivity (F8), audio compatibility (F9), video compatibility (F10), document reader
(F11) and price (F12) were taken. Price has been taken from the websites of respective
smartphones on date. Weights were generated by taking the average of different attributes
shown in Table 1.
Table 1
F1
4.14

Weights for different features of a smartphone


F2

F3

F4

F5

F6

F7

F8

F9

F10

F11

F12

4.13

3.97

4.21

4.27

4.39

4.15

4.40

4.25

4.18

3.64

4.62

10

S. Kumari et al.

Battery life

Screen
resolution

Dimension

Operating
system

Design and
appearance

Camera and
zoom

Customer
support

Web access and


connectivity

Audio
compatibility

Video
compatibility

Document
reader

Price

Rating of smartphones with respect to the features

Table 2

F1

F2

F3

F4

F5

F6

F7

F8

F9

F10

F11

F12

Weights

4.14

4.13

3.97

4.21

4.27

4.39

4.15

4.40

4.25

4.18

3.64

4.62

S1

4.76

4.42

4.51

4.61

4.54

4.49

4.73

4.78

4.47

4.75

4.85

25500

S2

3.88

3.79

4.92

4.75

4.04

4.08

4.54

4.75

4.13

4.21

4.88

11400

S3

4.77

4.31

4.23

4.08

4.77

4.62

4.27

4.08

4.73

4.77

4.15

13990

S4

4.41

4.70

4.48

4.67

4.59

4.26

4.26

4.48

4.48

4.37

4.48

13499

S5

4.05

3.36

3.90

4.34

4.81

4.81

4.79

4.76

4.71

4.41

4.60

9999

S6

4.74

3.83

4.09

4.78

4.48

3.91

4.65

4.57

4.57

4.57

4.48

4740

Price

Document
reader

Video
compatibility

Audio
compatibility

Web access and


connectivity

Customer
support

Camera and
zoom

Design and
appearance

Operating
system

Dimension

Battery life

Screen
resolution

Normalised matrix (rj is calculated for each xij)

Table 3

F1

F2

F3

F4

F5

F6

F7

F8

F9

F10

F11

F12

S1

0.44

0.44

0.42

0.41

0.41

0.42

0.42

0.43

0.40

0.43

0.43

0.71

S2

0.36

0.38

0.46

0.43

0.36

0.38

0.41

0.42

0.37

0.38

0.43

0.32

S3

0.44

0.43

0.40

0.37

0.43

0.43

0.38

0.36

0.43

0.43

0.37

0.39

S4

0.40

0.47

0.42

0.42

0.41

0.40

0.38

0.40

0.40

0.40

0.40

0.38

S5

0.37

0.34

0.36

0.39

0.43

0.45

0.43

0.42

0.43

0.40

0.41

0.28

S6

0.43

0.38

0.38

0.43

0.40

0.37

0.42

0.41

0.41

0.41

0.40

0.13

The data collected and displayed the most six preferred phones which are S1, S2, S3, S4,
S5 and S6. Surveyors were asked to rate the smartphones which they have based on
12 attributes on five point rating scale. The average was determined using these brands
shown in Table 2.
Each Xij was squared and roots were determined to calculate rj which can be shown in
following Table 3.
After determination of rj. All they rj were multiplied with respective weights (rj wj).
Table 4 shows the weighted normalised matrix.

Measuring brand preferences of smartphone using TOPSIS

11

Weighted normalised matrix

Battery life

Screen resolution

Dimension

Operating system

Design and
appearance

Camera and zoom

Customer support

Web access and


connectivity

Audio compatibility

Video compatibility

Document reader

Price

Table 4

F1

F2

F3

F4

F5

F6

F7

F8

F9

F10

F11

F12

S1

1.81

1.82

1.67

1.75

1.74

1.84

1.76

1.88

1.72

1.79

1.57

3.29

S2

1.47

1.56

1.83

1.80

1.55

1.67

1.69

1.87

1.58

1.59

1.58

1.47

S3

1.81

1.77

1.57

1.54

1.83

1.89

1.59

1.60

1.82

1.80

1.35

1.81

S4

1.68

1.94

1.66

1.77

1.76

1.75

1.59

1.76

1.72

1.65

1.45

1.74

S5

1.54

1.38

1.45

1.64

1.84

1.97

1.79

1.87

1.81

1.67

1.49

1.29

S6

1.80

1.58

1.52

1.81

1.72

1.60

1.73

1.79

1.75

1.72

1.45

0.61

Next step is to identify the Positive Ideal Solutions which was ascertained by taking the
maximum among the rj wj for all the attributes except F12 this was taken as minimum
because cost related features are considered as negative attribute in TOPSIS so as Price in
this. This gives the S'* shown in Table 5.
Table 5

Value of A* (positive ideal solution) for each feature

vj

F1

F2

F3

F4

F5

F6

F7

F8

F9

Ideal solution for A* 1.81 1.94 1.83 1.81 1.84 1.97 1.79 1.88 1.82

F10 F11 F12


1.8

1.58 0.61

After getting the positive ideal solutions (vj) each vij was subtracted and squared to find
out the roots of all the alternatives, shown in Table 6 using;
Si * =
Table 6

( v v
j

ij

1/ 2

)2

Separation from positive ideal solution for each smartphones

S1

0.36

S2

0.75

S3

0.65

S4

0.50

S5

0.81

S6

0.64

S'* is the positive ideal solutions. Similar steps were used to calculate negative ideal
solution, here vij is taken minimum value among the benefit attributes and maximum
among the cost attributes as shown in Table 7 using bold digits.

12

S. Kumari et al.
Weighted normalised matrix (negative ideal solution)

Battery life

Screen
resolution

Dimension

Operating
system

Design and
appearance

Camera and
zoom

Customer
support

Web access and


connectivity

Audio
compatibility

Video
compatibility

Document
reader

Price

Table 7

F1

F2

F3

F4

F5

F6

F7

F8

F9

F10

F11

F12

S1

1.81

1.82

1.67

1.75

1.74

1.84

1.76

1.88

1.72

1.79

1.57

1.91

S2

1.47

1.56

1.83

1.8

1.55

1.67

1.69

1.87

1.58

1.59

1.58

1.87
1.97

S3

1.81

1.77

1.57

1.54

1.83

1.89

1.59

1.6

1.82

1.8

1.35

S4

1.68

1.94

1.66

1.77

1.76

1.75

1.59

1.76

1.72

1.65

1.45

1.9

S5

1.54

1.38

1.45

1.64

1.84

1.97

1.79

1.87

1.81

1.67

1.49

1.97

S6

1.8

1.58

1.52

1.81

1.72

1.6

1.73

1.79

1.75

1.72

1.45

1.69

In Table 7, the benefits criteria are taken as minimum and cost criteria, i.e., price is taken
as maximum to ascertain the negative ideal solutions. The values for negative ideal
solution can be taken as seen in Table 8.
Table 8

Value of A (negative ideal solution) for each feature


F1

vj

F2

F3

F4

F5

F6

F7

F8

F9 F10 F11 F12

Negative ideal solution (A) 1.47 1.38 1.45 1.54 1.55 1.6 1.59 1.6 1.58 1.59 1.35 1.97

After getting the negative ideal solutions (vj) each vij was subtracted and squared to find
out the roots of all the alternatives using the following equation.
Si =

( v v
j

ij

1/ 2

)2

Separation from negative ideal solution is shown in Table 9.


Table 9

Separation from negative ideal solution for each smartphone

S1

0.85

S2

0.63

S3

0.74

S4

0.76

S5

0.65

S6

0.67

Calculated the relative closeness to the ideal solution Ci = Si / ( Si* + Si) shown in
Table 10.

Measuring brand preferences of smartphone using TOPSIS


Table 10

13

Relative closeness to the ideal solution

S1

0.701

S2

0.454

S3

0.533

S4

0.604

S5

0.446

S6

0.509

TOPSIS determines the best Alternative which is closer to 1 and also ranks among the
given alternatives. Ranks were determined by arranging the above results in descending
order shown in Table 11.
Table 11

Ranks for smartphone based on TOPSIS approach

Smartphone

Rank

S1

S2

S3

S4

S5

S6

Thus, this is the method of compensatory aggregation that compares a set of alternatives
by identifying weights for each criterion, normalising scores for each criterion and
calculating the geometric distance between each alternative and the ideal alternative,
which is the best score in each criterion. Here the most preferred brand, according to
TOPSIS comes to be S1.

Discussions and conclusions

This study has successfully implemented the methodology TOPSIS as MCDM in the
determination of the smartphone preference ranking. It simplifies the problem of
selecting the best alternatives among the number of smartphone by taking the importance
of each feature and satisfaction level involved in it. TOPSIS is a good tool to measure
this problem as it is based on information on attributes and also is the cardinal approach.
This study has used the CSI as a supplementary tool. The CSI is used to measure the
satisfaction level of the customer with the given phone. Data reveals that S1 has the more
satisfaction index level, i.e., 4.6226. The cause for this result can be many it can be in
terms of price, also from applications and android versions. Also, in TOPSIS approach,
S1 has the best ranking, which comprises of responses based on attributes and satisfaction
level. TOPSIS normalised the data and ranked the best preferred phones, which was S1
according to this study. Table 11 depicts, the smartphones are ranked based on relative
closeness to ideal solution. Lowest is given to S5.
The data collected was confined only to online users. Also the results are based on a
limited number of respondents so it may very when considering the large population
covering the entire section. This research paper aimed to use the TOPSIS multi-criterion

14

S. Kumari et al.

decision-making in the smartphone market, which can be used to select the best
alternatives with the help of simple mathematical calculation. Also, only those phones are
considered in the analysis, which were in common use at the time of the survey
(NovemberDecember 2013). Very costly smartphones are not considered due to the
limited number of available users in Indian markets. This paper may help the manager to
improve the performance of the smartphones and to check the market position of the
specific product. Also, this paper contributes to the application of TOPSIS as a tool for
comparison of the smartphones.

References
Amiri-Aref, M., Javadian, N. and Kazemi, M. (2012) A new fuzzy positive and negative ideal
solution for fuzzy TOPSIS, WSEAS Transactions on Circuits and Systems, Vol. 11, No. 3,
pp.92103.
Baumgart, D.C. (2011) Smartphones in clinical practice, medical education, Arch Intern Med.,
Vol. 171, No. 14, pp.12941296.
Gangurde, S.R. and Akarte, M.M. (2013) Customer preference oriented product design using
AHP-modified TOPSIS approach, Benchmarking: An International Journal, Vol. 20, No. 4,
pp.549564, Emerald Group Publishing Limited.
Genova, G.L. (2010) The anywhere office 1/4 anywhere liability, Business Communication
Quarterly, Vol. 73, No. 1, p.119.
Glasscock, N.F. and Wogalter, M.S. (2006) Evaluating preferences for mobile phone features,
Proceedings of the Human Factors and Ergonomics Society 50th Annual Meeting 2006.
Hong, G., Hu, L., Xue, D., Tu, L. and Xiong, L. (2008) Identification of the optimal product
configuration and parameters based on individual customer requirements on performance and
costs in one-of-a-kind production, International Journal of Production Research, Vol. 46,
No. 12, pp.32973326.
Hwang, C.L. and Yoon, K. (1981) Multiple Attributes Decision Making Methods and Applications,
Springer, Berlin, Heidelberg.
Hwang, C.L., Lai, Y.J. and Liu, T.Y., (1994) TOPSIS for MODM, European Journal of
Operational Research, Vol. 76, No. 3, pp.486500.
IDC (2013) [online] http://www.idc.com/getdoc.jsp?containerId=prIN24471213 (accessed March
2014).
Jahanshahloo, G.R., Lotfi, F.H. and Izadikhah, M. (2006) An algorithmic method to extend
TOPSIS for decision-making problems with interval data, Applied Mathematics and
Computation, Vol. 175, No. 2, pp.13751384.
Kumar, P. (2008) An integrated for model of AHP and 3PL evaluation, Asia-Pacific Business
Review, Vol. 4, No. 3, pp.1421.
Kumar, P. and Singh, R. K. (2012) A fuzzy AHP and TOPSIS methodology to evaluate 3PL in a
supply chain, Journal of Modelling in Management, Vol. 7, No. 3, pp.287303.
Lee, M.N., Lee, K.Y., Kwon, S.W. and Shin, J.H. (2012) Introducing and applying a modified
AHP (analysis hierarchy process) to analyse productivity at the construction site,
Gerontechnology: International Journal on the Fundamental Aspects of Technology to Serve
the Ageing Society, Vol. 11, No. 2, pp.415420.
Mishra, S., Dutta, S. and Mahapatra, S.S. (2013) Grey-based and fuzzy TOPSIS decision-making
approach for agility evaluation of mass customization systems, Benchmarking: An
International Journal, Vol. 20, No. 4, pp.440462.
Ramanathan, R. (2007) Supplier selection problem: integrating DEA with the approaches of total
cost of ownership and AHP, Supply Chain Management: An International Journal, Vol. 12,
No. 4, pp.258261.

Measuring brand preferences of smartphone using TOPSIS

15

Ting, D.H., Lim, S.F, Patanmacia, T.S., Low, C.G. and Ker, G.C. (2011) Dependency on
smartphone and the impact on purchase behavior, Young Consumers: Insight and Ideas for
Responsible Marketers, Vol. 12, No. 3, pp.193203.
Velasquez, M. and Hester, P.T. (2013) An analysis of multi-criteria decision making methods,
International Journal of Operations Research, Vol. 10, No. 2, pp.5666.
Yang, J. and Lee, H. (1997) An AHP Decision Model for Facility Location Selection, Vol. 15,
Nos. 9/10, pp.241254, Emerald Group Publishing Limited.

You might also like