See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/275043428

WEIGHTED INTUITIONISTIC FUZZY DELPHI METHOD

Article · August 2013

CITATIONS READS

3 228

2 authors, including:

Arindam Garai
Sonarpur Mahavidyalaya
33 PUBLICATIONS 125 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Intuitionistic Fuzzy T-set, Multi - objective optimization, Interactive optimization technique, Decision making under uncertainty View project

All content following this page was uploaded by Arindam Garai on 17 April 2015.

The user has requested enhancement of the downloaded file.

 Volume 4, No. 7, July 2013
Journal of Global Research in Computer Science
REVIEW ARTICLE
Available Online at www.jgrcs.info

WEIGHTED INTUITIONISTIC FUZZY DELPHI METHOD

Arindam Garai*1and Tapan Kumar Roy2
*1Department of Mathematics, Sonarpur Mahavidyalaya, Sonarpur, West Bengal, India
[email protected]
2
Department of Mathematics, Bengal Engineering and Science University, Howrah, West Bengal, India
[email protected]

Abstract: This paper presents weighted intuitionistic fuzzy Delphi method. In real life usage of Delphi method, information communicated by
experts may not be used with full and complete potential. Hence highly accurate and realistic conclusions cannot always be obtained. In
intuitionistic fuzzy Delphi method, communication with experts is the same as fuzzy Delphi method, yet an improved and elaborative statistical
tool is used to reach in better conclusions. Again, the experts use their individual competency and subjectivity. And competency and ability to
predict successfully varies extensively among experts. Thus different importance and hence weights should be assigned to them by the decision
maker. Hence more realistic and accurate prediction is obtained.
Key Word: Intuitionistic fuzzy Delphi method, Intuitionistic fuzzy decision making technique, Weighted Delphi technique.

Next, fuzzy set theory was further studied and intuitionistic

INTRODUCTION fuzzy set theory was introduced by K. T. Atanassov [3] [11].
And fuzzy Delphi method was further developed. In 2012, the
The Delphi Method [5] is a well structured communication authors studied Delphi technique under intuitionistic fuzzy
technique. It was originally developed as a systematic and environment [13].
interactive forecasting method that relies on a panel of experts. It is assumed till date that the capability and competency to
It belongs to the subjective-intuitive methods of foresight. predict are equal among experts. But this is not true always. It
Delphi was first developed in the 1940s by the Rand varies! Hence, in this study, the experts have been assigned
Corporation, Santa Monica, California, in operation research. different importance and hence weights. For simplicity,
Different approaches were tried, and to combat the many normalized weights are being used. This helps in forming
shortcomings, the Delphi Method was developed by Project smarter (!) sheaf of experts that leads to more effective and
RAND during the 1950-1960s (1959) by Olaf Helmer, Norman accurate prediction.
Dalkey and Nicholas Rescher [14].
The name can be traced back to the Delphic oracle, as DEFINITION
Woudenberg reports that the name „Delphi‟ was intentionally
coined by Kaplan, an associate professor of philosophy at the Delphi Method
UCLA working for the RAND corporation in a research effort Wechsler (1978) characterizes a „Standard Delphi Method‟ in
directed at improving the use of expert predictions in policy the following way: „It is a survey which is steered by a monitor
making. The temple was the locus of knowledge, i.e. the group, comprises several rounds of a group of experts, who are
Delphic oracle was probably the largest database of the ancient anonymous among each other and for whose subjective-
world [14]. intuitive prognoses a consensus is aimed at. After each survey
Overall the track record of the Delphi Method is mixed. There round, a standard feedback about the statistical group judgment
have been many cases when the method produced poor results. calculated from median and quartiles of single prognoses is
One may attribute this to poor application of the method and given and if possible, the arguments and counterarguments of
not to the weaknesses of the method itself. It must also be the extreme answers are fed back…‟ [14].
realized that in areas such as science and technology The Delphi Method is based on structural surveys and makes
forecasting, the degree of uncertainty is so great that exact and use of the intuitive and available information of the
always correct predictions are impossible, so a high degree of participants, who are mainly experts. Therefore it delivers
error is to be expected! qualitative as well as quantitative results and has beneath its
Despite these shortcomings, today the Delphi Method is a explorative, predictive even normative elements. There is not
widely accepted forecasting tool and has been used the one Delphi methodology but the applications are diverse.
successfully for thousands of studies in areas varying from There is agreement that Delphi is an expert survey in two or
technology forecasting to drug abuse [12]. more „rounds‟ in which in the second and later rounds of the
Several modifications and improvements have appeared in survey the results of the previous round are given as feedback.
Delphi method [1]. On the other hand, one of the largest shift Therefore, the experts answer from the second round on under
from traditional classical mathematics happened by the the influence of their colleagues‟ opinions. Thus, the Delphi
introduction of fuzzy set theory [6]. Fuzzy Delphi method was Method is a „relatively strongly structured group com-
developed to include and interpret the uncertainty involved in munication process, in which matters, on which naturally
experts‟ opinion [2]. unsure and incomplete knowledge is available, are judged upon
by experts‟.

© JGRCS 2010, All Rights Reserved 38

 Arindam Garai ,et al, Journal of Global Research in Computer Science, 4 (7), July 2013, 38-42

Fuzzy Set improved and elaborative statistical tool is used to reach in

A fuzzy subset à of X is defined by its membership function better conclusions. The experts use their individual competency
and subjectivity and are somehow uncertain to air their
Ã: X [0, 1] that assigns to every x X, a real number Ã
(x) in the closed unit interval [0, 1], where the value of à (x) at opinions. Hence, they tend to secure their opinions. Thus, they
x represents the grade of membership of x in à [6]. prefer degree of non-membership over degree of membership
and this is the very reason why use of intuitionistic fuzzy
Nearer the value of Ã(x) is unity, the grade of membership of
concepts is more relevant than fuzzy concepts. Moreover, by
x in Ã. When the membership function Ã(x) contains only two
using TIFNs, it is easier for an expert to study the realization
points 0 and 1, membership function Ã(x) is identical to the data which are nested within one another than TFNs. And, the
characteristic function : X [0, 1] and in that case à is a concept of sheaf of intuitionistic fuzzy number is an
crisp set. aggregation process which appears to be very convenient for
Intuitionistic Fuzzy Set the objectification of (somehow hazy) subjective opinions.
An intuitionistic fuzzy set [3] A in X is defined by A = {< x; On the other hand, among a group of experts, competency and
ability to successfully predict varies to great extent among
μA(x), νA(x) > | x X}, where μA: X [0, 1] and νA: X [0,
experts. It is well known that Nouriel Roubini had successfully
1] with the condition 0 ≤ (μA(x) + νA(x)) ≤ 1, where μA(x) and
predicted global recession of 2009 in as early as 2007. Many
νA(x) denote the degree of membership and non membership
economists called him Mr. Doom! Yet, it was exactly that! In
respectively [7].
real life situations, decision maker may assign varied
importance to experts‟ efficiency (from past experience or level
FUZZY DELPHI METHOD
of accuracy or any other pre defined criteria). Importances are
calculated in mathematics by assigning respective weights.
Fuzzy Delphi Method [5] was first introduced by Kaufman and
Weights are assigned to each expert. For simplicity, normalized
Gupta in 1988 and it was also proposed by Ishikawa et al.
weights are being used here. If there are n experts, their weights
(1993). Noorderhaben (1995) had indicated that applying the
may be assigned as
Fuzzy Delphi Method to group decision can solve the fuzziness n
of common understanding of expert opinions. The expert w 1 ,w 2 ...w n such that wi 1, wi 0 i=1,2...n.
prediction (or interval value) was then used to derive the fuzzy i 1

numbers, resulting in the Fuzzy Delphi Method. Hence, Fuzzy The steps of the proposed WIFDM are as follows:
Delphi Method is a generalization of the classical method [8]. Table II. Steps of Weighted Intuitionistic Fuzzy Delphi Method
It consists of the following steps:
Sr. Details
Table I. Steps of Fuzzy Dephi Method No.
Step The decision maker selects a panel of n experts. The ith expert is
Sr. No. Details 1 assigned a weight, say wi, depending on his competency, by the
Step 1 Experts are asked to provide the possible realization dates of a decision maker such that
certain event in science, technology, or business, namely: the n

earliest date, the most plausible date, and the latest date. The data wi 1, wi 0 i 1, 2...n.
i 1
given by the experts are presented to the moderator for fuzzy
averaging for forecasting. The expert Ei, i = 1, 2, …, n, are then requested to provide the possible
Step 2 First, the average (mean) is computed. Then for each expert the realization dates of a certain event in science, business or technology,
deviation between mean and respective data is computed. It is also viz. the earliest certain date ec1(i), the earliest uncertain date eu1(i), the
a triangular fuzzy number. The deviation is sent back to each of most plausible date mp1(i), the latest certain date lc1(i) and the latest
the expert for reevaluation. uncertain date lu1(i). Here „1‟ in the suffix indicates that this is the first
Step 3 Each expert again presents a new triangular number in second phase of forecasting process.
round. Next, the same process starting with Step 2 is repeated. Step Next, objective data is formed out of these subjective information by
The triangular averages are calculated once again and are 2 considering a triangular intuitionistic fuzzy number as follows: (Ei ;
substituted correspondingly. If necessary, new triangular numbers ec1(i), mp1(i), lc1(i); eu1(i), lu1(i)) with weight wi such that
n
are generated and their averages are calculated. The process could wi 1, wi 0 i 1, 2...n.
be repeated again and again until two successive means become i 1
reasonably close. Step These responses from n experts form a sheaf (Ei ; ec1(i), mp1(i), lc1(i);
Step 4 At a later time the forecasting may be reexamined by the same 3 eu1(i), lu1(i)), i = 1, 2, n. The mean of TIFN sheaf is then computed
process if there is important information available due to new (ec1m, mp1m, lc1m; eu1m, lu1m), keeping the weights assigned to experts
discoveries or any other misinterpretation in mind and for each expert the divergence is computed as follows: (Ei
; ec1m – ec1(i), mp1m – mp1(i), lc1m – lc1(i); eu1m – eu1(i), lu1m – lu1(i)).
INTUITIONISTIC AND WEIGHTED INTUITIONISTIC FUZZY Weighted arithmetic mean is used to calculate mean. Here these
divergence numbers can be positive, negative or null. This information
DELPHI METHOD (WIFDM) is then sent again to each individual expert for further prediction.

Intuitionistic fuzzy Delphi method was introduced in 2012 [13]. Step Each expert now gives decision maker a new TIFN (Ei ; ec2(i), mp2(i),
The arguments that can be used in favour of using triangular 4 lc2(i); e2(i), lu2(i)) and the process from Step 3 is repeated.
intuitionistic fuzzy numbers (TIFNs) in place of triangular
Step The process is continued until two successive means become
fuzzy numbers (TFNs) are that subjective information that may 5 reasonably close so that the Delphi manager is satisfied. The number
be transformed into objective values as required in Fuzzy of such iteration phases may also be limited a priori. There may be
Delphi Method cannot always be obtained. Subjective many variations of this procedure; e.g. the experts can be asked not to
information is more likely to be like quasi-objective data in increase the divergence without disturbing his unbiasness. Now, since
the word „close‟ is fuzzy, some in depth study is required. It can be
case of intuitionistic fuzzy number and hence the use of based on the concept of distance metric between intuitionistic fuzzy
intuitionistic fuzzy number is more justified. In case of numbers i.e. if necessary, a study of opinions from partial or full group
intuitionistic fuzzy Delphi method, communication with of experts is realized by calculating the distance between TIFN and
experts is the same as fuzzy Delphi method yet a more non-disjunctive group of experts are formed by finding maximum sub
relations of similarity.

© JGRCS 2010, All Rights Reserved 39

 Arindam Garai ,et al, Journal of Global Research in Computer Science, 4 (7), July 2013, 38-42

Step At a later time, the weights may be reassigned or experts are given Expert Earliest Earliest Most Latest Latest
6 equal importance; forecasting may be reexamined and reevaluated by Sr. No.. Uncertain Certain Plausible Certain Uncertain
same process in case of discovery or availability of new or important Date Date Date Date Date
information. eu2(i) ec2(i) mp2(i) lc2(i) lu2(i)
1 1995 1997 2003 2018 2021
CASE STUDY: TIME ESTIMATION FOR TECHNICAL
2 1995 1997 2004 2011 2013
REALIZATION OF AN INNOVATIVE PRODUCT
3 1998 1999 2005 2011 2013
The data required for the problem of the technological 4 1996 1998 2003 2008 2011
realization of a cognitive information processing computer (as 5 1997 2000 2005 2010 2014
used partially in literature for the sake of simplicity alone).
Opinions of five experts only are considered. Here it was In a similar way, the computation from this sheaf gives the
requested to a group of five computer experts to give a mean TIFN (ec2m, mp2m, lc2m; eu2m, lu2m) = (1997.85, 2004.1,
subjective estimation for the realization of new computing
2011.7; 1996.1, 2014.05) (1998, 2004, 2012; 1996, 2014).
technology in the format of intuitionistic fuzzy number i.e. it
Now the manager is satisfied because mean TIFN in both cases
will consist of the earliest certain date ec1(i), the earliest
are same. The process is stopped and the final TIFN is accepted
uncertain date eu1(i), the most plausible date mp1(i), the latest
as a combined conclusion of experts‟ opinions. This means that
certain date lc1(i) and the latest uncertain date lu1(i) for each
the realization of the invention will occur in time interval
expert Ei. It may be noted that the experts are not ranked
[1996, 2014] with the inside channel being [1998, 2004] and
equally and hence their opinions carry different weights. Let us
the most likely year for the realization is 2012.
take w1 = 0.15, w2 = 0.4, w3 = 0.3, w4 = 0.1, w5 = 0.05 so
5
Now, to find the non disjunctive group of experts, the distances
that wi 1, wi 0 i 1, 2...5. between experts‟ opinions are calculated. In literature, there is
i 1 no standard procedure to calculate the distance between TIFNs
The sheaf formed by experts‟ opinions is assumed to be as [9]. Here a technique described in by Arnold Kaufmann,
follows: Madan M. Gupta, is further developed [5]. Arnold Kaufmann,
Madan M. Gupta used d (Ni, Nj) to be the normalized distance
Table III. Initial Opinions by Five Experts
between two TFNs Ni and Nj with
1 l r
Earliest Earliest Most Latest Latest d ( Ni , N j ) ( Ni , N j ) ( Ni , N j )
Sr. 2( )
Uncertain Certain Plausible Certain Uncertain 2 ,
1
no. Date Date Date Date Date
eu1(i) ec1(i) mp1(i) lc1(i) lu1(i)
With Ni and Nj as respective TFNs given by experts i and j, l
is the left distance and r is the right distance, 2 and 1 are
1 1992 1995 2003 2020 2024
arbitrary values at the right and the left respectively chosen
2 1995 1997 2004 2010 2013
such that 0 d 1 [10].
3 1999 2000 2005 2010 2012 The normalized distance between two TIFNs Ei and Ej be
4 1997 1998 2003 2008 2010
1 | Eu (i ) Eu ( j ) | | Ec (i ) Ec ( j ) |
5 1992 1995 2010 2015 2019 d ( Ei , E j )
5( 2 1 ) | Lu ( i ) Lu ( j ) | | Lc ( i ) Lc ( j ) | | m p ( i ) m p ( j ) |
The computation from this sheaf gives the mean TIFN: (ec1m, Where 2 and 1 are proposed to be E u 2 m and Lu 2 m
mp1m, lc1m; eu1m, lu1m) = (1997.60, 2004.35, 2011.55; 1995.8,
respectively, provided 0 d 1. Else, the values of 2 and 1
2014.35) (1998, 2004, 2012; 1996, 2014). The deviations for
are suitably chosen so that the relation 0 d 1 holds. The
each expert are now calculated as in the following table [4].
results of the computations are tabulated for 2= E u 2 m = 1996
Table IV. Deviation for Each Expert at End of First Round

Expert eu1m(i) – ec1m(i) – mp1m(i) – lc1m(i) – lu1m(i) –

and 1= Lu 2 m = 2014.
Sr. No. eu1(i) ec1(i) mp1(i) lc1(i) lu1(i)
Table VI. Distances between Experts‟ Opinions
1 04 03 01 – 08 – 10
2 03 01 00 02 01
Expert Expert Expert Expert Expert
3 -03 -02 -01 02 02 Sr. No.
1 2 3 4 5
4 -01 00 01 04 04
5 04 03 -06 -03 – 05 Expert 1 0 0.200 0.275 0.275 0.275
Expert 2 0 0.075 0.100 0.100
Suppose that the manager is not satisfied with the mean (1998, Expert 3 0 0.125 0.050
2004, 2012; 1996, 2014). The deviations for each expert are Expert 4 0 0.125
given to respective expert and are requested to review his
Expert 5 0
previous forecast once again and a new sheaf of TIFNs is
obtained as follows.
It is to be noted that the minimum distance is d (E3, E5) = 0.050
Table V. Opinion at Second Round by Same Experts and the maximum distance is d (E1, E3) = d (E1, E4) = d (E1, E5)
Expert Earliest Earliest Most Latest Latest = 0.275. Now to find pair of experts for whom the distance is
Sr. No.. Uncertain Certain Plausible Certain Uncertain less than or equal to 0.1 (denoted by R), the table as below is
Date Date Date Date Date obtained.
eu2(i) ec2(i) mp2(i) lc2(i) lu2(i)
Table VII. Filtration of Experts‟ Opinions

© JGRCS 2010, All Rights Reserved 40

 Arindam Garai ,et al, Journal of Global Research in Computer Science, 4 (7), July 2013, 38-42

Result and discussion

Expert Expert Expert Expert Expert
Sr. No.
1 2 3 4 5 The results clearly indicate that customer is the king. It is
ultimately the choice of the Delphi manager i.e. the decision
Expert 1 NA maker. Experts are selected based on some pre defined criterion
Expert 2 NA R R R and are assigned importance. Conclusions depend not only on
the choice of experts but also the weights assigned on them. If a
Expert 3 NA R
good manager or interpreter is found, making futuristic
Expert 4 NA R decision will not be a stiff challenge! It is also biggest
Expert 5 NA weakness of the method. If the Delphi manager is unable to
select suitable experts or he is not justified enough to assign
Here d (i, i) = 0 for all i. So, this case is not considered. The proper importance (weights), incorrect prediction may be the
experts (2, 5) and (4, 5) have given almost same estimation. outcome. To explain and compare, the weakness and strength
Therefore the experts (2, 4, 5) form a subgroup of experts. of each method is discussed in the following table.
Similarly the experts (2, 5) and (3, 5) also have almost the same
Table X. Weighted Intuitionistic Fuzzy Delphi Method
estimation. Hence experts (2, 3, 5) form another sub group
from our group of experts. For another upper limit of the metric Method Methodology Weakness and strength
d, different class of experts may be obtained. Traditional Experts give independent Takes more time to collate
Delphi opinions; expert opinions as Survey
Another metric to calculate distance between pair of TIFNs. Method Data are analyzed must be repeated multiple
statistically and are times. So, the cost is high.
communicated to The survey recovery rate is
Distance between TIFNs [7] Ei and Ej may also be defined experts; low.
d ( Ei , E j ) Experts‟ reviews are In pushing for a consensus,
analyzed and this process it‟s easy to misinterpret
1 max(| E ci E cj |,| Lic Lcj |) is repeated until expert opinion.
i j i
3( 2 1 ) min(| E u E |,| L
u u Luj |) | M ip M pj | convergence.

Clearly 0 ≤ d ≤ 1. In this case, another table is formed by

Fuzzy Delphi Experts give independent Saves on survey time and
measuring distances between every pair of experts‟ opinions as Method opinions; hence saves cost by
follows: Subjective information reducing number of
are converted into surveys; increases
Table VIII. Another Distance Measure between Experts‟ Opninions objective data using questionnaire recovery
fuzzy number; rate.
A fuzzy statistical Experts can better express
Sr. Expert Expert Expert Expert Expert analysis is done and are their opinions, ensuring the
No. 1 2 3 4 5 communicated to completeness and
experts; consistency of the group
1 0 0.148 0.222 0.203 0.222 Experts‟ reviews are opinions as it takes into
analyzed and this process account the fuzziness that
2 0 0.056 0.278 0.093 is repeated until outcome cannot be avoided during
converges to a reasonable the survey process.
3 0 0.130 0.037
solution
4 0 0.093 Intuitionistic Communication with Reduces number of
5 0 Fuzzy experts is the same as surveys rapidly and
Weighted fuzzy Delphi Method yet increases questionnaire
Delphi an improved and recovery rate. So, the cost
It is to be noted that the minimum distance is d (3, 5) = 0.037 Method elaborative statistical tool is lower than Fuzzy Delphi
and the maximum distance is d (2, 4) = 0.278. Now to find pair is used to reach in better Method.
conclusions. Takes into account the
of experts for whom the distance is less than or equal to 0.1 Subjective information is degree of non-membership
(denoted by R‟), the table as below is obtained: more likely to be like a values that cannot be
quasi-objective data in avoided during the survey
Table IX. Filtration of Experts‟ Opinions case of intuitionistic process. Hence, it does not
fuzzy number; misinterpret experts‟
Sr. Expert Expert Expert Expert Expert Intuitionistic fuzzy original opinions and
No. 1 2 3 4 5 statistical analysis is provides a true reflection
done and is of their response.
1 NA communicated to Decision maker may not be
2 NA R‟ _____ R‟ experts; able to assign suitable
3 NA R‟ Importance and hence weights to expert.
4 NA R‟ weights are assigned to Moreover successful
5 NA each expert; Experts‟ prediction of one event
reviews are analyzed in does not guarantee success
detail and this process is in another prediction.
For instance, the experts (2, 3), (2, 5) and (3, 5), (4, 5) have repeated until outcome
given almost same estimation. Opinion of expert 2 is more converges to a reasonable
solution.
similar to expert 3 and 5 but not to expert 4. On the other hand,
prediction of expert 5 is similar to expert 3 and expert 4. Yet
expert 3 does not give similar prediction as expert 4.

© JGRCS 2010, All Rights Reserved 41

 Arindam Garai ,et al, Journal of Global Research in Computer Science, 4 (7), July 2013, 38-42

REFERENCES [9] E. Szmidt and J. Kacprzyk, “Distances between intuitionistic

fuzzy sets,” Fuzzy Sets and Systems, vol. 114, pp. 505–518,
[1] N. Dalkey and O. Helmer, “An experimental application of the 1997.
Delphi method to the use of experts,” Management Science, vol. [10] P. Grzegorzewski, “Metrics and orders in space of fuzzy
9, pp. 458–467, 1963. numbers,” Fuzzy Sets and Systems, vol. 97, pp. 83–94, 1998.
[2] R. E. Bellman and L. A. Zadeh, “Decision making in a fuzzy [11] K. Atanassov, “Two theorems for intuitionistic fuzzy sets,”
environment,” Management Science, vol. 17, B141–B164, 1970. Fuzzy Sets and Systems, vol. 110, pp. 267–269, 2000.
[3] K. Atanassov, “Intuitionistic fuzzy sets,” Fuzzy Sets and [12] H. C. Chu and G. J. Hwang, “Delphi-based approach to
Systems, vol. 20, pp. 87–96, 1986. developing expert systems with the cooperation of multiple
[4] D. Dubois and H. Prade, “The mean value of a fuzzy number,” experts,” Expert Systems with Applications, vol. 34, pp.2826–
Fuzzy Sets and Systems, vol. 24, pp. 279–300, 1987. 2840, 2008.
[5] A. Kaufmann and M. M. Gupta, Fuzzy mathematical models in [13] T. K. Roy and A. Garai, “Intuitionistic fuzzy delphi method:
engineering and management science, Elsevier Science Inc., more realistic and interactive forecasting tool,” Notes on
New York, 1988. Intuitionistic Fuzzy Sets, vol. 18, no. 2, 2012.
[6] G. J. Klir and T. A. Folger, Fuzzy Sets, uncertainty, and [14] Wikipedia contributors. Delphi method. Wikipedia. The Free
information, Prentice-Hall International, 1988. Encyclopedia. [13 June, 2012 09:57 UTC] (accessed 17th June,
[7] K. Atanassov, “Ideas for intuitionistic fuzzy sets equations, 2013). (Contributions from website)
inequalities and optimization,” Notes on Intuitionistic Fuzzy
Sets, vol. 1, No. 1, pp. 17–24, 1995.
[8] H. M. Hsu and C. T. Chen, “Aggregation of fuzzy opinions
under group decision making,” Fuzzy Sets and Systems, vol. 79,
pp. 279–285, 1996.

SHORT BIODATA OF ALL THE AUTHOR

Arindam Garai
Me@ passionately love to mentor using innovation, mixing
positive attitude and enthusiasm of a born leader and with strong
desire to be a part of success in my chosen academic life.
Books, especially on self-help, mystery are my first love.
Confidence and mental power have been gained from those
printed words.
Chess is my favourite game. It has improved my ability to think
logically.
Research Area: Fuzzy Set, Intuitionistic Fuzzy Optimization,
Fuzzy and Intuitionistic Fuzzy Decision Making, Delphi Method
etc.
Mail: [email protected]
Mob.: +91-9932890115

Dr. Tapan Kumar Roy

Me@ guide my students professionally as well as personally.
Research Area: Fuzzy and Intuitionistic Fuzzy set Theory,
Inventory, Transportation, Reliability Optimization, Portfolio
Optimization, Fuzzy and Stochastic Optimization etc.
Mail: [email protected]
Mob.: +91-9477419380

© JGRCS 2010, All Rights Reserved 42

View publication stats