Malaya Journal of Matematik, Vol. S, No.

1, 431-434, 2020

https://doi.org/10.26637/MJM0S20/0080

Comparative study of fuzzy assignment problem

with various ranking
R. Srinivasan1 *, G. Saveetha2 and T. Nakkeeran3

Abstract
In this study, we compare to the methodology for assignment problem with fuzzy optimal solutions. The
comparisons of the fuzzy optimal solutions are considered as triangular fuzzy numbers. Ranking techniques
are used for the triangular fuzzy numbers and so fuzzy numbers are converted to crisp values. Then we also
proceeded to Hungarian methodology is to resolve this type of fuzzy assignment problem. The Numerical
examples are effectively during this study.
Keywords
Triangular fuzzy numbers; Assignment problem; Fuzzy assignment problem; Ranking Techniques; Hungarian
method.
AMS Subject Classification
03E72.
1,2,3 Department of Ancient Science, Tamil University, Thanjavur-613010, Tamil Nadu, India.

*Corresponding author: 1 [email protected]; 2 [email protected]

Article History: Received 01 February 2020; Accepted 29 March 2020 c 2020 MJM.

Contents 1. Introduction
Assignment problem is a specific form of linear program-
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431
ming optimization problem that deals with one-to-one alloca-
2 Preliminaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431 tion of the different resources to the different jobs. The total
2.1 Fuzzy Set . . . . . . . . . . . . . . . . . . . . . . . . . . 431 cost of optimizing the full benefit of the allocation. Hungarian
approach is Kuhn’s most traditional problem solving method.
2.2 Fuzzy Number . . . . . . . . . . . . . . . . . . . . . . . 431
The problem with the assignment occurs because of an indi-
2.3 α-cut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432 vidual or a computer’s varying ability to perform a given task
2.4 Triangular Fuzzy Number . . . . . . . . . . . . . . . 432 of work. The question of assignment arises due to the vary-
ing capacity of an individual or computer to perform a given
3 Assignment Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
task or work. In this article, we tend to compare the fuzzy
4 Fuzzy Assignment Problem . . . . . . . . . . . . . . . . . . . . . . . 432 assignment problem to position various ranking methods.
5 Ranking Methods for Fuzzy Assignment Problem432
5.1 Centroid of a Generalized Triangular . . . . . . . 432 2. Preliminaries
5.2 Magnitude Ranking Method . . . . . . . . . . . . . 432 2.1 Fuzzy Set
5.3 Robust’s ranking technique . . . . . . . . . . . . . . 432 Let X be non–empty set. A fuzzy set  in X, mapping a
5.4 Sub Interval Average Method . . . . . . . . . . . . 432 membership function in the range µÂ (x) : X → [0, 1], for each
5.5 Pascal Triangular Approach . . . . . . . . . . . . . 432 x ∈ X. Then the fuzzy set is defined as  = {x, µÂ (x)\x ∈ X}.

6 Numerical Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 2.2 Fuzzy Number

7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 A fuzzy number  on R̂ in the real line set  : R̂ → I where
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433 Â(x) is the membership function of setÂ. Then the fuzzy
number  is defined by
(i) Â is normal.
 Comparative study of fuzzy assignment problem with various ranking — 432/434

(ii) Â is convex. Table 4.1

(iii) Â is upper semi continuous X ∈ R̂. 1 2 3 ... j ... M
1 B11 B12 B13 ... B1 j ... B1n
2.3 α-cut 2 B21 B22 B23 ... B2 j ... B2n
An α-cut of a fuzzy set P is a Crisp set Pα comprising all the ... ... ... ... ... ... ... ...
elements of the universal set X with a membership grade P i Bi1 Bi2 Bi3 ... Bi j ... Bin
greater than or equal to a given value thus ... ... ... ... ... ... ... ...
P = {x ∈ X/µA (x) > α} , 0 6 α 6 1 N Bn1 Bn2 Bn3 ... Bn j ... Bnn

2.4 Triangular Fuzzy Number

A triangular fuzzy number P = (a1 , a2 , a3 ) specified as mem- 5. Ranking Methods for Fuzzy
bership function Assignment Problem

0, if x 6 a

 5.1 Centroid of a Generalized Triangular

 (x − a)
, if a 6 x 6 b The Centroid of a triangle fuzzy number à = (a, b, c; w) as


P(x) = (b − a)
  
a+b+c w
(x − c) GÃ = , . The ranking function is known as the
, if b 6 x 6 c

 3 3
(b − c)




 generalized triangle fuzzy number à = (a, b, c; w) that maps
0, if x > c
the set of all fuzzy numbers to a set of real numbers is defined
as [1]
3. Assignment Problem


a+b+c w

R Ã = .
The problem of assignment can be represented as a n × n cost 3 3
matrix [Ci j ] of real numbers as shown in Table 3.1
5.2 Magnitude Ranking Method
Table 3.1 Triangular fuzzy number is defuzzified by ranking method
1 2 3 ... j ... M
for an arbitrary triangular fuzzy number (a, b, c) = (a1 , b1 , c1 )
1 B11 B12 B13 ... B1 j ... B1n magnitude of triangular fuzzy number is given by equation [2].
2 B21 B22 B23 ... B2 j ... B2n
 
Z1
... ... ... ... ... ... ... ... 1
Mag(a, b, c) = (c1 3a1 − b1 ) f (r)dr .
i Bi1 Bi2 Bi3 ... Bi j ... Bin 2
0
... ... ... ... ... ... ... ...
N Bn1 Bn2 Bn3 ... Bn j ... Bnn 5.3 Robust’s ranking technique
Given a convex triangluar fuzzy number S,
Mathematically assignment problem can be expressed as 
Z1


R(s) =  0.5 sLα , sUα dα 

n n
Minimize z = ∑ ∑ Bi j xi j 0
i=1 j=1
where sLα , sUα is the α-level cut of the fuzzy number S de-


n
Subject to ∑ xi j = 1, i = 1, 2, 3, . . . , n termines the Robust’s Ranking Index. The ranking index of
i=1 the Robust provides the representive value of the S fuzzy
n
number [3].
∑ xi j = 1, j = 1, 2, 3, . . . , n (3.1)
j=1
5.4 Sub Interval Average Method
xi j ∈ {0, 1} Let  = (a, b, c) be a triangular fuzzy number, where a, b, c
( is a real numbers. The ranking function is generally defined
1, if the ith person is assigned the jth job by [4].
xi j =
0, otherwise
(i + 1) n n
(i + 1)
R Ai = ∑ ai = ∑ ai .
2xi i=1 2i(i + 1)/2 i=1
4. Fuzzy Assignment Problem
5.5 Pascal Triangular Approach
The generalized fuzzy assignment problem can be repre-
We recall the triangular Pascal method for the triangular fuzzy
sented with in the form of n × n fuzzy cost matrix Ai j as given
numbers derived from [5]
in Table 4.1.
The generalized trapezoidal fuzzy
a + 2b + c
  numbers cost or time P Â = .
(1) (2) (3) (4)
[Ai j ] Ai j = Ai j , Ai j Ai j , Ai j ; wi j . 4

432
 Comparative study of fuzzy assignment problem with various ranking — 433/434

Table 5.1
Optimal
Solution
S. No. Ranking Ranking Formula Matrices Formulation
(By Hungarian
Methods
  Method)
13 10 11 10
R1 12 13 11 12
R(ã) = 0.5 aLα , aUα dα


1 Robust’s 
 8 10 10
 46
0 8
Ranking
11 9 11 8
 
13 10 11 10
a + 2b + c 12 13 11 12
2 Pascal P(A) = 
 8 10 10
 46
4 8
Triangular
11 9 11 8
Approach  
13 10 11 10

(i + 1) n (i + 1) n 12 13 11 12
3 Sub Interval R Ai = ∑ ai = ∑ ai 
 8 10 10
 46
2xi i=1 2i(i + 1)/2 i=1 8
Average
11 9 11 8
Method  
1  8.25 6 6.75 6
1 R  7.5 8.25 6.75 7.5
4 Magnitude Mag(a, b, c) = c + 3a − b f (r)dr 
 4.5
 28.5
2 0 6 6 4.5
Ranking
6.75 5.25 6.75 4.5
 
  4.33 3.33 3.67 3.33

a+b+c w 4.00 4.33 3.67 4.00
5 Centroid R Ã = 
2.67 3.33
 15.33
3 3 3.33 2.67
Ranking
3.67 3.00 3.67 2.67

6. Numerical Example References

Here we are going to solve fuzzy assignment problem to [1] R.Q. Mary and D. Selvi, Solving Fuzzy Assignment Prob-
allocate 4 jobs to different 4 machines, the fuzzy assignment lem Using Centroid Ranking Method. International Jour-
cost Ci j is given below nal of Mathematics And its Applications, 55(2018) 7–13.
[2] D. Selvi and G. Velammal, Method for Solving Fuzzy As-
 
(10, 13, 16) (7, 10, 13) (8, 11, 14) (7, 10, 13) signment Problem Using Magnitude Ranking Technique,
 (9, 12, 15) (10, 13, 16) (8, 11, 14) (9, 12, 15) International Journal of Applied and Advanced Scientific
 
 (5, 8, 11) (7, 10, 13) (7, 10, 13) (5, 8, 11)  Research, (2017), 16–20.
[3] K. Kalaiarasi, S. Sindhu and M. Arunadevi, Optimization
(8, 11, 14) (6, 9, 12) (8, 11, 14) (5, 8, 11)
of fuzzy assignment model with triangular fuzzy numbers
Solution: Given Triangular fuzzy cost matrix is balanced using Robust Ranking technique, International Journal
one. Above problem can be solved by using Centroid Rank- of Innovative Science, Engg. Technology, 1(3)(2014), 10–
ing Method, Magnitude Ranking Method, Robust’s Ranking 15.
[4] J.P. Singh and N.I. Thakur, A novel method to solve
Technique, Sub Interval Average Method and Pascal Triangu-
lar Approach (See Table 5.1). assignment problem in fuzzy environment, Industrial
Engineering Letters, 5(2)(2015), 31–35.
[5] S.K. Khaddar Babu and K. Karthikeyen, Statistical
Optimization for Generalized Fuzzy Number, Interna-
7. Conclusion
tional Journal of Modern Engineering Research (IJMER),
In this paper, the fuzzy assignment problem for the place- 3(2)(2013), 647–651.
[6] V. Ramadas, and P. Shanmugam, An Effective Fuzzy
ment of four candidates for four totally different posts is
successfully resolved. The comparisons of the completely Ranking Method In Convex Triangular Fuzzy Soft Num-
different ranking methods are successfully comparable. The bers, International Journal of Applied and Advanced Sci-
numerical examples are strengthening to the comparisons. entific Research. 3(1)(2018), 2456–3080.
[7] Stephen Dinakar, Kamalanathan and Rameshan, Sub in-
terval average method for ranking of linear fuzzy num-

433
 Comparative study of fuzzy assignment problem with various ranking — 434/434

bers, International Journal of Pure and Applied Mathe-

matics, 4(6)(2017), 119–130.
[8] T.A. Thakra, O.K. Chaudhri and N.R. Dhawade, Place-
ment of Staff in LIC using Fuzzy Assignment Problem,
International Journal of Mathematics Trends and Tech-
nology, 53(4)(2018), 259–266.
[9] X. Wang, Fuzzy optimal assignment problem, Fuzzy
Math., 3(1987), 101–108.

?????????
ISSN(P):2319 − 3786
Malaya Journal of Matematik
ISSN(O):2321 − 5666
?????????

434