Integrated Intelligent Research (IIR) International Journal of Computing Algorithm

Volume: 04 Issue: 02 December 2015 Pages:76-78

ISSN: 2278-2397

Solving Assignment Problem Using L-R Fuzzy

Numbers
S.JohnsonSavarimuthu1, K.Babu2
1
Associate Professor, Department of Mathematics, St.Joseph’s college of Arts and Science Cuddalore, India.
2
Department of Mathematics, St.Joseph’s college of Arts and Science Cuddalore, India.
Email: [email protected]

Abstract-In this paper we determine a new method to solving A fuzzy number Ā=(m,n,ɑ,β) L-R is said to be an L-R fuzzy
assignment problem using L-R fuzzy parameters. This method number if . Where L and R are continuous, non-increasing
requires finding out the minimal cost reach optimality compared to the functions that define the left and right shapes of µ A(x)
existing methods available in the literature. Numerical examples show respectively and L(0)=R(0)=1.
that the fuzzy assignment ranking method offers an effective way for
handling the fuzzy assignment problem.

Keywords-Assignment problem, fuzzy quantifier, L-R fuzzy number,

optimal solution, λ-cut

I. INTRODUCTION

In this work we investigate a more assignment problem

namelyrealastic the assignment problem with fuzzy costs or
time since the objective values of the function by ranking
method. A method named YAGER’S method is proposed to
find fuzzy optimal solution of fuzzy assignment problems by
representing all parameters as fuzzy quantifiers which are
replaced by L-R fuzzy numbers. The advantages of the A. Definition
proposed method solved and the results obtained are also Let Ā=(m,n,⍺,β)L-R be an L-R fuzzy number. Then the crisp set
discussed the proposed method is easy to understand and to A⍺ = {x∈X:µĀ(x)≥ λ} and λ be a real number in the interval
apply for finding an optimal solution of assignment problem in [0,1] then [m-⍺ L-1(λ) , n+βR-1(λ)] is said to be λ-cut of Ā.
real life situation or a vagueness in everyday life.
B. Yager’s ranking approach
II. DEFINITION A number of ranking approaches have been proposed for
comparing fuzzy numbers. In this paper, Yager’s ranking
A fuzzy set is characterized by a membership function approach is used for fuzzy numbers. This approach involves
mapping elements of a domain X→[0,1] is a mapping called relatively simple computational and is easily understandable.
the degree of membership value of x∈X in the fuzzy set A. This approach involves a procedure for ordering fuzzy
Those membership grades are often represented by real numbers in which a ranking index Ɍ(Ã)is calculated for an LR
numbers ranging from [0,1]. flat fuzzy number à =(m,n,⍺,β)LR from its λ-cut Aλ= [m-⍺L-1(λ),
n+βR-1(λ)] according to the following formula.
A. Triangular fuzzy number 1 1
1
Triangular fuzzy number is defined on A =(a, b, c) are real Ɍ(Ã)= (
2 ∫ m-⍺L-1(λ)dλ) +∫ ¿ ¿+βR (λ))dλ)
-1

numbers and its member function is given below 0 0

~
Let à and B be two LR flat fuzzy numbers then
(i)
~ ~
Ã≥ B if Ɍ(Ã)≥ Ɍ( B )
(ii)
~ ~
ÿ B if Ɍ(Ã)¿Ɍ( B )
(iii)
~ ~
Ã= B if Ɍ(Ã)=Ɍ( B )

C. Linearity property of Yager’s ranking index

~
Let Ã=(m1,n1,⍺1,β1)LR and B =(m2,n2,⍺2,β2)LR be two LR flat
fuzzy numbers and k1,k2 be two non negative real numbers.
~
Using definition, the λ-cut AλandBλcorresponding à and B are
III. DEFINITION Aλ = [m1-⍺1 L-1(λ), n1+β1 R-1(λ)] and Bλ=[m2-⍺2 L-1(λ), n2+β2 R-
1
(λ)] using the property, (δ1A1 + δ2A2)λ= δ1(A1)λ + δ2(A2)λ ∀

1
Integrated Intelligent Research (IIR) International Journal of Computing Algorithm
Volume: 04 Issue: 02 December 2015 Pages:76-78
ISSN: 2278-2397

δ1, δ2 ∈ (is the set of real numbers), the λ-cut (k1A Solution- First we show that how to calculate Y (4,6,7,9) by
~
+k2A)corresponding to k1Ã +k2 B is applying definition(3.2)Yager’s Ranking method.
(k1A+k2B)λ= [k1m1 + k2m2 –(k1⍺1+k2⍺2)L-1(λ), k1n1 +k2n2 +(k1β1
+k2β2)R-1(λ)] The membership value of first we choose L-R fuzzy number
~
usingYager’s ranking index Ɍ(k1Ã+k2 B ) corresponding to
(4,6,7,9) is
1 1
~ 1
fuzzy number (k1Ã+k2 B ) is:
1 1
Ɍ(Ã)= (
2 ∫ m-⍺L-1(λ)dλ) +∫ ¿ ¿+βR-
~ 1 0 0

2 ∫
Ɍ(k1Ã+k2 B ) = k1( m1-⍺1L-1(λ)dλ) +∫ ¿ ¿ 1+β1 R-1(λ))dλ ) 1
(λ))dλ)
0 0 1
1 1 Ɍ(c11)= ¿(
1 2
+ k2(∫ m2-⍺2L-1(λ)dλ) +∫ ¿ ¿ 2+β2 R-1(λ))dλ )
2 0 1 1
~
=k1Ɍ(Ã)+k2Ɍ( B )
0
∫ 4−7( 12 )dλ ¿+∫(6+9( 12 ) dλ)¿]
0 0

~ ~ 1
Similarly it can be proved that Ɍ(k1Ã+k2 B ) = k1Ɍ(Ã)+k2Ɍ( B ) = ¿(
2
1 1

Table 1.Fuzzy assignment costs ∫ 2

0
4−(
7
) dλ ¿+∫ (6+ 92 dλ) ¿
0
()
1
Person = ¿(
→ 2
1 1

( )
P1 P2 … Pj … Pn
Job↓
∫ ( 2 ) dλ ¿+∫ 12+9
8−7
2
dλ ¿ ¿ ¿
0 0
J1 c11 c12 … cij … c1n
1 1

⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞ = ¿(
1
2 ∫ ( 12 )dλ ¿+∫ ( 21
0 0 2
dλ)¿ ( )
Ji cj1 cj2 … cjj … cjn 1
1
⁞ ⁞ ⁞ ⁞ ⁞ ⁞ ⁞
= ¿(
2 ∫ ( 22
2
)dλ ¿
0
1
1
= ¿(∫ (11 λ)dλ ¿
Jn cn1 cn2 … cnj … cnn

Where, cij = 2 0
(mij,nij,⍺ij,βij)LR =5.5

Example-Let us consider a fuzzy assignment problem rows Proceeding by similarly Yager’sRanking method
representing 5 persons P1, P2, P3, P4, P5 and columns
representing 5 jobs Job1, job2, job3, job4, job5. The problem Ɍ(c12)=2, Ɍ(c13)=2.67, Ɍ(c14) =6, Ɍ(c15)=6.5,
can find the minimal cost. Ɍ(c21)=5.58, Ɍ(c22)=5.92, Ɍ(c23)=0.5, Ɍ(c24)=5.5, Ɍ(c25)=9.67,
Ɍ(c31)=5.25, Ɍ(c32)=2.67,Ɍ(c33)=8.5,Ɍ(c34)=5.5,Ɍ(c35)=2.38
Table:2 Ɍ(c41)=11.75, Ɍ(c42)=7,Ɍ(c43)=5.67,Ɍ(c44)=3.67,Ɍ(c45)=8.5,
C P1 P2 P3 P4 P5 Ɍ(c51)=1.12,Ɍ(c52)=3.67, Ɍ(c53)=4.75,Ɍ(c54)=5.38, Ɍ(c55)=1.75.
ij
Ɍ(c*1 2) = Ɍ(c*23)=Ɍ(c*35) = Ɍ(c*44) = Ɍ(c*51)= 1
Ɍ(c*11) =Ɍ(c*13) = Ɍ(c*14) =Ɍ(c*15) =Ɍ(c*21) = Ɍ(c*22) =Ɍ(c*23) =Ɍ(c*25)
=Ɍ(c 31) =Ɍ(c 32) = Ɍ(c 33) =Ɍ(c 34) = Ɍ(c 41) = Ɍ(c 42) = Ɍ(c 43)
* * * * * * *
J1 (4,6,7,9) (3,4,6,9 (4,5,7,10 (3,5,6,7 (5,7,10,
LR )LR )LR )LR 12)LR = Ɍ(c*
)
45 = Ɍ(c *
)
52 = Ɍ(c*
)
53 = Ɍ(c* )
54 = Ɍ(c *
)
55 = 0

J2 (4,7,10,1 (3,7,10, (2,3,10,1 (3,7,9,1 (4,10,13 Table:3 optimal table

4)LR 13)LR 2)LR 3)LR ,15)LR
Cij P1 P2 P3 P4 P5
J3 (5,7,10,1 (2,4,10, (7,9,13,1 (5,7,12, (4,5,7,1
4)LR 13)LR 5)LR 15)LR 1)LR J1 5.5 2 2.67 6 6.5

J4 (7,9,13,1 (6,7,9,1 (7,9,10,1 (6,7,10, (7,9,10, J2 5.58 5.9 0.5 5.5 9.67
4)LR 0)LR 3)LR 13)LR 12)LR 2

J5 (2,3,5,9) (5,6,7,1 (4,6,9,12 (5,7,9,1 (2,3,5,7) J3 5.25 2.6 8.5 5.5 2.38
LR 0)LR )LR 3)LR LR 7

J4 11.75 8 7 5.6 8.5

2
Integrated Intelligent Research (IIR) International Journal of Computing Algorithm
Volume: 04 Issue: 02 December 2015 Pages:76-78
ISSN: 2278-2397

J5 1.12 3.6 4.75 5.3 1.75

7 8

With optimal objective value Ɍ(Cij) = 11.67

Which represents a optimal cost in other words the optimal
assignment is

J1→ P2, J2→ P3, J3→ P5, J4→ P4, J5→ P1

The fuzzy optimal cost is
C12 + C23 + C35 +C44 + C15 = (3,4,6,9)LR + (2,3,10,12)LR +
(4,5,7,11)LR +(6,7,10,13)LR
+ (2,3,5,9)LR
= 2 + 0.5 + 2.38 + 5.67 + 1.12
= 11.67

IV. CONCLUSION

Thus we have recalled some basic definitions and we have

considered fuzzy assignment problem and introduced the
yager’s ranking method solution for these problems using
triangular fuzzy numbers, trapezoidal fuzzy numbers, L-R
fuzzy parameters and also, we have proposed a new algorithm
for solving these problems directly by use of yager’s ranking
function.Finally, we have presented a method deal with
sensitivity analysis of fuzzy assignment problem, and the
presented method is illustrated with help of some numerical
examples. In sensitivity analysis, the fuzzy optimal solution is
again minimal cost, even though some variations are
considered in the assignment problems. As this method
consumes less time and is very easy to understand and apply,
so it will be very helpful for decision makers who are dealing
with logistic and supply chain problems.

REFERENCES
[1] L. A ,Zadeh, fuzzy sets, information and control, vol. 8,1965,pp. 338-
353.
[2] R.Nagarajan and A.Solairaju computing improved fuzzy optimal
Hungarian assignment problems with fuzzy cost under Robust Ranking
Techniques IJCA vol.6-no.4 sep2010.
[3] A.Nagoorgani and V.N. Mohamed ,Solution of a Fuzzy Assignment
Problem by using a New Ranking Method IJFMA vol.2, 2013, 8-16 , 6
june 2013.
[4] Andreae T. 2001. On the travelling salesman restricted to inputs
satisfying a relaxedInequality. Networks, 38:59-67