Professional Documents
Culture Documents
Materials Today: Proceedings: K. Kalaiarasi, H. Mary Henrietta, M. Sumathi
Materials Today: Proceedings: K. Kalaiarasi, H. Mary Henrietta, M. Sumathi
a r t i c l e i n f o a b s t r a c t
Article history: The paper discusses about minimizing the total cost in an economic order quantity by applying geometric
Received 24 November 2020 programming. In order for a smooth functioning ofinventory management system the assortment of
Accepted 7 December 2020 stocks and various costs along with demand have to be perfectly balanced. The sustainable model con-
Available online xxxx
siders the carbon emissions with limitations on warehouse capacity. The parameter demand is fuzzified
by using hexagonal fuzzy numbers and varying the defuzzification methods and results are analyzed.
Keywords: Numerical calculations are done by applying python programming. Also, sensitivity analysis is done
Economic order quantity (EOQ)
for the demand parameter and compared among with three defuzzification methods donein both crisp
Hexagonal fuzzy numbers
Ranking method
and fuzzy environment.
Graded-mean Ó 2020 Elsevier Ltd. All rights reserved.
Signed-distance Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in
Python Materials Science, Technology and Engineering.
https://doi.org/10.1016/j.matpr.2020.12.209
2214-7853/Ó 2020 Elsevier Ltd. All rights reserved.
Selection and peer-review under responsibility of the scientific committee of the Emerging Trends in Materials Science, Technology and Engineering.
Please cite this article as: K. Kalaiarasi, H. Mary Henrietta and M. Sumathi, Comparison of hexagonal fuzzy numbers with varied defuzzification methods in
optimization of EOQ using GP technique, Materials Today: Proceedings, https://doi.org/10.1016/j.matpr.2020.12.209
K. Kalaiarasi, H. Mary Henrietta and M. Sumathi Materials Today: Proceedings xxx (xxxx) xxx
qðBH þ alÞ
1 ð1Þ
2
The dual problem is taken as
b1 b b b4
BH þal 2 FðBT þadÞ 3 BH þal
Max dðbÞ ¼ FðBTqþadÞ 2b2 b3 M 1 2M 2 b
– (2)
4
Fig. 1. Comparison.
With
b1 þ b2 ¼ 1
In this paper, the crisp inventory model is solved using the geo-
metrical programming method and hexagonal fuzzy numbers are b1 þ b2 b3 þ b4 ¼ 0
used to solve the fuzzy inventory model with three different
The degree of difficulty for the problem is 4 1 1 ¼ 2 which
defuzzification methods. We have applied python coding for calcu-
will result in having an infinite number of solutions.
lating the crisp and fuzzified values which reduce the time con-
sumed when compared with manual calculations. The program is b2 ¼ 1 b1
designed in a way that the user can choose any of the three cases
directly. b1 þ ð1b1 Þ b3 þ b4 ¼ 0
2
K. Kalaiarasi, H. Mary Henrietta and M. Sumathi Materials Today: Proceedings xxx (xxxx) xxx
1=2 1=2 F
b1 b2 F 1 þF 2 þF 3 þF 4 þF 5 þF 6
ðB b3
F ðBT þ adÞ 1 BH þ al 1 b4 þ b3 1 þF 2 þF 3 þF 4 þF 5 þF 6 ðBT þadÞ BH þal T þadÞ
¼e Max dðwÞ ¼ 6 6
2F ðBT þ adÞ 1 þ b4 b3
q 2b2 b3 M 1
M1 b3
b4
BH þal
1=2 1=2 2M2 b4
BH þ al 1 2F ðBT þ adÞ 1 þ b4 b3 with
¼e
2M 2 b4 BH þ al 1 b4 þ b3
b1 þ b2 ¼ 1
F ðBT þ adÞðBH þ alÞ b1 þ b2 b3 þ b4 ¼ 0
¼ e2
2M 1 M2 b3 b4
!1þb42b3 1b42þb3
F ðBT þ adÞðBH þ alÞ 2 F 1 þF 2 þF 3 þF 4 þF 5 þF 6
ðBT þ adÞ BH þ al
) b3 b4 ¼ dðb3 ; b4 Þ ¼ 6
2M1 M 2 e2 1 þ b4 b3 1 b4 þ b3
F þF þF þF þF 5 þF
!b 3 b
F ðBT þ adÞ
1 2 3
6
4 6
ðBT þ adÞ BH þ al 4
¼ b1 d ðbÞ b3 M1 2M 2 b4
q
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
qðBH þ alÞ 2 F 1 þF 2 þF 3 þF 4 þF 5 þF 6
ðBT þ adÞ
¼ b2 d ðbÞ q¼ 6
2 ðBH þ alÞ
F ðBT þ adÞ s ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
¼ b1 dðbÞ F 1 þF 2 þF 3 þF 4 þF 5 þF 6
q ½ð B T þ a d Þ ½B H þ a l
minTC ¼ 6
0:5
qðBH þ alÞ
¼ b2 dðbÞ
2
3.3. Approach (ii)
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2F ðBT þ adÞb2 Fuzzy approach through hexagonal fuzzy numbers and defuzzi-
q¼
ðBH þ alÞb1 fication was carried out by signed-distance method
F
!1þb42b3
2 1 þ3F 2 þ2F 3 þ2F 4 þ3F 5 þF 6
ðBT þ adÞ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
dðb3 ; b4 Þ ¼ 12
2 F 1 þ3F 2 þ2F 312 þ2F 4 þ3F 5 þF 6
ðBT þ adÞ
1 þ b4 b3 q¼
!b 3 ðBH þ alÞ
1b4 þb3 F þ3F 2 þ2F 3 þ2F 4 þ3F 5 þF 6
BH þ al 2 1
12
ðBT þ adÞ
min
1 b4 þ b3 b3 M1 sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
b F þ3F þ2F þ2F þ3F 5 þF
BH þ al 4 1 2 3 4 6
½ðBT þ adÞ½BH þ al
TC ¼ 12
2M 2 b4 0:5
Table 1
4. Python approach
import math
print(‘‘Enter the crisp values of the parameters”) The numerical calculations for crisp and fuzzy EOQ are esti-
Bt=float(input(‘‘Bt=”)) mated using the below-mentioned python coding in Table 1 and
a=float(input(‘‘alpha=”)) Table 2. The program gets the input for all the parameters used
g=float(input(‘‘sigma=”)) in the total cost and computes the hexagonal fuzzy numbers with
Bh=float(input(‘‘Bh=”))
three different defuzzification methods and the results are tabu-
u=float(input(‘‘meu=”))
while True: lated. The version of python used 3.7.5 [MSC v.1916 64 bit
print(‘‘enter the F value”) (AMD64)] on win32 (Table 3).
f=float(input(‘‘F=”))
print(‘‘enter the fuzzified value of F ”)
f1=float(input(‘‘F1=”)) 5. Numerical Illustration:
f2=float(input(‘‘F2=”))
f3=float(input(‘‘F3=”)) The crisp values for the parameters are taken as Wang and Ye
f4=float(input(‘‘F4=”))
(2018) and fuzzified using hexagonal fuzzy numbers and defuzzied
f5=float(input(‘‘F5=”))
f6=float(input(‘‘F6=”)) under three cases with a sensitivity analysis
nu=2*f*(Bt+(a*g)) F ¼ 5000,BT ¼ 1000=deliv ery, BH ¼ 5=unit=year, a ¼ 30=ton,
din=(Bh+(a*u)) d ¼ 0:4ton=deliv ery, l ¼ 0:04ton=unit=year.The estimated crisp
bot=nu/din
value q ¼ 1277:59 and TC ¼ 7921:11 and compared with the fuzzi-
print(‘‘crisp value”)
print(‘‘final value(q)=”,math.sqrt(bot)) fied values.
nu=f*(Bt+(a*g))*(Bh+(a*u))
tot=nu/0.5
print(‘‘final value(tc)=”,math.sqrt(tot))
6. Conclusion
print(‘‘——————————————————————————————————”)
print(‘‘1.ranking method”) The minimization of the total cost of an economic order quan-
print(‘‘case 1:”) tity by using geometric programming was derived and applying
nu=2*((f1+f2+f3+f4+f5+f6)/6)*(Bt+(a*g))
hexagonal fuzzy numbersfor fuzzification of the demand and
din=(Bh+(a*u))
bot=nu/din defuzzification by three different defuzzification methods such as
print(‘‘final value(q)=”,math.sqrt(bot)) ranking method, signed-distance and graded-mean representation
nu=((f1+f2+f3+f4+f5+f6)/6)*(Bt+(a*g))*(Bh+(a*u)) method by using python programming. A numerical comparison
tot=nu/0.5 was done between fuzzy and crisp forms in a non-linear program-
print(‘‘final value(tc)=”,math.sqrt(tot))
ming problem and found that all the values are exactly equal.
4
K. Kalaiarasi, H. Mary Henrietta and M. Sumathi Materials Today: Proceedings xxx (xxxx) xxx
Table 3
Sensitivity Analyses.
Variations Values Hexagonal Numbers Crisp output Ranking method Signed distance Graded-mean
Q TC q TC Q TC q TC
50% 2500 (2350,2400,2450,2550,2600,2650) 903.398 5601.071 903.398 5601.071 903.398 5601.071 903.398 5601.071
25% 3750 (3450,3550,3650,3850,3950,4050) 1106.432 6859.883 1106.432 6859.883 1106.432 6859.883 1106.432 6859.883
10% 4500 (4200,4300,4400,4600,4700,4800) 1212.036 7514.625 1212.036 7514.625 1212.036 7514.625 1212.036 7514.625
No 5000 (4700,4800,4900,5100,5200,5300) 1277.598 7921.111 1277.598 7921.111 1277.598 7921.111 1277.598 7921.111
+ 10% 5500 (5200,5300,5400,5600,5700,5800) 1339.956 8307.731 1339.956 8307.731 1339.956 8307.731 1339.956 8307.731
+ 25% 6250 (6100,6150,6200,6300,6350,6400) 1428.398 8856.071 1428.398 8856.071 1428.398 8856.071 1428.398 8856.071
+ 50% 7500 (7200,7300,7400,7600,7700,7800) 1564.732 9701.34 1564.732 9701.34 1564.732 9701.34 1564.732 9701.34
Declaration of Competing Interest comparison with Lagrangian and Kuhn-Tucker method through sensitivity
analysis, J. Model Based Res. 1 (3) (2020) 1–12.
[11] K. Kalaiarasi, M. Sumathi, H. Mary Henrietta, Optimizing EOQ using geometric
The authors declare that they have no known competing finan- programming with varying fuzzy numbers by applying python, J. Critical Rev.
cial interests or personal relationships that could have appeared 7 (18) (2020) 596–603.
[12] G.A. Kochenberger, Inventory models: optimization by geometric
to influence the work reported in this paper.
programming, Decision Sci. 2 (2) (1971) 193–205.
[13] W.J. Lee, Optimal order quantities and prices with storage space and inventory
References investment limitations, Comput. Ind. Eng. 26 (3) (1994) 481–488.
[14] G.S. Mahapatra, T.K. Mandal, G.P. Samanta, EPQ model with fuzzy coefficient of
[1] C.S. Beightler, D.T. Philips, Applied geometric programming, John Wiley and objective and constraint via parametric geometric programming, IJOR 17 (4)
Sons, New York, 1976. (2013) 436–448.
[2] Y.-C. Chen, A probabilistic approach for traditional EOQ model, J. Inform. [15] K.S. Park, Fuzzy Set Theoretical Interpretation of economic order quantity, IEEE
Optim. Sci. 24 (2003) 249–253. Trans. Systems Man. Cybernet SMC 17 (1987) 1082–1084.
[3] D. SetiyaWidodol, Dana Marsetiya Utama, Analisis Model Sustainable Eonomi [16] E.L. Peterson, Geometric programming, Siam Rev. 18 (1974) 37–46.
Order Quantity DenganMempertimbangkanEmisiKarbon Dan Batasan [17] T.K. Roy, M. Maiti, A fuzzy EOQ model with demand-dependent unit cost under
Kapasitas Gudang UntukMenekan Total BiayaPersediaan, Teknik 40 (3) limited storage capacity, Eur. J. Oper. Res. 99 (2) (1997) 425–432.
(2019) 169–175. [18] S. Wang, B. Ye, A comparison between just-in-time and economic order
[4] R.J. Duffin, Cost Minimization Problems Treated by Geometric Means, Oper. quantity models with carbon emissions, J. Cleaner Prod. 187 (2018) 662–671.
Res. 10 (5) (1962) 668–675. [19] X. Wang, W. Tang, R. Zhao, Random fuzzy EOQ model with imperfect quality
[5] R.J. Duffin, E.L. Peterson, C.M. Zener, Geometric programming – theory and items, Fuzzy Optim Decis Making 6 (2) (2007) 139–153.
application, John Wiley, New York, 1967. [20] R. Wilson, A scientific routine for stock control, Harvard Business Review 13
[6] K. Durai, A. Karpagam, A new membership function on hexagonal fuzzy (1934) 116–128.
numbers, Int. J. Sci. Res. 5 (2016) 1129–1131. [21] L.A. Zadeh, Fuzzy sets, Inf. Control 8 (3) (1965) 338–353.
[7] F. Harris, Operations and cost, AW Shaw Co., Chicago, 1913. [22] H.-J. Zimmerman, Using fuzzy sets in operational research, Eur. J. Oper. Res. 13
[8] C.H. Hsieh, Optimization of fuzzy production inventory models, Inf. Sci. 146 (3) (1983) 201–216.
(2002) 29–40. [23] C.M. Zener, A Mathematical aid in optimizing engineering design, Proc.
[9] T.R. Jefferson, C.H. Scott, Avenues of geometric programming. New Zealand, National Acad. Sci. USA 47 (4) (1961) 537–539.
Oper. Res. Int. J. 6 (1978) 109–136. [24] C.M. Zener, A further mathematical aid in optimizing engineering design, Proc.
[10] K. Kalaiarasi, M. Sumathi, H. Mary Henrietta, A. Stanley Raj, Determining the National Acad. Sci. USA 48 (4) (1962) 518–522.
efficiency of fuzzy logic EOQ inventory model with varying demand in