Finalized Basepaper
Finalized Basepaper
A R T I C L E I N F O A B S T R A C T
Keywords: In this paper, a deterministic fuzzy inventory model with quadratic demand pattern, time-
Multi item deterministic model dependent holding cost, and time-dependent deterioration has been created. We have taken the
Quadratic demand rate of shortage as fully backlogged. In traditional inventory model, most of the cost parameters
Time dependent deterioration and holding cost
were taken as crisp. But in real world, fuzzy cost parameters are more realistic to appear instead
Hexagonal fuzzy number
Graded mean integration method
of crisp parameters. Due to uncertainty we have considered all the cost parameters as hexagonal
Neutrosophic and pythagorean hesitant fuzzy number. We have regarded all of the cost parameters as hexagonal fuzzy numbers due to
programming uncertainty. Our primary goal is to reduce the total average cost, and for that we’ve covered two
optimization techniques: the Pythagorean Hesitant Fuzzy Programming Approach (PHFPA) and
the Neutrosophic Hesitant Fuzzy Programming approach (NHFP).
Through extensive numerical experiments and sensitivity analysis, the effectiveness and
practicality of the proposed model are demonstrated. Comparative studies with traditional in
ventory models underscore the advantages of the Neutrosophic and Pythagorean Hesitant Fuzzy
Programming Approach in managing multi-item inventory systems with quadratic demand pat
terns, highlighting its robustness in handling uncertain and ambiguous information.
1. Introduction
Now a days it is a big challenge for a company to manage inventory properly. Deterioration is a natural phenomenon, so we can not
avoid its effect while we are developing an inventory model. Spoiled items cannot be used further to make more products. While
storing fruits, vegetables etc. a large amount of its physical material decays and spoils. The procedure of evaporation tells upon the
physical state of highly volatile substances as gasoline, fuel, alcohol, turpentine, etc. All the electronic materials as well as photo
graphic films, radioactive materials, grains etc. also degrade over time and they gradually loss their potential. Therefore it is very much
important for inventory modelers to consider the effect of deterioration while developing a model.
The fuzzy set technique is associated effectively dealing with uncertainty and imprecise information. In 1965, Zadeh [1] first
introduced Fuzzy sets theory. For many real-world problems, the EOQ model is a standard approach. In most cases, researchers have
investigated various EOQ models in a fuzzy environment to optimize the objective function. Inventory control policies play an
important role in the manufacturing process considering many manufacturing systems. Various models have been developed to
achieve a balance between having higher inventory, which causes high holding costs, and having few inventory, which causes stock
* Corresponding author.
E-mail address: [email protected] (K. Das).
https://doi.org/10.1016/j.rico.2023.100367
Received 31 July 2023; Received in revised form 14 October 2023; Accepted 16 December 2023
Available online 20 December 2023
2666-7207/© 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
outs and poor customer service. The main objective is to reduce inventory-related average costs over time. Crisp is not exact but fuzzy
set theory provides more realistic results for inventory problems. The conventional EOQ model presumes that manufactured goods are
non-perishable and have an indefinite useful life. Health kits, chemicals, medicines, volatile substances etc. deteriorate rapidly over
time. As a result, over time, these products are no longer able to perform their primary purpose and quickly become obsolete.
Consequently, the significance of deterioration related losses cannot be avoided.
In 1998, Chang et al. [2], described an Economic reorder point for fuzzy backorder quantity. In 2011, Uthayakumar [3], studied
fuzzy economic production quantity model for weibull deteriorating items with ramp type of demand. In 2010, Kazemi et al. [4]
described an inventory model with backorders with fuzzy parameters and decision variables. Rajoria et al. [5], in 2015 established an
EOQ model for decaying items with power demand, partial backlogging and inflation. After that in 2015, Pal et al. [6] have introduced
a production inventory model for deteriorating item with ramp type demand allowing inflation and shortages under fuzziness. Many
researchers have contributed significantly in recent years to inventory challenges with variable deterioration and slightly varying
demand, which is an important component of inventory systems. It can be a constant, random variable with a known probability
density or it can vary over time based on selling prices, purchasing costs, and customer behavior. The rate of demand for many things
like mobile phones, televisions, attractive clothes etc. fluctuates due to various reasons. Some items have increasing demand
throughout time. On the other hand, the demand for some items has decreased due to more attractive products. Garai et al. [7], in 2019
have studied a fully fuzzy inventory model with price-dependent demand and time varying holding cost under fuzzy decision variables.
Padiyar et al. [8], in 2021, described an inventory system with price dependent consumption for deteriorating items with shortages
under fuzzy environment. In 2021, Srivastava and Bajaj [9], established an optimal inventory management system for deteriorating
items with linear demand, shortages and partial backlogging in a triangular fuzzy setup. Rajput et al. [10] described a fuzzy opti
mization of a production model with CNTFN demand rate under trade-credit policy in 2022.
Furthermore, important situations may happen when the delegation of membership and non-membership degrees is not be
favorable with a single value, however, a set of different possible values may effectively express the uncertainty. For this, Pythagorean-
hesitation fuzzy set is introduced with a combination of Pythagorean fuzzy set and hesitancy fuzzy set respectively. Biswas et al. [11],
in 2016 also presented some distance measures of single valued neutrosophic hesitant fuzzy set and their applications to multiple
attribute decision making. Yager and Abbasov [12] described the Pythagorean membership grades, complex numbers, and decision
making in 2013. In 2009, Torra and Narukawa [13], described about hesitant fuzzy sets and decision. In 2018, Bharati [14] has
discussed hesitant fuzzy computational algorithm for multiobjective optimization problems. In 2014, Ye [15], introduced
multiple-attribute decision-making method under a single-valued neutrosophic hesitant fuzzy environment. In 2018, Ahmad et al. [16]
also designed a single valued neutrosophic hesitant fuzzy computational algorithm for multi objective nonlinear optimization problem.
The uncertainty present in discrete case decision-making problems has been widely studied in the literature and solved efficiently in
the basis of fuzzy sets, intuitive fuzzy sets, non-ambiguous fuzzy sets and Pythagorean fuzzy decision sets etc. Most of the studies based
on Pythagorean fuzzy decision sets are limited to discrete case optimization problems and have not been applied to continuous cases.
Moreover the configuration of Pythagorean-hesitant fuzzy programming which is depending on Pythagorean-hesitant fuzzy decision
set is very new technique to solve multi-objective functions. Liang and Xu [17], in 2017, established the new extension of TOPSIS
method for multiple criteria decision making with hesitant Pythagorean fuzzy sets. Wei and Lu [18], in [2017], have established Dual
hesitant Pythagorean fuzzy amacher aggregation operators in multiple attribute decision making. In 2020, Adhami and Ahmad [19],
have described interactive Pythagorean hesitant fuzzy computational algorithm for multiobjective transportation problem under
uncertainty.
In this present article due to uncertainty, it is very much realistic to take the cost parameters as fuzzy number. But without
defuzzifing the cost parameters we can not optimize an inventory problem. In 2015, Rajarajeshwari et al. [20] have described A New
Operation on Hexagonal Fuzzy Number. Chen and Hsieh [21] studied the graded mean integration representation (GMIR) of general
fuzzy numbers while in 2006, Chen et al. [22], used some notions of GMIR of L-R type fuzzy numbers. In 2015, Deb and De [23] gave an
optimal solution of a fully fuzzy linear fractional programming problem by using graded mean integration representation method. In
2022, Mueen et al. [24], have described performance measure of multiple-channel queueing systems with imprecise data using graded
mean integration for trapezoidal and hexagonal fuzzy number.
In 2023 [25], Alzabut, Jehad & Khan, Hasib & Nie, Yufeng & Alkhazzan, Abdulwasea introduced a stochastic SIRS modeling of
transport-related infection with three types of noises. In 2023 [26], Khan, Hasib, Jehad Alzabut, Wafa F. Alfwzan, and Haseena Gulzar
described Nonlinear Dynamics of a Piecewise Modified ABC Fractional-Order Leukemia Model with Symmetric Numerical Simula
tions. Recently in 2023 [27], Hasib Khan, Jehad Alzabut, Haseena Gulzar, describe Existence of solutions for hybrid modified
ABC-fractional differential equations with p-Laplacian operator and an application to a aterborne disease model.In 2023 [28], Shah
Hussain, Osman Tunç, Ghaus ur Rahman, Hasib Khan, Elissa Nadia, describe a Mathematical analysis of stochastic epidemic model of
MERS-corona & application of ergodic theory. Khan, Hasib, Jehad Alzabut, Haseena Gulzar, Osman Tunç, and Sandra Pinelas [29]
introduced the On System of Variable Order Nonlinear p-Laplacian Fractional Differential Equations with Biological Application in
2023.In 2023 [30], Telli, Benoumran, Mohammed Said Souid, Jehad Alzabut, and Hasib Khan have described The Existence and
Uniqueness Theorems for a Variable-Order Fractional Differential Equation with Delay. Also in 2023 [31], Hasib Khan, Jehad Alzabut,
Osman Tunç, Mohammed K.A. Kaabar established A fractal–fractional COVID-19 model with a negative impact of quarantine on the
diabetic patients.
Abu Arqub in 2017 [32] have described the Adaptation of reproducing kernel algorithm for solving fuzzy Fredholm–Volterra
integrodifferential equations. In 2020 [33], Alshammari, Mohammad, Mohammed Al-Smadi, Omar Abu Arqub, Ishak Hashim, and
Mohd Almie Alias have discussed about Residual Series Representation Algorithm for Solving Fuzzy Duffing Oscillator Equations. In
2021 [34], Jagdev Singh, Banan Maayah, Mohammed Alhodaly have described The Reproducing kernel approach for numerical
2
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
solutions of fuzzy fractional initial value problems under the Mittag–Leffler kernel differential operator. In 2021 [35], Jagdev Singh,
Mohammed Alhodaly have established the Adaptation of kernel functions-based approach with Atangana–Baleanu–Caputo distributed
order derivative for solutions of fuzzy fractional Volterra and Fredholm integrodifferential equations.
In this present article we have considered a deterministic inventory model with cost parameters as hexagonal fuzzy number. We
have solved the inventory model in two different techniques namely neutrosophic hesitant fuzzy programming technique (NHFP) and
Pythagorean hesitant fuzzy programming approach. (PHFPA).
The contribution and background in this research article:
Contribution: This research paper contributes to the field of inventory management by offering a powerful decision-making tool
for practitioners and researchers alike. The incorporation of neutrosophic and Pythagorean hesitant fuzzy sets into the inventory
modeling process provides a more realistic representation of uncertainty, leading to improved inventory control strategies and cost
reduction in dynamic and uncertain business environments.
An inventory model with quadratic demand and time-dependent holding costs and deterioration, solved by neutrosophic hesitant
fuzzy programming (NHFP) and Pythagorean hesitant fuzzy programming approach (PHFPA), is a specialized approach to inventory
management that addresses the complexities and uncertainties associated with inventory systems. Let’s explore the background of this
model and the two techniques-
Background of the Inventory Model:
1. Quadratic Demand: In this model, the demand for the product is assumed to follow a quadratic pattern, meaning that the
demand rate is a function of time that can be represented by a quadratic equation. This is a departure from traditional inventory models
that often assume constant or linear demand.
2. Time-Dependent Holding Costs: The holding costs associated with maintaining inventory are not assumed to be constant but
instead vary with time. These costs may increase or decrease over time due to factors such as storage costs, interest rates, and storage
capacity limitations.
3. Deterioration: Deterioration refers to the reduction in the quality or value of the inventory items over time. This model in
corporates deterioration effects, which are particularly relevant for perishable or time-sensitive products.
4. Neutrosophic Hesitant Fuzzy Programming (NHFP): NHFP is a mathematical programming technique that deals with un
certainty and vagueness in decision-making. Neutrosophic sets are used to represent the degree of truth, indeterminacy, and falsity of
information. Hesitant fuzzy sets are employed when decision-makers are uncertain or hesitant about assigning membership values.
NHFP combines these concepts to handle complex, uncertain, and hesitant information in the context of inventory optimization.
5. Pythagorean Hesitant Fuzzy Programming Approach (PHFPA): Similar to NHFP, PHFPA also deals with uncertain and
hesitant information but uses Pythagorean fuzzy sets. Pythagorean fuzzy sets include both membership degrees and non-membership
degrees, allowing for a more comprehensive representation of uncertainty. PHFPA extends traditional fuzzy programming to address
complex optimization problems under uncertainty.
The combination of the quadratic demand, time-dependent holding costs, and deterioration in this inventory model acknowledges
the real-world complexities faced by organizations in managing their inventory systems. By incorporating these factors, the model
aims to optimize inventory decisions while considering the uncertainties associated with demand and cost parameters.
2. Mathematical preliminaries
Here we have discussed some important definitions which are quite vital for this article.
Fuzzy Sets:
Let us take X as a domain space or universe of discourse. Then the Fuzzy set A
̃ is a subset of X and is defined by
μà : X→[0, 1]
3
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
here,
n
TAn (x) = {σ |σ ∈ T
̃ ̃ An (x)} ∈ [0, 1], is called truth membership hesitant of x ∈ X to the set A
n
Ĩ
An (x) = {τ| ∈ τ TAn (x)} ∈ [0, 1], is called indeterminacy membership hesitant of x ∈ X to the set A
̃
n
F̃
An (x) = {δ|δ ∈ FAn (x)} ∈ [0, 1], is called falsity membership hesitantof x ∈ X to the set A
̃
With,
Where σh (x) ∈ [0,1] is representing the possible Pythagorean hesitant membership (truth) degree of the element x ∈ X to the set Ph .
δh (x) ∈ [0, 1] is representing the possible Pythagorean hesitant nonmembership (falsity) degree of the element x ∈ X to the set Ph .
With 0 ≤ β, γ ≤ 1 and 0 ≤ β2 + γ2 ≤ 1,
Here, β ∈ σh (x), γ ∈ δh (y) ∀ x ∈ X.
Hexagonal Fuzzy Number:
A hexagonal fuzzy number is a type of fuzzy number used in fuzzy logic and fuzzy set theory to represent uncertain or imprecise
values. It is characterized by a hexagonal-shaped membership function that assigns degrees of membership to a particular value within
a given range. The hexagonal membership function typically has six vertices, each representing a specific membership grade.
Hexagonal fuzzy number with an weight (w=1 here), is denoted by
A
̃ H = (a1 ,a2 ,a3 ,a4 ,a5 ,a6 ; 1); Where, {a1 ,a2 ,a3 ,a4 ,a5 and a6 } ∈ R(real number), With the membership function μ (x) is defined by:
ÃH
4
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
⎧
⎪
⎪ 0, x < a1
⎪
⎪ ( )
⎪
⎪
⎪
⎪ 1 x − a1
⎪
⎪ , a1 ≤ x ≤ a2
⎪
⎪ 2 a2 − a1
⎪
⎪
⎪
⎪ ( )
⎪
⎪ 1 1 x − a2
⎪
⎪ + , a2 ≤ x ≤ a3
⎪
⎪
⎨ 2 2 a3 − a2
⎪
μÃH (x) = 1, a3 ≤ x ≤ a4
⎪
⎪ ( )
⎪
⎪ 1 x − a4
⎪
⎪
⎪
⎪ 1− , a4 ≤ x ≤ a5
⎪
⎪ 2 a5 − a4
⎪
⎪ ( )
⎪
⎪
⎪
⎪ 1 a6 − x
⎪
⎪ , a5 ≤ x ≤ a6
⎪
⎪ 2 a6 − a5
⎪
⎪
⎪
⎩ 0, x > a6
3. Mathematical model
Here we will take some standard notations which will help us to formulate the inventory model.
5
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
3.2. Assumptions
Demand(Di ) = a + bt + ct 2
4. Here we have considered deterioration as function time θi (t) = αt. 0 < α < 1.
5. The backlogging rate is fully backlogged.
6. The holding cost is depends on time Hi (t)=hi.t.
Since backlogging rate is fully backlogged, the demand function is similar throughout the time period t ∈ [0,t2i ]. At time t = 0, the
inventory level is its maximum state that is Imax . For customer demand as well as product’s deterioration inventory starts to decrease
throughout the time period t ∈ [0,t1i ]. At time t = t1i inventory level goes to zero and shortages begin from that time. At the end of time
cycle t = t2i , inventory reached to the maximum shortage level. Which is shown in the following Fig. 1.
Here we represent the level of inventory at any time t ∈ [0, t2i ] by Iri (t) where r=1, 2 and i=1, 2,…..,n.
Then the governing differential equations are as follows.
dI1i
+ θi (t)I1 = − Di ; 0 ≤ t ≤ t1i (1)
dt
dI2i
+ Di = 0; t1i ≤ t ≤ t2i . (2)
dt
With the initial conditions,
I1i (0) = Imax ; I1i (t1i ) = 0; I2i (t2i ) = − Si (3)
From (1) & (2) we get,
dI1i ( )
= − αi tI1i − ai + bi t + ci t2 , 0 ≤ t ≤ t1 (4)
dt
dI2i ( )
= − ai + bi t + ci t2 , t1i ≤ t ≤ t2i (5)
dt
Solving (4) & (5) with the initial condition (3) given above we get
{ }
bi ( 2 ) ci ( 3 ) αi a i ( 3 ) bi ( 4 ) ci ( )
I1i (t) = ai (t1i − t) + t1i − t2 + t1i − t3 + t1i − t3 + t1i − t4 + t1i5 − t5 ; 0 ≤ t ≤ t1i (6)
2 3 2 3 4 5
bi ( 2 ) ci ( 3 3 )
I2i (t) = ai (t1i − t) + t − t2 + t1i − t ; t1i ≤ t ≤ t2i (7)
2 1i 3
Now, at t = 0; I1i (t) = Imax .
From (6) we get,
( )
bi ci αi ai 3 bi 4 ci 5
Imax = a1 t1i + t1i2 + t1i3 + t1 + t1i + t1i
2 3 2 3 4 5 (8)
.
Putting t = t2i in I2i (t) we get the total shortages quantity (S)
[ ]
bi ( 2 ) ci ( )
Si = − ai (t1i − t2i ) + t1i − t2i2 + t1i3 − t2i3 (9)
2 5
Total ordering quantity = Imax + Si .
Ordering cost (O.C): Inventory ordering cost, also known as order cost or procurement cost, refers to the expenses incurred by a
6
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
business or organization when they place orders for inventory items. These costs are associated with the process of replenishing stock
levels to meet demand and can be a significant part of the overall inventory management costs.
Let us take the total ordering cost as
O.Ci = Oi for i = 1, 2, ……, n (10)
Holding cost (H.C):
Inventory holding cost, also known as carrying cost, refers to the expenses associated with storing and maintaining inventory
within a business or organization. These costs can be a significant component of the overall cost of managing inventory and are
important to consider when determining the optimal level of inventory to hold.
Total inventory holding cost in the time interval t ∈ [0, t1i ] is given by
∫t1i {( ) ( )}
ai 2 bi 3 ci 4 αi ai 4 bi 5 ci 6
hi tI1i (t)dt = hi t1i + t1i + t1i + t1i + t1i + t1i ; 0 ≤ t ≤ t1i , (11)
2 3 4 2 4 5 5
0
for i = 1, 2, ……, n.
Shortage cost (S.C):
Inventory shortage cost, also known as stock-out cost or stock-out expense, refers to the costs and negative consequences that a
business incurs when it runs out of inventory and is unable to meet customer demand.
Shortage cost for the time t ∈ [t1i , t2i ] is as follows
∫t2i
S.Ci = Si I2i (t)dt Where i = 1, 2, ……, n. (12)
t1i
[ { } { } { }
1( 2 ) bi 2 1( 3 ) ci 1( 4 )
= − Si ai t1i (t2i − t1i ) − t2i − t1i2 + t1i (t2i − t1i ) − t2i − t1i3 + t1i 3 (t2i − t1i ) − t2i − t1i4 ; t1i ≤ t ≤ t2i , for i
2 2 3 3 4
= 1, 2, ……, n (13)
Considering total average cost for each item as objective function, we have a Multi-Objective Inventory Model (MOIM) given below
Minimize {TAC1(t11 , t21 ), TAC2(t12 , t22 ), …………………., TACn(t1n , t2n )} (15)
[ {( 2 ) αi ( ai 4 )} { }
t1i ≥ 0, t2i ≥ 0where, TACi (t1i , t2i ) = t12i Oi + hi a2i t1i + b3i t1i
3
+ c4i t1i
4
+ 2 4 t1i + b5i t1i 5
+ c5i t1i
6
− Si [ai t1i (t2i − t1i ) − 1 2
(t
2 2i
− t1i
2
) +
bi
{2 } {3 } ( 3 )
2
t1i (t2i − t1i ) − 13 (t2i
3
− t1i
3
) + c3i t1i (t2i − t1i ) − 14 (t2i
4
− t1i
4
) + di a3i t1i + b4i t1i
4
+ c5i t1i
5
, for i = 1, 2, ……, n.
It is very much obvious that cost parameter’s value (ordering cost, holding cost, shortages cost, deterioration cost) may change after
certain interval of time. To tackle such types of problems we can consider fuzzy number for these parameters. Due to uncertainty we
are considering hexagonal fuzzy number for all of the costs parameters. Here ‘w’ represents the weight for the corresponding fuzzy
costs parameters.
( )
Hexagonal Fuzzy ordering cost Õi = O1 , O2 , O3 , O4 , O5 .O6 ; wOi , 0 ≤ wOi ≤ 1;
i i i i i i
7
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
( )
Hexagonal Fuzzy holding cost ̃
hi = h1i , h2i , h3i , h4i , h5i , h6i ; whi , 0 ≤ whi ≤ 1;
( )
Hexagonal Fuzzy shortage cost S̃i = Si1 , Si2 , Si3 , Si4 , Si5 , Si6 ; wSi , 0 ≤ wSi ≤ 1;
( )
Hexagonal Fuzzy deterioration cost d̃i = di1 , di2 , di3 , di4 , di5 , di6 ; whi , 0 ≤ wdi ≤ 1;
Having these hexagonal fuzzy number our crisp inventory model (15) transform into the following fuzzy model.
{( ) ( )} [ { }
̃ i (t1i , t2i ) = i [Õi + h̃i ai 2 bi 3 ci 4 αi ai 4 bi 5 ci 6 1( 2 )
TAC t1i + t1i + t1i + t1i + t1i + t1i − S̃i ai t1i (t2i − t1i ) − t2i − t1i2
t2i 2 3 4 2 4 5 5 2
{ } { } ( )
bi 2 1( 3 3
) ci 3 1( 4 4
)
̃ ai 3 bi 4 ci 5
+ t1i (t2i − t1i ) − t2i − t1i + t1i (t2i − t1i ) − t2i − t1i + di t1i + t1i + t1i ; for i
2 3 3 4 3 4 5
= 1, 2, ……, n. (16)
Now we are interested in finding the solution for our inventory model. Which is given by the following equation.
̃1 (t11 , t21 ), TAC
Min {TAC ̃2 (t12 , t22 ), ………………., TAC
̃n (t1n , t2n ) (17)
t1i ≥ 0, t2i ≥ 0 .
[ {( ) ( )} { } {
Where, TAC ̃ i (t1i , t2i ) = i O ̃i + h̃i ai t2 + bi t 3 + ci t 4 + αi ai t 4 + bi t 5 + ci t6 − S̃i [ai t1i (t2i − t1i ) − 1 2 2
− t1i bi 2
t1i
t2i 2 1i 3 1i 4 1i 2 4 1i 5 1i 5 1i 2 (t2i ) + 2 (t2i −
} { } ( )
t1i ) − 13 (t2i
3 3
− t1i ) + c3i t1i 3 (t2i − t1i ) − 14 (t2i4
− t1i4
) + d̃i a3i t1i
3
+ b4i t1i
4
+ c5i t1i
5
; for i = 1, 2, ……, n.
Graded Mean Integration Method (Defuzzyfication technique)
1 1
Let us take a hexagonal fuzzy numbers A ̃ H (z) = (a1 , a2 , a3 , a4 , a5 , a6 ; w). Let L− are R− are the left and right inverse function
respectively. Then we can define the graded mean integration as
∫w ( − 1 ) / ∫w
L (x) + R− 1 (x)
̃H ) =
P(A h dh hdh, (18)
2
0 0
t1i ≥ 0, t2i ≥ 0.
[ {( ) ( )} { } {
Where, TAC ̂ ai t2 + bi t 3 + ci t4 + αi ai t4 + bi t 5 + ci t6 1 2 2 bi 2
t2i Oi + hi − Ŝi [ai t1i (t2i − t1i ) − − t1i t1i
̂ i (t1i , t2i ) = i ̂
2 1i 3 1i 4 1i 2 4 1i 5 1i 5 1i 2 (t2i ) + 2 (t2i −
} { } ( )
t1i ) − 13 (t2i
3 3
− t1i ) + c3i t1i 3 (t2i − t1i ) − 14 (t2i
4 4
− t1i ) + d̂i a3i t1i
3
+ b4i t1i
4
+ c5i t1i
5
; for i = 1, 2, ……, n.
8
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
5. Solution procedure
To solve (21) we will solve one objective first and ignore the others. Now using lingo software we will find the value for each
objective function. Having these total average cost we now formulate the following pay-of matrix.
TAC1 (t11 , t21 ) TAC2 (t12 , t22 ) …… … TACn (t1n , t2n )
⎤
⎡
( 1 1) ( 1 1) ( 1 1)
( 1 1
)
TAC1∗ t11 , t21 TAC2 t11 , t21 …….. TACn t11 , t21 ⎥
t11 , t21⎢
( 2 2) ⎢ ( 2 2) ( ) ( )⎥
⎥
t12 , t22 ⎢ TAC1 t12 , t22 TAC1∗ t122 , t22
2
……… TACn t122 , t22
2
⎥
⎢ ⎥
⎢ ⎥
………… ⎢ ……… ⎢ ………… ………….. …………… ⎥ ⎥
⎣ ⎥
t122 , t22
2
) ( ) ( ) ( 2 2 )⎦
TAC1 t122 , t22
2
TAC2 t122 , t22
2
………… ∗
TACn t12 , t22
Using (22), now we are defining truth-hesitant membership (or hesitant membership) function, indeterminacy-hesitant mem
bership function, and falsity-hesitant membership (or hesitant non-membership) function as follows
Truth hesitant-membership function:
⎧
⎪
⎪ 1
⎪
⎪ ( k )t
⎨ U − (TAC k (t1k , t2k ))
t for TACk (t1k , t2k ) < Lk
ThE−1 (TACk (t1k , t2k )) = σ1 ( )t ( )t for Lk ≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪ Uk − Lk
⎪
⎪ for TACk (t1k , t2k ) > Uk
⎩
0
⎧
⎪
⎪ 1
⎪
⎪ ( )t
⎨ U k − (TACk (t1k , t2k ))t for TACk (t1k , t2k ) < Lk
ThE−2 (TACk (t1k , t2k )) = σ2 ( )t ( )t for L k
≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪ Uk − Lk
⎪
⎪ for TACk (t1k , t2k ) > Uk
⎩
0
⎧
⎪
⎪ 1
⎪
⎪ ( )t
⎨ U k − (TACk (t1k , t2k ))t if TACk (t1k , t2k ) < Lk
ThE−n (TACk (t1k , t2k )) = σn ( )t ( )t i f L k
≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪ Uk − Lk
⎪
⎪ if TACk (t1k , t2k ) > Uk
⎩
0
1
k t t for TACk (t1k , t2k ) < Lk − sk
(U ) − (TAC k (t1k , t2k ))
IhE−2 (TACk (t1k , t2k ))= {τ2 t for Uk − sk ≤ TACk (t1k , t2k ) ≤ Uk
(Sk )
if TACk (t1k , t2k ) > Uk
0
………………………………………………………………………………….
⎧
⎪
⎪ 1
⎪
⎨ ( U k )t − (TACk (t1k , t2k ))t
⎪ for TACk (t1k , t2k ) < Lk − sk
IhE−n (TACk (t1k , t2k )) = τn ( )t for U k − sk ≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪ Sk
⎪
⎪ for TACk (t1k , t2k ) > Uk
⎩
0
9
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
……………………………………..
⎧
⎪
⎪ 0
⎪
⎨ ( TACk (t1k , t2k )t − Lk )t − (rk )t
⎪ for TACk (t1k , t2k ) < Lk + rk
FhE−n (TACk (t1k , t2k )) = δn ( )t ( )t for Lk + rk ≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪
⎪ Uk − Lk − (rk )t for TACk (t1k , t2k ) > Uk
⎪
⎩
1
Here sk , rk ∈ (0, 1)∀ k = 1, 2, 3, ……, n represents the tolerance value of indeterminacy and falsity respectively. h− Stands for the
minimization type hesitant objective function.
Where parameters σ n , τn , δn ∈ (0, 1) ∀n = 1, 2, ….., n representing the hesitant values for membership (truth), indeterminacy, non-
membership (falsity) function and is assigned by nth expert in Hesitant fuzzy environment.
ThE−1 (TACk (t1k , t2k )), IhE−1 (TACk (t1k , t2k )), FhE−1 (TACk (t1k , t2k )) are representing the truth (membership), indeterminacy and falsity(non-
membership) hesitant membership degree respectively assigned by 1st expert.
ThE−2 (TACk (t1k , t2k )), IhE−2 (TACk (t1k , t2k )), FhE−2 (TACk (t1k , t2k )) are representing the truth (membership), indeterminacy and falsity(non-
membership) hesitant membership degree respectively, assigned by 2st expert.
……………………………
ThE−n (TACk (t1k , t2k )), IhE−n (TACk (t1k , t2k )), FhE−n (TACk (t1k , t2k )) are representing the truth (membership), indeterminacy and falsity(non-
membership) hesitant membership degree respectively, assigned by nst expert.
Here we will formulate our multi-objective inventory model using the above membership functions.
∑n
1 μi
Max
n
∑n
1 θi
Max
n
∑n
1 ϑi
Max
n
Subject to,
ThE−i (TACk (t1k , t2k )) ≥ μi ; IhE−i (TACk (t1k , t2k )) ≥ θi ; FhE−i (TACk (t1k , t2k )) ≤ ϑi
ThE−1 (TACk (t1k , t2k )) ≥ μ1 , ThE−2 (TACk (t1k , t2k )) ≥ μ2 , …., ThE−n (TACk (t1k , t2k )) ≥ μn
IhE−1 (TACk (t1k , t2k )) ≥ θn , IhE−2 (TACk (t1k , t2k )) ≥ θ2 , ……., IhE−n (TACk (t1k , t2k )) ≥ θn
FhE−1 (TACk (t1k , t2k )) ≤ ϑ1 , FhE−2 (TACk (t1k , t2k )) ≤ ϑ2 , …., FhE−n (TACk (t1k , t2k )) ≤ ϑn ;
0 ≤ μ1 + μ2 + ……….. + μn ≤ 1;
0 ≤ θ1 + θ2 + ……….. + θn ≤ 1;
10
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
0 ≤ ϑ1 + ϑ2 + ……….. + ϑn ≤ 1
Pythagorean hesitant fuzzy programming technique is designed for solving multi-objective functions. Pythagorean hesitant fuzzy
programming technique try to maximize Pythagorean hesitant membership (Truth-membership) degree and minimize the Pythago
rean hesitant non-membership (falsity-membership) degree.
Using (22) the lower and upper bound, we are defining a linear membership (truth-membership) and non-membership (falsity-
membership) functions under Pythagorean-hesitant fuzzy
Pythagorean-hesitant membership (Truth membership) functions:
⎧
⎪
⎪ 1 if TACk (t1k , t2k ) < Lk
⎪
⎪
⎨
Uk − TACk (t1k , t2k )
ThE−1 (TACk (t1k , t2k )) = σ 1 if Lk ≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪ U k − Lk
⎪
⎪
⎩
0 if TACk (t1k , t2k ) > Uk
⎧
⎪
⎪ 1 if TACk (t1k , t2k ) < Lk
⎪
⎪
⎨
Uk − TACk (t1k , t2k )
ThE−2 (TACk (t1k , t2k )) = σ2 if Lk ≤ TACk (t1k , t2k ) ≤ Uk
⎪
⎪ U k − Lk
⎪
⎪
⎩
0 if TACk (t1k , t2k ) > Uk
Where parameters σ n , δn ∈ (0, 1) ∀n = 1, 2, ….., n representing set of hesitant values for membership (truth) and non-membership
(falsity) function and is assigned by nth expert under Pythagorean-hesitant fuzzy environment.
ThE−1 (TACk (t1k , t2k )), FhE−1 (TACk (t1k , t2k )) Stands for representing the Pythagorean-hesitant membership (truth-membership) and non-
membership (falsity-membership) degrees respectively assigned by 1st expert.
ThE−2 (TACk (t1k , t2k )), FhE−2 (TACk (t1k , t2k )) Stands for representing the Pythagorean-hesitant membership (truth-membership) and non-
membership (falsity-membership) degrees respectively assigned by 2nd expert.
ThE−n (TACk (t1k , t2k )), FhE−n (TACk (t1k , t2k )) Stands for representing the Pythagorean-hesitant membership (truth-membership) and non-
membership (falsity-membership) degrees respectively assigned by nth expert.
Using the linear membership function we have,
11
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
subject to ,
[ En ]2 [ ]2 [ ]2
Th− (TACk (t1k , t2k )) ≥ μ21 , ThE−n (TACk (t1k , t2k )) ≥ μ22 , …, ThE−n (TACk (t1k , t2k )) ≥ μ2n
[ En ]2 [ ]2 [ ]2
Fh− (TACk (t1k , t2k )) ≤ ϑ21 , FhE−n (TACk (t1k , t2k )) ≤ ϑ22 , …, FhE−n (TACk (t1k , t2k )) ≤ ϑ2n
μ2n ≥ ϑ2n ; 0 ≤ μ21 , μ22 , …, μ2n ≤ 1; 0 ≤ ϑ21 , ϑ22 , …, ϑ2n ≤ 1; 0 ≤ μ2n + ϑ2n ≤ 1, ∀n.
Hhere μ1 , μ2 , …, μn and ϑ1 , ϑ2 , …ϑn are representing the degree of satisfaction and dissatisfaction for each objective function under
the different expert’s opinions. The average difference between satisfaction and dissatisfaction degree is known as new achievement
function under Pythagorean hesitant fuzzy programming environment.
6. Numerical illustration
For numerical illustration, we are taking two objective function at a time. Here we have also considered the values of some cost
parameters as hexagonal fuzzy number and others parameters values as crisp.
̃1 (t11 , t21 ), TAC
Minimize {TAC ̃2 (t12 , t22 )}
t1i ≥ 0, t2i ≥ 0 .
Where,
{( ) ( )} [ { }
̃ i (t1i , t2i ) = i [Õi + h̃i ai 2 bi 3 ci 4 αi ai 4 bi 5 ci 6 1( 2 )
TAC t1i + t1i + t1i + t1i + t1i + t1i − S̃i ai t1i (t2i − t1i ) − t2i − t1i2
t2i 2 3 4 2 4 5 5 2
{ } { } ( )
bi 2 1( 3 3
) ci 3 1( 4 4
)
̃ ai 3 bi 4 ci 5
+ t1i (t2i − t1i ) − t2i − t1i + t1i (t2i − t1i ) − t2i − t1i + di t1i + t1i + t1i for i
2 3 3 4 3 4 5
= 1, 2.
Optimality condition:
̃ i (t1i , t2i ) with respect to each variable (t1i , t2i ) are ∂TACi and
Here we compute the first-order partial derivatives of the function TAC ∂t1i
∂TACi
∂t2i .
Now putting ∂TAC
∂t1i = 0 and
i ∂TACi
∂t2i = 0 we find the critical value of t1i , t2i .
2 2
(2 )2
∂TACi
Having these values we have > 0 and ∂ ∂TAC
∂t1i t2
i ∂ TACi
. ∂t2 − ∂∂t1iTAC
∂t2i
i
> 0.
1i 2i
Which shows us the minimum value for the total average cost TAC ̃ i ( t1i , t2i ).
In the following Table 1 we have considered the parameters whose values are hexagonal in nature.
The values of crisp parameters are given below.
After defuzzifying all the hexagonal fuzzy number by using the graded mean integration (18) and (19), we get the final optimal
solution in CRISP, NHFP AND PHFPA techniques with the help of Lingo Software (Table 3).
Graphical representation of the total average cost in above three methods have been given below.
From Fig. 2 we can conclude that the value of total average cost for two items in CRISP and NHFP technique are almost same. But
PHFPA technique gives little different value for total average cost. We can see the optimal values of time and total average cost from the
Table 3. It is also clear from the Table 3 that each methods (NHFP, PHFPA) give the optimal value which shows the stability of the
techniques we used to solve the proposed model.
Table 1
Input data (the cost parameters as hexagonal fuzzy number).
Parameters Item
Item1 Item2
12
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
Fig. 2. Graphical representation of the total average cost for two items.
7. Sensitivity analysis
In this section we are interested to see how sensitive our Inventory model is according to the changes of different parameters value.
Here we will discuss about the changes of total average cost for the parameters Ordering cost (Oi ), Holding cost (hi ) only.
7.1. Here we will see how the total average cost changes when we change the value of ordering cost (Oi ). For this we are taking all
the defuzzified cost parameters value and all the crisp parameters value from Tables 1 and 2.
Graphical representation of total average cost for two items given below.
From the above two figure (Figs. 3 and 4) we can conclude that whenever we are increasing the value of ordering cost corre
sponding total average cost is also increasing.
7.2. Here we will see how the total average cost changes when we change the value of holding cost (hi ). For this we are taking all the
defuzzified cost parameters value and all the crisp parameters value from Tables 1 and 2.
Graphical representation of total average cost for two items given below.
From the above two figure (Figs. 5 and 6) we can conclude that whenever we are increasing the value of holding cost, the cor
responding TACi (i=1,2) is also increasing (Tables 4,5).
8. Conclusion
In this article we have discussed a multi-objective inventory model for time dependent holding cost and deterioration. In the virtue
of customer demand, we have taken a quadratic demand function which is also time dependent. To handle uncertainty in cost pa
rameters, we have considered all the cost parameters as hexagonal fuzzy number. This research paper introduces an innovative in
ventory management model that addresses the complex challenges of multi-item inventory systems characterized by quadratic
demand patterns. Traditional inventory models often struggle to accommodate the uncertainties and vagueness inherent in real-world
demand and supply scenarios. To overcome these limitations, this study proposes a novel approach integrating Neutrosophic and
Pythagorean Hesitant Fuzzy Programming techniques into a deterministic inventory framework. After seeing the results of total
average cost, we can conclude that the performance of these two technique are almost the same. The acquired result (From Table 3)
specify the validity and stability of our inventory model.
Future Research Directions: The model can be implemented by using a ramp-type demand function, power demand can also be
taken into account here. For uncertainty, the values for the cost parameters can be considered as triangular fuzzy number, trapezoidal
fuzzy number, pentagonal fuzzy number and probabilistic function may also be taken here for more realization. To defuzzify the fuzzy
numbers researcher can use these techniques namely Centroid Method (Center of Area Method),Mean of Maxima (MOM), Bisector
Method: Smallest of Maxima (SOM), Largest of Maxima (LOM), Weighted Average of Maxima (WAM), Probability Weighted Moments
(PWM) etc. for the defuzzification instead of Graded mean integration method.
Some limitations proposed model:
An inventory model with quadratic demand, time-dependent holding costs and deterioration has certain limitations and com
plexities that can make it challenging to apply in real-world situations. Here are some of the limitations of this type of inventory model:
1. Complexity of the Model: Managing inventory with quadratic demand, time-dependent holding costs, and deterioration in
volves solving nonlinear equations and optimization problems, which can be computationally intensive and complex. This complexity
can make it challenging to implement and use in practice.
2. Data Requirements: To accurately model time-dependent holding costs and deterioration, you need historical data on these
Table 2
Cost parameter’s values in crisp.
b1 c1 α1 a2 b2 c2 α2
70 50 12 0.5 75 55 14 0.55
13
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
costs and deterioration rates, which may not always be readily available or easy to obtain.
3. Sensitivity to Parameter Values: The performance of the model can be highly sensitive to parameter values such as the co
efficients in the quadratic demand function, deterioration rates, and holding cost functions. Small changes in these parameters can lead
to significant changes in the optimal order quantity and total cost.
4. Limited Applicability: This type of inventory model may not be suitable for all types of products. It is primarily applicable to
products with quadratic demand patterns and deterioration over time. Many products have demand patterns that do not follow a
quadratic function, which limits the applicability of this model.
5. Assumption of Stationarity: The model assumes that the parameters governing demand, holding costs, and deterioration rates
remain stationary over time. In reality, these parameters may change due to various factors such as market conditions, seasonality, and
technological advancements.
6. Optimality Assumption: The model aims to find an optimal order quantity that minimizes total costs. However, achieving this
optimality may be challenging in practice due to factors like uncertainty in demand, lead times, and production constraints.
7. Limited Practical Guidance: While the model can provide optimal order quantities and cost estimates, it may not offer practical
guidance on inventory management strategies, such as reorder points, safety stock levels, or order timing, which are important for day-
to-day operations.
8. Limited Incorporation of Risk: The model typically does not explicitly account for risk factors, such as demand variability,
supply chain disruptions, or economic uncertainties, which can impact inventory decisions in practice.
9. Computational Resources: Solving the optimization problem associated with this type of model may require significant
14
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
Table 3
Optimal solution in CRISP, NHFP AND PHFPA techniques.
Methods t11 t21 TAC1(t11
∗
, t21
∗
) t12 t22 TAC2(t12
∗
, t22
∗
)
computational resources, especially for large-scale inventory systems, which can be a limitation in terms of time and computational
power.
Authors’ contribution
Both of the authors equally contributed equally to the above work. Both of them read and approved the final version of the
15
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
Table 4
Optimal solution in CRISP, NHFP AND PHFPA techniques.
Methods Ordering cost(O1 ) Ordering cost(O1 ) TAC1(t11
∗
, t21
∗
) t11
∗
t21
∗
TAC2(t12
∗
, t22
∗
) t12
∗
t22
∗
Table 5
The changes of total average cost depending on holding cost (hi ).
Methods Holding cost (1) Holding cost (h2 ) TAC1(t11
∗
, t21
∗
) t11
∗
t21
∗
TAC2(t12
∗
, t22
∗
) t12
∗
t22
∗
manuscript.
Not applicable.
Not applicable.
Funding
No funding to be reported.
The authors declare the following financial interests/personal relationships which may be considered as potential competing
interests:
Kausik Das reports financial support, administrative support, and writing assistance were provided by University of Kalyani. Kausik
Das reports a relationship with University of Kalyani that includes: non-financial support and speaking and lecture fees. Kausik Das has
patent pending to NA. No funding to be reported.
Data availability
Acknowledgment
Let us convey our humble gratitude to the department of mathematics, University of Kalyani for their cordial and helpful coop
eration by providing us with financial support through DSE-PURSE (Phase-II).
16
K. Das and S. Islam Results in Control and Optimization 14 (2024) 100367
References
[1] Zadeh, L.A. (1965). Fuzzy sets. Published online, 16 p. English. (OCR-слой). [Original Published: Information & Control. 8, 338-353],[Department of Electrical
Engineering and Electronics Research Laboratory, University of California, Berkeley, California].
[2] Chang SC, Yao JS, Lee HM. Economic reorder point for fuzzy backorder quantity. Eur J Oper Res 1998;109(1):183–202.
[3] Uthayakumar M. Fuzzy economic production quantity model for weibull deteriorating items with ramp type of demand. Int J Strateg Decis Sci (IJSDS) 2011;2
(3):55–90.
[4] Kazemi N, Ehsani E, Jaber MY. An inventory model with backorders with fuzzy parameters and decision variables. Int J Approx Reason 2010;51(8):964–72.
[5] Rajoria YK, Saini S, Singh S. EOQ model for decaying items with power demand, partial backlogging and inflation. Int J Appl Eng Res 2015;10(9):22861–74.
[6] Pal S, Mahapatra G, Samanta G. A production inventory model for deteriorating item with ramp type demand allowing inflation and shortages under fuzziness.
Econ Model 2015;46:334–45.
[7] Garai T, Chakraborty D, Roy TK. Fully fuzzy inventory model with price-dependent demand and time varying holding cost under fuzzy decision variables.
J Intell Fuzzy Syst 2019;36(4):3725–38.
[8] Padiyar S, Vikram S, Singh SR, Punetha N. Inventory system with price dependent consumption for deteriorating items with shortages under fuzzy environment.
Int J Sustain Agric Manag Inform 2021;7(3):218–31.
[9] Srivastava S, Bajaj RK. Optimal inventory management system for deteriorating items with linear demand, shortages and partial backlogging in a triangular
fuzzy setup. Int J Serv Oper Manag 2021;39(1):98–120.
[10] Rajput N, Pandey R, Chauhan A. Fuzzy optimisation of a production model with cntfn demand rate under trade-credit policy. Int J Math Oper Res 2022;21(2):
200–20.
[11] Biswas P, Pramanik S, Giri BC. Some distance measures of single valued neutrosophic hesitant fuzzy set sand their applications to multiple attribute decision
making. New trends in neutrosophic theory and applications. Brussels: Pons Editions; 2016. p. 27–34. pages.
[12] Yager RR, Abbasov AM. Pythagorean membership grades, complex numbers, and decision making. Int J Intell Syst 2013;28(5):436–52.
[13] Torra V, Narukawa Y. On hesitant fuzzy sets and decision. In: Proceedings of the IEEE international conference on fuzzy systems; 2009.
[14] Bharati SK. Hesitant fuzzy computational algorithm for multiobjective optimization problems. Int J Dyn Control 2018;1–8:6.
[15] Ye J. Multiple-attribute decision-making method under a single-valued neutrosophic hesitant fuzzy enviruniment. J Intell Syst 2014;24:23–36.
[16] Ahmad F, Adhami AY, Smarandache F. Single valued neutrosophic hesitant fuzzy computational algorithm for multi objective nonlinear optimization problem.
Neutrosophic Sets Syst 2018;22:76.
[17] Liang D, Xu Z. The new extension of TOPSIS method for multiple criteria decision making with hesitant Pythagorean fuzzy sets. Appl Soft Comput 2018;60:
167–79.
[18] Wei G, Lu M. Dual hesitant Pythagorean fuzzy amacher aggregation operators in multiple attribute decision making. Arch Control Sci 2017;27(3):365–95.
[19] Adhami AY, Ahmad F. Interactive Pythagorean hesitant fuzzy computational algorithm for multiobjective transportation problem under uncertainty. Int J
Manag Sci Eng Manag 2020;15(4):288–97.
[20] Rajarajeswari P, Sangeetha M. An effect for solving fuzzy transportation problem using hexagonal fuzzy numbers. Int J Res Inf Technol 2015;6:295–307.
[21] Chen SH, Hsieh CH. Graded mean integration representation of generalized fuzzy number. J Chin Fuzzy Syst 1999;5(2):1–7.
[22] Chen SH, Wang ST, Chang SM. Some properties of graded mean integration representation of LR type fuzzy numbers. Tamsui Oxf J Math Sci 2006;22(2):
185–208. Aletheia University.
[23] Deb M, De PK. Optimal solution of a fully fuzzy linear fractional programming problem by using graded mean integration representation method. Appl Appl
Math Int J (AAM) 2015;10(1):571–87.
[24] Mueen Z, Zaibidi NZ, Ramli R. Performance measure of multiple-channel queueing systems with imprecise data using graded mean intregration for trapezoidal
and hexagonal fuzzy number. J Comput Innov Anal 2022;1:1–13. Number 2.
[25] Alzabut J, Khan H, Nie Y, Alkhazzan A. A stochastic SIRS modeling of transport-related infection with three types of noises. AEJ Alex Eng J 2023;76:16. https://
doi.org/10.1016/j.aej.2023.06.049.
[26] Khan H, Alzabut J, Alfwzan WF, Gulzar H. Nonlinear dynamics of a piecewise modified ABC fractional-order leukemia model with symmetric numerical
simulations. Symmetry 2023;15(7):1338. https://doi.org/10.3390/sym15071338.
[27] Khan H, Alzabut J, Gulzar H. Existence of solutions for hybrid modified ABC-fractional differential equations with p-Laplacian operator and an application to a
aterborne disease model. Alex Eng J 2023;70:665–72. https://doi.org/10.1016/j.aej.2023.02.045.
[28] Hussain S, Tunç O, ur Rahman G, Khan H, Nadia E. Mathematical analysis of stochastic epidemic model of MERS-corona & application of ergodic theory. Math
Comput Simul 2023;207:130–50. https://doi.org/10.1016/j.matcom.2022.12.023.
[29] Khan H, Alzabut J, Gulzar H, Tunç O, Pinelas S. On system of variable order nonlinear p-Laplacian fractional differential equations with biological application.
Mathematics 2023;11(8):1913. https://doi.org/10.3390/math11081913.
[30] Telli B, Souid MS, Alzabut J, Khan H. Existence and uniqueness theorems for a variable-order fractional differential equation with delay. Axioms 2023;12(4):
339. https://doi.org/10.3390/axioms12040339.
[31] Khan H, Alzabut J, Tunç O, Kaabar MKA. A fractal–fractional COVID-19 model with a negative impact of quarantine on the diabetic patients. Results Control
Optim 2023;10:100199. https://doi.org/10.1016/j.rico.2023.100199. 2023.
[32] Abu Arqub O. Adaptation of reproducing kernel algorithm for solving fuzzy Fredholm–Volterra integrodifferential equations. Neural Comput Appl 2017;28:
1591–610. https://doi.org/10.1007/s00521-015-2110-x.
[33] Alshammari M, Al-Smadi M, Arqub OA, Hashim I, Alias MA. Residual series representation algorithm for solving fuzzy duffing oscillator equations. Symmetry
2020;12(4):572. https://doi.org/10.3390/sym12040572.
[34] Singh J, Maayah B, Alhodaly M. Reproducing kernel approach for numerical solutions of fuzzy fractional initial value problems under the Mittag–Leffler kernel
differential operator. Math Models Appl Sci 2021:7965–86. https://doi.org/10.1002/mma.7305.
[35] Singh J, Alhodaly M. Adaptation of kernel functions-based approach with Atangana–Baleanu–Caputo distributed order derivative for solutions of fuzzy
fractional Volterra and Fredholm integrodifferential equations. Math Models Appl Sci 2021:7807–34. https://doi.org/10.1002/mma.7228.
Further reading
[1] Bellmaaan RE, Zadeh LA. Decision making in fuzzy environment. Manag Sci 1970;17:141–64.
[2] Rajarajeshwari PA, Sudha S, Karthika R. A new operation on hexagonal fuzzy number. Int J Fuzzy Logic Syst 2013;3:15–26.
[3] Mueen Z, Ramli R, Zaibidi NZ. Parametric nonlinear programming approach with fuzzy queues using hexagonal membership functions. J Comput Theor Nanosci
2017;14(10):4979–85.
[4] Garg H. Hesitant pythagorean fuzzy sets and their aggregation operators in multiple attribute decision-making. Int J Uncertain Quantif 2018;8(3):267–89.
17