ArticlePDF Available

The pythagorean fuzzy Frank aggregation operators based on isomorphism Frank t-norm and s-norm and their application

Authors:
  • Hunan university of technology and business

Abstract

Based on the automorphism on the unit interval, an isomorphism Frank t-norm and its dual s-norm are created, which are applicable to Pythagorean fuzzy environment, and the generalized operational laws with respect to pythagorean fuzzy set(PFS) are defined and some properties of these laws are also investigated. Then, the Pythagorean fuzzy Frank weighted averaging(PFFWA) operator and Pythagorean fuzzy Frank weighted geometric(PFFWG) operator are proposed by using the new operational laws. Some desirable properties of these operators are proved. A new approach to solve the Pythagorean fuzzy multiple attribute decision making problem is developed by using the PFFWA and PFFWG operators. By solving the problem of airline service quality assessment, the proposed method is compared with existing methods. The analysis results demonstrate the feasibility and flexibility of the proposed method, and verify its ability to feedback the attitude of the decision maker.
33 8 Vol.33 No.8
20188Control and Decision Aug. 2018
文章编号:1001-0920(2018)08-1471-10 DOI: 10.13195/j.kzyjc.2017.0532
基于同构Frank t-模与s-模的勾股模糊
Frank集结算子及其应用
1,2,李延来1, 1,钱桂生1,2,吕红霞1
(1. 西南交通大学 交通运输与物流学院,成都 6100312. 香港城市大学 系统工程与工程管理系,香港 999077)
:运用单位区间上的自同构构造一种适用于勾股模糊环境下的同构 Frank t-模与其对偶 s-,进而定义勾
股模糊集的广义运算法则,并探究新法则的相关性质.应用新的运算法则提出勾股模糊 Frank 加权平均 (PFFWA)
算子与勾股模糊 Frank 加权几何 (PFFWG) 算子,证明算子的相关性质.利用 PFFWAPFFWG 算子提出一种解决
勾股模糊多属性决策问题的新方法.通过解决航空公司服务质量评估问题,对比分析新方法与现存的决策方法,
进而表明新方法的可行性和灵活性,并验证了新方法具有反馈决策者态度特征的能力.
关键词:勾股模糊数;Frank t-模与s-模;PFFWA算子;PFFWG算子
中图分类号:C934 文献标志码:A
The pythagorean fuzzy Frank aggregation operators based on
isomorphism Frank t-norm and s-norm and their application
YANG Yi1,2, LI Yan-lai1, DING Heng1, QIAN Gui-sheng1,2, LYU Hong-xia1
(1. School of Transportation and LogisticsSouthwest Jiaotong UniversityChengdu 610031China2. Department
of Systems Engineering and Engineering ManagementCity University of Hong KongHong Kong 999077China)
Abstract: Based on the automorphism on the unit interval, an isomorphism Frank t-norm and its dual s-norm are created,
which are applicable to Pythagorean fuzzy environment, and the generalized operational laws with respect to pythagorean
fuzzy set(PFS) are defined and some properties of these laws are also investigated. Then, the Pythagorean fuzzy Frank
weighted averaging(PFFWA) operator and Pythagorean fuzzy Frank weighted geometric(PFFWG) operator are proposed
by using the new operational laws. Some desirable properties of these operators are proved. A new approach to solve the
Pythagorean fuzzy multiple attribute decision making problem is developed by using the PFFWA and PFFWG operators.
By solving the problem of airline service quality assessment, the proposed method is compared with existing methods.
The analysis results demonstrate the feasibility and flexibility of the proposed method, and verify its ability to feedback
the attitude of the decision maker.
Keywords: pythagorean fuzzy numberFrank t-norm and s-normPFFWA operatorPFFWG operator
0
经典的模糊集(Fuzzy set, FS)[1] 用隶属度刻画
元素归属于集合的模糊程度.作为模糊集的拓展,
觉模 糊集 (Intuitionistic fuzzy set, IFS)[2] 同时考虑了
隶属度µ和非隶属度v两方面信息,且满足条件µ+
v1. IFS 能更加直观地描述决策者对事物的不确
定性,且在多属性决策 (MADM) 问题中得到了广泛
的应用[3-8] .文献 [9-10] 在分析 MADM 实际案例
考虑如下情形:属性评价值的隶属度与非隶属度之
和大 1, 隶属度与非隶属度的平方和小于1. ,
IFS的约束条件可知, IFS 并不适用于上述情形.
, Yager [9-10] (Pythagorean
fuzzy set, PFS), PFS 的隶属度与非隶属度的约束条
件为 µ2+v21,其适用于上述情形. Yager[10] 进一
步揭露了任意的IFS皆为PFS. 从几何的角度上看,
觉模糊集拓展到勾股模糊集的过程即为三角形约束
区域扩大到四分之圆约束区域.因此,在实际决策问
题中,勾股模糊集为评价者提供了更灵活的评价尺
度和工具,使其不再局限于提供隶属度和非隶属度之
和小于1的评价值,进而增强了处理多属性决策问题
收稿日期:2017-05-01修回日期:2017-08-23.
基金项目:国家自然科学基金项目(71373222, 71371156);西南交通大学拔尖人才培育项目(A0520502051601-6).
作者简介:杨艺 (1990), ,博士生,从事模糊决策的研究;李延来 (1971), ,教授,博士生导师,从事智能控制
与应用等研究.
通讯作者. E-mail: yanlaili@home.swjtu.edu.cn
1472 33
的能力.现存的文献研究表明了勾股模糊集备受关
,且能够有效地解决多属性决策问题[11-22]. Zhang
[11] 构建了拓展的勾股模糊 TOPSIS 决策方,并将
其应用于勾股模糊信息环境下的决策问题中. Peng
[12] 基于勾股模糊集的运算法则提出了勾股模糊集
结算子,并将其应用于股票投资的决策问题中.针对
专家权重未知的勾股模糊群决策问题, Zhang[13] 利用
所提出的相似测度获取专家权重,进而解决选择最优
光伏电池的实际问题.刘卫锋[15-16] 提出广义勾股
模糊加权集结算子与交叉勾股集结算子,将所提出集
结算子应用于国内航空公司服务质量评估的多属性
决策问题中.李德清等[17] 通过分析勾股模糊集的结
构特点提出勾股模糊集的距离测度,借助 TOPSIS
法建立了基于所提出距离测度的决策方法,并利用该
方法处理投资选择问题. Ma [18] 提出了对称勾股模
糊加权集结算子,并将其应用于航空公司服务质量评
估问题中. Ren [19] 提出了拓展的勾股模糊 TODIM
决策方法,并将其应用于亚洲基础设施投资银行选择
投资对象的实际决策问题中.彭新东等[20] 将语义集
拓展到勾股模糊语义集,并将其应用于软件开发项目
评价的决策问题中. Peng[21] 构建出拓展的勾股模
MABAC 方法,并将其应用于项目投资选择问题中.
模糊集的运算规则(Operationallaws)在模糊集理
论的发展过程中扮演着重要角色,对于勾股模糊集而
,其基本运算规则的研究相对甚少[11,18,22],尤其是
广义的运算规则.值得注意的是, Frank t-模和s-[23]
作为定义运算规则的重要工具,因其能在特定条件下
退化成Lukasiewicz t-模和 s-模以及Algebraic t-模和
s-,而被应用于定义各类模糊集的运算规则,如直
觉模糊集[24]、区间直觉模糊集[25]、犹豫模糊集[26]
区间直觉语义集[27]、三角区间二型模糊集[28].
Frank t-模和 s-模定义的运算规则较为广义,决策
可以根据参数的不同取值获取不同的运算规则,因此
能够更灵活地处理决策问题.然而,经过实例分析,
直接利用Frank t-模和 s-模定义勾股模糊集的运算规
,则会导致运算规则不满足封闭性.为此,经过剖析
直觉模糊集和勾股模糊集之间的关系,本文挖掘出单
位区间内的一个自同构,然后利用该自同构构造出
适用于勾股模糊集的一种同构Frank t-模与其对偶
s-,进而定义勾股模糊集的广义运算规则,并研究
新运算规则的特殊形式和基本性质.
集结算子在多属性决策过程中发挥着重要作
.决策者利用集结算子将多个评价信息融合成单
一的综合评价值后,再利用评价值的排序方法对综
进行 ,进而获取备选方案的排序,选择
符合要求的最优方案.针对勾股 模糊环境 MADM
的多个评价信息的集结问题,本文将基于同构Frank
t-模和 s-模定义的运算规则提出勾股模糊Frank
权平均 (PFFWA) 算子和勾股模糊 Frank 加权平均
(PFFWG) 算子,探讨其特殊形式和基本性质,并提出
一种基于PFFWA PFFWG 算子的勾股模糊环境下
的多属性决策方法.在部分勾股模糊环境下的多属
性决策问题中,不同决策者的心理特征并不统一,
乐观性和悲观性并非一成不变.面对此种情形,单一
固定的算子并不适用,而具有反馈决策者心理特征
能力的广义集结算子显得尤为重要.本文所提出带
参数的Frank算子是基于勾股模糊 Frank 运算定义的,
在特定条件下可以退化现存的部分算子,且具备广义
性和灵活性.利用所提出决策方法对航空公司的服
务质量进行评定和择优,以验证Frank 算子的这些性
,并深入分析算子中参数对于评估过程的影响,
用参数刻画决策者的心理特征.为了进一步表明该
方法的可信性和优点,将现存的决策方法与所提出方
法进行对比分析.
1󳹕备知󳆗
1.1 勾股模糊集的相关概念
给出直觉模糊集与勾股模糊集的相关定义.
定义 1[2] X={x1, x2,··· , xn}为给定的集
,X上的直觉模糊集I定义为
I={⟨xi, µI(xi), vI(xi)⟩|xiX},
其中 µI(xi)vI(xi)分别为元素 xi 于集 I的隶
属度和非隶属度.二元组 (µI(xi), vI(xi)) 称为直觉
(IFN), 便,直觉模糊数简记为β=
(µI, vI),其中µI, vI[0,1],µI+vI1.
定义 2[9-10] X={x1, x2,··· , xn}为给定的
集合,X上的勾股模糊集P定义为
P={⟨xi, µP(xi), vP(xi)⟩|xiX},
µP(xi)vP(xi)分别 为元 xi集合 P
隶属和非隶属. 元组 (µP(xi), vP(xi)) 称为
股模 糊数 (PFN), 方便, 股模 糊数 简记 α=
(µP, vP),其中µP, vP[0,1],µ2
P+v2
P1.
α= (µ, v),α1= (µ1, v1)α2= (µ2, v2)3
个勾股模糊数, Yager [9-10] 定义了如下基本运算:
1) α1α2= (max{µ1, µ2},min{v1, v2});
2) α1α2= (min{µ1, µ2},max{v1, v2});
3) αc= (v, µ).
进一,勾股模糊数之间的自然偏序关系定
义如:µ1µ2,v1v2,则称 α1α2.知若
α1α2,αc
2αc
1.
8 :基于同构Frank t-模与s-模的勾股模糊Frank集结算子及其应用 1473
定义3[12] αi= (µi, vi)(i= 1,2)为两个勾股
模糊 ,: 1) s(α1)< s(α2),α1α2; 2)
s(α1) = s(α2),h(α1)< h(α2),α1α2; 3)
s(α1) = s(α2),h(α1) = h(α2),α1α2.
s(αi) = µ2
iv2
ih(αi) = µ2
i+v2
i分别为αi(i= 1,2)
的记分函数[11] 和精确度函数[12].
由定义3易知,α1α2,α1α2α1=
α2.
1.2 Frank s-模与t-
定义 4[29] φ: [a, b][a, b],φ是严格递
增的连续函数, 满足 φ(a) = a,φ(b) = b,则称 φ
[a, b]上的自同构.
定义5[29] N: [0,1] [0,1],N单调减,
N(0) = 1,N(1) = 0,则称 N为否定函数.若否定函
N严格递减且连续,则称 N为严格否定函数.若严
格否定函数N满足复原性质x[0,1],N(N(x)) =
x,则称N为强否定函数.
若强否定函数N(x) = 1 x,则称为标准否定函
,简记为 NI.若强否定函数为 N(x) = 1x2,
称为勾股否定函数[9-10],简记为 NP.
1[30] N为强否定函数当且仅当存在自
φ: [0,1] [0,1],使得 x[0,1],N(x) =
φ1(NI(φ(x))),其中NI为标准否定函数.
定理 1表明,任意的强否定函数都可以通过单位
区间上的自同构构建.
定义 6[29] T: [0,1] ×[0,1] [0,1] :
1) 对称性:x, y [0,1], T (x, y) = T(y, x); 2) 单调
:x1x2, y1y2, T (x1, y1)T(x2, y2); 3)
合律:x, y, z [0,1], T (x, T (y, z)) = T(T(x, y), z );
4) 边界条件:x[0,1], T (x, 1) = x.则称 Tt-.
定义 7[29] S: [0,1] ×[0,1] [0,1] :
1) 对称性:x, y [0,1],S(x, y) = S(y, x); 2)
:x1x2, y1y2,S(x1, y1)S(x2, y2); 3)
合律:x, y, z [0,1],S(x, S (y, z)) = S(S(x, y), z );
4) 边界条件:x[0,1], S(x, 0) = 0.则称 Ss-
(t-余模).
George [31] 提出了对偶三元组的概念 (T, S,
N),TS关于 N对偶,将这样的三元组称为对偶
三元组.
8[31] TS分别 t-模和 s-,TS
关于否定函数N对偶当且仅当
T(x, y) = N(S(N(x), N (y))),
S(x, y) = N(T(N(x), N (y))).
Frank[23] 提出了Frank t-模和s-,定义为
TF(x, y) = h1
F(hF(x) + hF(y)) =
logτ1 + (τx1)(τy1)
τ1,(1)
SF(x, y) = g1
F(gF(x) + gF(y)) =
1logτ1 + (τ1x1)(τ1y1)
τ1.
(2)
其中:τ(1,),标准否定函数NI(x) = 1x,hF(x)
=log((τ1)/(τx1))为连续严格递减函数,gF(x)
=hF(NI(x)) 为连续严格递增函数,hFgF分别为
Frank t-模和 s-模的生成元.显然, Frank t-模和 Frank
s-SF关于 NI.根据定义 9, (TF, SF, NI)
Frank对偶三元组.
特别地,τ1, Frank t-模退化为Algebraic
t-T(x, y) = xy, Frank s-模退 化为 Algebraic s-
S(x, y) = x+yxy.
2勾股模糊集的广义运算法则
虽然Frank t-模和 s-模已成功应用于定义直觉模
糊集的运算规则,但经实例分析其并不适用于勾股模
.为此,本文基于标准否定函数与勾股否定函数分
析了直觉模糊集与勾股模糊集之间的关系,确定了单
位区间内的一个自同构,进而利用该自同构构造出
适用于勾股模糊集的一种同构Frank t-模与其对偶
s-,并定义了勾股模糊集的广义运算规则.
2.1 一种同构Frank t-模与其对偶s-
1α1= (0.6,0.8)α2= (0.8,0.6)
个勾股模糊数.若利用Frank t- s-模定义勾股模
糊数的加法运算,则有
α1Fα2= (SF(µ1, µ2), TF(v1, v2)).
,τ= 2 ,α1Fα2= (0.933,0.467).
,(0.933)2+ (0.467)2>1,因此 α1Fα2是勾
模糊数,即该运算规则不满足封闭性.
1表明,基于Frank t-模与s-模定义的勾股模糊
集的运算规则不满足封闭性,因此 Frank t-模与 s-
不适用于勾股模糊集.
下面引入单位区间上的一个自同构构造出适
于勾股模糊集的一种同构Frank t-模与其对偶s-,
在此之前,回顾同构t-模与s-模的基本定义.
定义 9[29] T是一 t-,φ[0,1]上的一
自同构,若函数Tφ为映射Tφ: [0,1]2[0,1],且满足
Tφ(x, y) = φ1(T(φ(x), φ(y))),(3)
Tφ仍为t-,并称TφT的同构t-.
10 [29] S是一个s-,φ[0,1] 上的一
个自同构,若函数Sφ为映射Sφ: [0,1]2[0,1],
满足
1474 33
Sφ(x, y) = φ1(S(φ(x), φ(y))),(4)
Tφ仍为s-,并称SφS的同构t-.
NI与勾股否定函数
NP对直觉模糊数和勾股模糊数之间的关系进行探
.I= (µI, vI)为直觉模糊数,其约束条件µI+
vI1可以等价为µINI(νI).对于勾股模糊数
P= (µP, vP),其约束条件 (µP)2+ (vP)21可以等
价为 µPNP(νP).由定 1可得,勾股否定函数可
通过标准否定函数和[0,1]上的自同构φ(x) = x2
,NP(x) = φ1(NI(φ(x))).
1下文中符号φ均视作为自同构φ(x) = x2.
基于自 同构 φ(x) = x2提出一个新的对偶三元
(TF,φ, SF , NP),TF,φ 为同构 Frank t-,SF 为同
Frank s-,下面给出相关定义.
定义11 TF,φ : [0,1] ×[0,1] [0,1],且满足
TF,φ(x, y ) = φ1(TF(φ(x), φ(y))),(5)
则称TF,φ Frank t-TF的同构t-,简称同构Frank
t-,其中φ(x) = x2[0,1]上的自同构.
根据定义11和式(1),可得
TF,φ(x, y ) = h1
F,φ(hF (x) + hF,φ (y)) =
logτ1 + (τx21)(τy21)
τ1,(6)
其中hF,φ(t) = hF(φ(t)) TF 的生成元.
定义12 SF,φ : [0,1] ×[0,1] [0,1],且满足
SF,φ(x, y ) = φ1(SF(φ(x), φ(y))),(7)
SF,φ Frank s-SF s-,
Frank s-,其中φ(x) = x2[0,1]上的自同构.
根据定义12和式(2),可得
SF,φ(x, y ) = g1
F,φ(gF (x) + gF,φ (y)) =
1logτ1 + (τ1x21)(τ1y21)
τ1,(8)
其中gF,φ(t) = gF(φ(t)) SF 的生成元.
2TF SF,φ 11
12 所给的同构Frank t-模和同构 Frank s-, :
1) TF,φ(x, y ) = NP(SF,φ(NP(x), NP(y))); 2) SF,φ (x,
y) = NP(TF,φ(NP(x), NP(y))). NP(t) =
1t2为勾股否定函数.
该定理只需证明1) 即可,类似可以证明
2). 根据定义12,
SF,φ(x, y ) = φ1(SF(φ(x), φ(y))),
NP(x) = φ1(NI(φ(x))).
进而有
SF,φ(NP(x), NP(y)) =
φ1(SF(φ(NP(x)), φ(NP(y)))) =
φ1(SF(φ(φ1(NI(φ(x)))), φ(φ1(NI(φ(y)))))) =
φ1(SF(NI(φ(x)), NI(φ(y)))).
因此
NP(SF,φ(NP(x), NP(y))) =
φ1(NI(φ(φ1(SF(NI(φ(x)), NI(φ(y))))))) =
φ1(NI(SF(NI(φ(x)), NI(φ(y))))) =
φ1(TF(φ(x), φ(y))) = TF,φ(x, y ).2
根据定理2和定义8可得,TF,φ SF 关于勾股
否定函数NP对偶,因此(TH,φ, SH,φ , NP)仍为对偶三
元组,将其称为同构Frank对偶三元组.
Frank 对偶三元组(TF, SF, NI)与同构 Frank
偶三元组(TF,φ, SF , NP)之间的关系体现为:
1) TF,φ(x, y ) = φ1(TF(φ(x), φ(y)));
2) SF,φ(x, y ) = φ1(SF(φ(x), φ(y)));
3) NP(x) = φ1(NI(φ(x))).
上述三元组的特点在于TF,φSF,φ NP皆基于
自同构φ(x) = x2构造.
2.2 勾股模糊集的广义运算法则
定义 13 αi= (µi, vi)(i= 1,2) 为两个勾股
模糊数,τ(1,),λ > 0.基于同构Frank 对偶三
元组(TH,φ, SH,φ , NP)的勾股模糊集的运算规则定义
:
1) α1F I α2= (SF,φ(µ1, µ2), TF,φ(v1, v2)) =
1logτ1 + (τ1µ2
11)(τ1µ2
21)
τ1,
logτ1 + (τv2
11)(τv2
21)
τ1,
2) α1F I α2= (TF,φ(µ1, µ2), SF,φ(v1, v2)) =
logτ1 + (τµ2
11)(τµ2
21)
τ1,
1logτ1 + (τ1v2
11)(τ1v2
21)
τ1,
3) λα1= (g1
F,φ(λgF (µ1)), h1
F,φ(λhF (v1))) =
(1logτ(1 + (τ1µ2
11)λ/(τ1)λ1),
logτ(1 + (τv2
11)λ/(τ1)λ1) ),
4) αλ
1= (h1
F,φ(λhF (µ1)), g1
F,φ(λgF (v1))) =
(logτ(1 + (τµ2
11)λ/(τ1)λ1),
1logτ(1 + (τ1v2
11)λ/(τ1)λ1) ).
8 :基于同构Frank t-模与s-模的勾股模糊Frank集结算子及其应用 1475
特别地,τ1,则定义13中的运算规则退化为:
1) α1F I α2= (1(1 µ2
1)(1 µ2
2), v1v2);
2) α1F I α2= (µ1µ2,1(1 v2
1)(1 v2
2) );
3) λα1= (1(1 µ2
1)λ, vλ
1);
4) αλ
1= (µλ
1,1(1 v2
1)λ).
易知,上述运算法则为Zhang[11] 定义的勾股模
糊集的运算法则.
定理 3(封闭)αi= (µi, vi)(i= 1,2)为两
个勾股模糊数,13 中的运算法则均满足封闭性,
α1F I α2α1F I α2λα1αλ
1仍为勾股模糊数.
证明 下面证明 α1F I α2λα1为勾股模糊数,
其余的可以类似证明.由于αi= (µi, vi)(i= 1,2)
勾股模糊数,µiNP(vi)(i= 1,2).根据定义7
定义8,
SF,φ(µ1, µ2)
SF,φ(NP(v1), NP(v2)) =
NP(TF,φ(NP(NP(v1)), NP(NP(v2)))) =
NP(TF,φ(v1, v2)),
因此α1F I α2为勾股模糊数.
根据生成元的单调性,
g1
F,φ(λgF (µ1))
g1
F,φ(λgF (NP(µ1))) =
NP(h1
F,φ(λhF (NP(NP(µ1))))) =
NP(h1
F,φ(λhF (µ1))).
因此λα1为勾股模糊数.2
定理 4αi= (µi, vi)(i= 1,2) 为两个勾股模
糊数,则有:
1) α1F I α2=α2F I α1;
2) α1F I α2=α2F I α1;
3) λα1F I λα2=λ(α1F I α2),λ > 0;
4) αλ
1F I αλ
2= (α1F I α2)λ,λ > 0;
5) λ1α1F I λ2α1= (λ1+λ2)α1,λ1, λ2>0;
6) αλ1
1F I αλ2
1= (α1)λ1+λ2,λ1, λ2>0.
根据定义13易证定理14,此处不再详述.
3勾股模糊 Frank加权平均算子与勾股模
Frank加权几何算子
为了集结带权重的勾股模糊数组,本节基于所
定义的勾股模糊Frank 运算规则提出勾股模糊Frank
加权 平均 (PFFWA) 算子和勾股模糊 Frank 权几
(PFFWA) 算子.
3.1 勾股模糊Frank加权平均算子
定义 14 αi= (µi, vi)(i= 1,2,··· , n)
股模糊数组,勾股模糊 Frank 加权平(PFFWA)
为映射ΘnΘ,满足
PFFWA(α1, α2,·· · , αn) = n
F I
i=1
(wiαi).(9)
其中:Θ为所有勾股模糊数组成的集合;(w1, w2,··· ,
wn)T(α1, α2,·· · , αn)的权重向量,满足 wi>0,
n
i=1
wi= 1.
定理 5αi= (µi, vi)(i= 1,2,··· , n)为勾股
模糊数组,则有
PFFWA(α1, α2,·· · , αn) =
g1
F,φn
i=1
wigF,φ(µi), h1
F,φn
i=1
wihF,φ(vi)=
1logτ1 +
n
i=1
(τ1µ2
i1)wi,
logτ1 +
n
i=1
(τv2
i1)wi.
基于 定义 13 中的运算法,利用数学归纳法可
以证明定理 5. 下面探究算子的基本性质:闭性、
调性、幂等性、有界性.
6αi= (µ1i, v1i)(i= 1,2,··· , n)
βi= (µ2i, v2i)(i= 1,2,·· · , n)为两组勾股模糊数
,则有:
1) 封闭: PFFWA(α1, α2,· ·· , αn)勾股模糊
;
2) 单调性:αiβi(i= 1,2,·· · , n),
PFFWA(α1, α2,·· · , αn)
PFFWA(β1, β2,·· · , βn); (10)
3) 幂等性:αi=α= (µ, v)(i= 1,2,··· , n),
PFFWA(α1, α2,·· · , αn) = α;(11)
4) 有界性:若令
αmin = (min
i{µ1i},max
i{v1i}),
αmax = (max
i{µ1i},min
i{v1i}),
αmin PFFWA(α1, α2,·· · , αn)αmax .(12)
下面 探究勾 股模糊 Frank 权平均算子的特殊
情形.
定理 7αi= (µi, vi)(i= 1,2,··· , n)为勾股
模糊数组,:
1) lim
τ1PFFWA(α1, α2,·· · , αn) =
1
n
i=1
(1 µ2
i)wi,
n
i=1
(vwi
i);
1476 33
2) lim
τ→∞ PFFWA(α1, α2,·· · , αn) =
n
i=1
wiµ2
i,
n
i=1
wiv2
i.
证明 1) 根据常用的等价无穷小,x0,
xln(1 + x),ax1xln a(a>0),因此,τ1
,
ln 1 +
n
i=1
(τ1µ2
i1)wi
n
i=1
(τ1µ2
i1)wi,
τ1µ2
i1 = e(1µ2
i)ln τ1(1 µ2
i)ln τ.
进而有
lim
τ1
1logτ1 +
n
i=1
(τ1µ2
i1)wi=
1lim
τ1ln 1 +
n
i=1
(τ1µ2
i1)wiln τ=
1lim
τ1n
i=1
(τ1µ2
i1)wiln τ=
1lim
τ1n
i=1
((1 µ2
i)ln τ)wiln τ=
1
n
i=1
(1 µ2
i)wi.
同理可证
lim
τ1
logτ1 +
n
i=1
(τv2
i1)wi=
n
i=1
(vwi
i).
因此
lim
τ1PFFWA(α1, α2,·· · , αn) =
1
n
i=1
(1 µ2
i)wi,
n
i=1
(vwi
i).
2) 根据洛必达法则,
lim
τ→∞
1logτ1 +
n
i=1
(τ1µ2
i1)wi=
1lim
τ→∞
n
i=1
wi(1µ2
i)τµ2
i
τ1µ2
i1n
i=1
(τ1µ2
i1)wi
(1/τ)1 +
n
i=1
(τ1µ2
i1)wi
=
1lim
τ→∞
n
i=1
(τ1µ2
i1)win
i=1
wi(1µ2
i)τ1µ2
i
τ1µ2
i1
1 +
n
i=1
(τ1µ2
i1)wi
=
1
n
i=1
wi(1 µ2
i) =
n
i=1
wiµ2
i.
同理可证
lim
τ→∞
logτ1 +
n
i=1
(τv2
i1)wi=
n
i=1
wiv2
i.
因此
lim
τ→∞ PFFWA(α1, α2,·· · , αn) =
n
i=1
wiµ2
i,
n
i=1
wiv2
i.2
上述定理表明.当参数 τ1, PFFWA 算子退
化为现存的勾股模糊加权平均(PFWA) [9-10];
参数 τ , PFFWA 算子退化为现存的勾股模
加权幂平均(PFWPM)算子[9-10].
3.2 勾股模糊Frank加权几何算子
定义 15 αi= (µi, vi)(i= 1,2,··· , n)
股模糊数组,勾股模糊Frank 加权几何(PFFWG)
为映射ΘnΘ,满足
PFFWG(α1, α2,·· · , αn) = n
F I
i=1
(αwi
i).(13)
其中:Θ为所有勾股模糊数组成的集合;(w1, w2,··· ,
wn)T(α1, α2,·· · , αn)的权重向量,满足 wi>0,
n
i=1
wi= 1.
定理 8αi= (µi, vi)(i= 1,2,··· , n)为勾股
模糊数组,
PFFWG(α1, α2,·· · , αn) =
h1
F,φn
i=1
wihF,φ(µi), g1
F,φn
i=1
wigF,φ(vi)=
logτ1 +
n
i=1
(τµ2
i1)wi,
1logτ1 +
n
i=1
(τ1v2
i1)wi.
9αi= (µ1i, v1i)(i= 1,2,··· , n)
βi= (µ2i, v2i)(i= 1,2,·· · , n)为两组勾股模糊数
,则有:
1) 封闭: PFFWG(α1, α2,··· , αn)为勾股模
;
2) 单调性:αiβi(i= 1,2,·· · , n),
PFFWA(α1, α2,·· · , αn)
8 :基于同构Frank t-模与s-模的勾股模糊Frank集结算子及其应用 1477
PFFWG(β1, β2,·· · , βn); (14)
3) 幂等性:αi=α= (µ, v)(i= 1,2,··· , n),
PFFWG(α1, α2,·· · , αn) = α;(15)
4) 有界性:若令
αmin = (min
i{µ1i},max
i{v1i}),
αmax = (max
i{µ1i},min
i{v1i}),
αmin PFFWG(α1, α2,·· · , αn)αmax .(16)
类似于PFFWA 算子,可以获取 PFFWG 算子的如
下特殊情形.
定理 10 αi= (µi, vi)(i= 1,2,··· , n)
股模糊数组,:
1) lim
τ1PFFWG(α1, α2,·· · , αn) =
n
i=1
(vwi
i),
1
n
i=1
(1 µ2
i)wi;
2) lim
τ→∞ PFFWG(α1, α2,·· · , αn) =
n
i=1
wiµ2
i,
n
i=1
wiv2
i.
定理10的证明类似定理7,此处不再详述.
上述定理表:当参数 τ1, PFFWG 算子退
化为现存的勾股模糊加权几何(PFWG) 算子[9-10];
参数 τ , PFFWG算子退化为现存的勾股模糊
加权幂平均(PFWPM)算子[9-10].
下述定理阐述了PFFWA 子与 PFFWG 算子之
间的关系.
11 αi= (µi, vi)(i= 1,2,··· , n)
股模 ,: 1) PFFWG(α1, α2,··· , αn) =
(PFFWA((α1)c,(α2)c,·· · ,(αn)c))c; 2) PFFWA(α1,
α2,·· · , αn) = (PFFWG((α1)c,(α2)c,· · · ,(αn)c))c.
其中αc
i= (vi, µi), i = 1,2,· · · , n.
4基于勾股模糊 Frank算子的多属性决策
方法
本节 将提出 的勾股模糊 Frank 子应用于勾股
模糊环境下的多属性决策问题中.在多属性决策问
题中,设备选方案集为 X={X1, X2,··· , Xm},属性
(准则)集为 C={C1, C2,··· , Cn}.决策者根据属性
Cj(j= 1,2,·· · , n)对方案 Xi(i= 1,2,· · · , m)进行
评价,进而给出决策矩阵 D= (αij )m×n,其中 αij =
(µij , vij )为勾股模糊.属性集的权重向量为W=
(w1, w2,·· · , wn),
n
j=1
wj= 1,wj>0, j =
1,2,·· · , n.
Step 1:根据属性权重向量W=(w1, w2,·· · , wn),
利用 勾股模 Frank 加权平 (PFFWA) 算子或勾股
模糊Frank加权几何(PFFWG)算子
αi=n
F I
j=1
(wjαij ), αi=n
F I
j=1
(αwj
ij )(17)
集结备 选方案 Xi(i= 1,2,··· , m)所对应的评价值
αij (j= 1,2,·· · , n),进而获取综合评价值 αi(i=
1,2,·· · , m).
Step 2: 利用定义3中的排序方法对综合评价值
αi(i= 1,2,·· · , m)进行排序.
Step 3:根据综合评价值αi(i= 1,2,·· · , m)的排
序对方案集X={X1, X2,··· , Xm}行排序,
最优方案.
5实例分析
考虑航空公司的服务质量评价问题[11],评价对
象为 4家国内航空公司 X={X1, X2, X3, X4},评价
属性 (准则)包括:机票服务 C1 检和登机服 C2
客舱服务 C3响应性C4. 4 个属性的详情见文献[11].
评价委员会给出属性的权重向量为W= (0.15,0.25,
0.35,025),评价对象Xi(i= 1,2,3,4) 在属性Cj(j
= 1,2,3,4)下的评价值 αij = (µij , vij )(i, j = 1,2,
3,4) 为勾股模糊数,进而构成评价矩阵 D=
(αij )4×4,详见表 1. 利用所提出的多属性决策方法选
出服务质量最优的航空公司.
1决策矩阵D
C1C2C3C4
X1(0.9,0.3) (0.7,0.6) (0.5,0.8) (0.6,0.3)
X2(0.4,0.7) (0.9,0.2) (0.8,0.1) (0.5,0.3)
X3(0.8,0.4) (0.7,0.5) (0.6,0.2) (0.7,0.4)
X4(0.7,0.2) (0.8,0.2) (0.8,0.4) (0.6,0.6)
Step 1:根据权重向量W=(0.15,0.25,0.35,025)
和评价矩阵 D= (αij )4×4,利用 PFFWA算子 (τ= 2)
集结αij (j= 1,2,3,4),
αi=
1logτ1 +
4
j=1
(τ1µ2
ij 1)wj,
logτ1 +
4
j=1
(τv2
ij 1)wj,
进而获取综合评价值
α1= (0.677 8,0.511 2),
α2= (0.756 2,0.212 0),
α3= (0.688 8,0.333 6),
α4= (0.748 3,0.338 4).
1478 33
Step 2: 定义 3
αi(i= 1,2,3,4)进行排序.记分函数值
s(α1) = 0.198 0,
s(α2) = 0.526 8,
s(α3) = 0.363 1,
s(α4) = 0.445 4,
因此α2α4α3α1.
Step 3:根据综合评价值αi(i= 1,2,·· · , m)的排
序获取方案集的排序,
X2X4X3X1,
因此最优方案为X2.
为了探究参数对决策过程的影响,考虑参数范围
τ(1,50]方案的排序情况综合分析.
利用 PFFWA 算子集 结方案 Xi(i= 1,2,3,4)
应的评价值αij (j= 1,2,3,4)获取综合评价值 αA
i=
PFFWA(αi1, αi2, αi3, αi4)(i= 1,2,3,4),其对应的
记分函数值s(αA
i)(i= 1,2,3,4) 与参数τ(1,50]
之间的关系如图1所示.
0 10 20 30 40 5 0
τ
0.1
0.2
0.3
0.4
0.5
0.6
s X( )
1
s X( )
2
s X( )
3
s X( )
4
!"#$%
1基于PFFWA 算子下的记分函数值
利用 PFFWG 算子集结方案Xi(i= 1,2,3,4)
应的评价值 αij (j= 1,2,3,4)获取综合评价值 αG
i=
PFFWG(αi1, αi2, αi3, αi4)(i= 1,2,3,4),其对应的
记分函数值s(αG
i)(i= 1,2,3,4) 与参 τ(1,50]
之间的关系如图2所示.
0 10 20 30 40 5 0
τ
s X( )
1
s X( )
2
s X( )
3
s X( )
4
0
0.1
0.2
0.3
0.4
!"#$%
2基于PFFWG算子下的记分函数值
将综 合评价值 αA
i(i= 1,2,3,4) 对应的记分函
数值 s(αA
i)(i= 1,2,3,4) αG
i(i= 1,2,3,4) 对应
的记分函数值s(αG
i)(i= 1,2,3,4)作差值,进而获取
si=s(αA
i)s(αG
i)(i= 1,2,3,4),并给出 si(i=
1,2,3,4)与参数之间的关系如图3所示.
0 10 2 0 30 40 5 0
τ
X1
X2
X3
X4
0. 02
0. 06
0. 10
0. 14
0. 18
0. 22
!"#$%&'
3PFFWA PFFWG 算子下的记分函数值之差
分别利用PFFWA 算子、PFFWG算子、PFWPA
子、PFWA 算子和 PFWG 算子对方案X1对应的评价
α1j(j= 1,2,3,4)进行集结,获取方案X1对应综合
评价值αA
1αG
1αPFWPA
1αPFWA
1αPFWG
1,研究综
合评价值的记分函数值s(αA
1)s(αG
1)s(αPFWPA
1)
s(αPFWA
1)s(αPFWG
1)与参数之间的关系如图4所示.
0 10 20 30 40 5 0
τ
s(PF FWA)
s(PF FWG)
s(PF WPA)
s(PF WA)
s(PF FWG)
0
0.1
0.2
0.3
!"#$%
4基于不同算子下方案X1的记分函数值
由图14可见:
1) 根据图 1, τ(1,50] ,空公司的排序为
X2X4X3X1,方案 Xi(i= 1,2,3,4) 对应
记分函数值s(αA
i)(i= 1,2,3,4)关于参数 τ呈递减趋
.
2) 根据图 2, τ(1,50] ,空公司的排序为
X4X2X3X1,服务质量最优的公司为X4,
方案 Xi(i= 1,2,3,4) 对应的记分函数值 s(αG
i)(i=
1,2,3,4)关于参数τ呈递增趋势.
3) 根据 3, PFFWA 算子获取的记分函数值与
PFFWG算子获取的记分函数值之间的差值 si(i=
1,2,3,4) 关于 参数 τ(1,50] 递减,由图 1 2
,s(αA
i)(i= 1,2,3,4) 关于参数递减,s(αG
i)(i=
1,2,3,4)关于参数递增.
4) τ(1,50] ,si>0(i= 1,2,3,4),
明由 PFFWA 算子获取的综合评价值大于PFFWG
子获取的综合评价值αA
iαG
i(i= 1,2,3,4).
PFFWG 算子而言, PFFWA 算子获取的综合评价
值较大,适用于乐观的决策者,参数可以刻画乐观水
, PFFWG算子较适用于悲观的决策者,参数可以刻
画决策者的悲观水平.
8 :基于同构Frank t-模与s-模的勾股模糊Frank集结算子及其应用 1479
5) 根据图4,τ(1,50],
s(αPFWG
1)< s(αG
1)< s(αPFWPA
1)<
s(αA
1)< s(αPFWA
1),
与定理7和定理10保持一致.根据定理7,
lim
τ1s(αA
1) = s(αPFWA
1),
lim
τ→∞ s(αA
1) = s(αPFWPA
1).
根据定理10,
lim
τ1s(αG
1) = s(αPFWG
1),
lim
τ→∞ s(αG
1) = s(αPFWPA
1).
为了进一步验证本文所提出决策方法的可行
和可信性,将所提出方法与现存的决策方法进行对比
分析.利用现存的勾股集结算子和决策方法解决上
述算例中多属性决策问题,获取的排序结果如表2
.由表2可见,基于表中8类算子所获取的方案排序
X2X4X3X1,最优方案为 X2.利用 Zhang
[11] TOPSIS 方法获取的方案排序为X2X3
X4X1,虽然方 X4X3的排序略有变化,但最
优方案仍为X2,充分表明了本文所提出的PFFWA
子和PFFWG算子具有一定的可行性.事实上,根据上
述分析所述, Yager [9-10] 提出的PFWPA算子、PFWA
算子和PFWG算子为本文所提出算子的特殊形式,
明本文所提出算子具有退化性.相对表中的其他算
子而言,本文算子具有一定的灵活性, PFFWA子中
的参数可以表征乐观决策者的乐观程度, PFFWG
子中的参数可以表征悲观决策者的悲观程度.
2决策方法与方案排序结果
算子与决策方法 方案排序
本文的 PFFWA PFFWG算子 X2X4X3X1
Yager[9-10] PFWPA 算子 X2X4X3X1
Yager[9-10] PFWA PFWG算子 X2X4X3X1
Zhang Xu[11] TOPSIS 方法 X2X3X4X1
刘卫锋等[16] PFIWA 算子 X2X4X3X1
Ma Xu[18] SPFWG/SPFWA 算子 X2X4X3X1
6
鉴于Frank t-模与 s-模可有效应用于定义各类模
糊集的广义运算规则,本文构造出适用于定义勾股模
糊集广义运算规则的一种同构Frank t-模与其对偶
s-.勾股模糊集的广义运算规则具有兼容性,能够
退化成文献[11] 定义的运算规则.进一步,针对带
权重向量的勾股模糊数组,基于新的运算规则提出两
类勾股模糊 Frank 集结算子,研究其基本性质,揭露
了所提出算子在特定条件下可以退化为文献[9-10]
所提出的勾股模糊集结算子.利用基于新算子提出
的决策方法对航空公司的服务质量进行评估,深入分
析了算子参数对于决策过程的影响,并剖析了新提出
算子在该案例中的内在关联.
参考文献(References)
[1] Zadeh L A. Fuzzy sets[J]. Information & Control, 1965,
8(3): 338-353.
[2] Atanassov K T. Intuitionistic fuzzy sets[J]. Fuzzy Sets &
Systems, 1986, 20(1): 87-96.
[3] 万树平.直觉模糊多属性决策方法综述 [J]. 控制与
, 2010, 25(11): 1601-1606.
(Wan S P. Survey on intuitionistic fuzzy multi-criteria
decision making approach[J]. Control and Decision,
2010, 25(11): 1601-1606.)
[4] 曾祥添,李登峰,余高锋.面向偏好冲突的直觉模糊
多属性群体决策方法[J]. 控制与 决策, 2015, 30(11):
2108-2112.
(Zeng X T, Li D F, Xu G F. Intuitionistic fuzzy
multi-attribute group decision-making method for
preference conflicting[J]. Control and Decision, 2015,
30(11): 2108-2112.)
[5] 陈华友,何迎,周礼刚,.广义直觉模糊加权交叉
影响平均算子及其在多属性决策中的应用[J]. 控制与
决策, 2014, 29(7): 1250-1256.
(Chen H Y, He Y D, Zhou L G, et al. Generalized
intuitionistic fuzzy interaction averaging operators and
their applications to multi-attribute decision making[J].
Control and Decision, 2014, 29(7): 1250-1256.)
[6] 周伟,何建敏,余德建.直觉模糊群决策中专家权重
确定的一种精确方法[J]. 控制与决策, 2013, 28(5):
716-725.
(Zhou W, He J M, Xu D J. Accurate method of obtaining
decision expert weights in intuitionistic fuzzy group
decision making[J]. Control and Decision, 2013, 28(5):
716-725.)
[7] 江红 ,建敏,亚明,.基于直觉模糊集和证
据理论的群决策方法[J]. 控制与决策, 2012, 27(5):
752-756.
(Jiang H L, He J M, Zhuang Y M, et al. Approach to group
decision making based on intuitionistic fuzzy sets and
evidence theory[J]. Control and Decision, 2012, 27(5):
752-756.)
[8] 陶长琪,凌和良.基于 Choquet 积分的模糊数直觉模
糊数多属性决策方法[J]. 控制与决策, 2012, 27(9):
1381-1386.
(Tao C Q, Ling H L. Approach for
multiple attribute decision-making with fuzzy-
number-intuitionistic-fuzzy-number based on Choquet
integral[J]. Control and Decision, 2012, 27(9): 1381-
1480 33
1386.)
[9] Yager R R, Abbasov A M. Pythagorean membership
grades, complex numbers, and decision making[J]. Int
J Intelligent Systems, 2013; 28: 436-452.
[10] Yager R R. Pythagorean membership grades in
multicriteria decision making[J]. IEEE Trans on Fuzzy
Systems, 2014, 22: 958-965.
[11] Zhang X L, Xu Z S. Extension of TOPSIS to multiple
criteria decision making with Pythagorean fuzzy sets[J].
Int J Intelligent Systems, 2014, 29: 1061-1078.
[12] Peng X D, Yang Y. Some results for Pythagorean fuzzy
sets[J]. Int J Intelligent Systems, 2015, 30: 1133-1160.
[13] Zhang X L. A novel approach based on similarity measure
for pythagorean fuzzy multiple criteria group decision
making[J]. Int J of Intelligent Systems, 2015, 31(6):
593-611.
[14] Yang Y, Ding H, Chen Z, et al. A note on extension
of TOPSIS to multiple criteria decision making with
pythagorean fuzzy sets[J]. Int J of Intelligent Systems,
2015, 31(1): 68-72.
[15] 刘卫锋,常娟,何霞.广义毕达哥拉斯模糊集成算子及
其决策应用[J]. 控制与决策, 2016, 31(12): 2280-2286.
(Liu W F, Chang J, He X. Generalized Pythagorean
fuzzy aggregation operators and applications in decision
making[J]. Control and Decision, 2016, 31(12): 2280-
2286.)
[16] 刘卫锋,杜迎雪,常娟.毕达哥拉斯模糊交叉影响集
成算子及其决策应用[J]. 控制与决策, 2017, 32(6):
1033-1040.
(Liu W F, Du Y X, Chang J, Pythagorean fuzzy
interaction aggregation operators and applications in
decision making[J]. Control and Decision, 2017, 32(6):
1033-1040.)
[17] 李德 ,文艺.勾股模糊集的距离测度及其在多
属性决策中的应用[J]. 控制与决策, 2017, 32(10):
1817-1823.
(Li D Q, Zeng W Y. Distance measures of Pythagorean
fuzzy sets and their applications in multiattribute
decision making[J]. Control and Decision, 2017, 32(10):
1817-1823.)
[18] Ma Z M, Xu Z S. Symmetric pythagorean fuzzy weighted
geometric/averaging operators and their application
in multicriteria decisionmaking problems[J]. Int J of
Intelligent Systems, 2016, 31(12): 1198-1219.
[19] Ren P J, Xu Z S, Gou X J. Pythagorean fuzzy TODIM
approach to multi-criteria decision making[J]. Applied
Soft Computing, 2016, 42(5): 246-259.
[20] 彭新东,杨勇. Pythagorean 模糊语言集多属性群
决策方法[J]. 计算机工程与应用, 2016, 52(23): 50-54.
(Peng X D, Yang Y. Multiple attribute group decision
making methods based on Pythagorean fuzzy linguistic
set[J].Computer Engineering and Applications, 2016,
52(23): 50-54.)
[21] Peng X D, Yang Y. Pythagorean fuzzy choquet integral
based MABAC method for multiple attribute group
decision making[J]. Int J of Intelligent Systems, 2016,
31(10): 989-1020.
[22] Gou X J, Xu Z S, Ren P J. The properties of continuous
pythagorean fuzzy information[J]. Int J of Intelligent
Systems, 2016, 31(5): 401-424.
[23] Frank M J. On the simultaneous associativity of F(x, y)
and x+yF(x, y)[J]. Aequationes Mathematicae, 1979,
19(1): 194-226.
[24] Zhang X, Liu P D, Wang Y M. Multiple attribute group
decision making methods based on intuitionistic fuzzy
frank power aggregation operators[J]. J of Intelligent &
Fuzzy Systems, 2015, 29(5): 2235-2246.
[25] Zhang Z M. Interval-valued intuitionistic fuzzy Frank
aggregation operators and their applications to multiple
attribute group decision making[J]. Neural Computing &
Applications, 2017, 28(6): 1471-1501.
[26] Qin J D, Liu X W, Pedrycz W. Frank aggregation
operators and their application to hesitant fuzzy multiple
attribute decision making[J]. Applied Soft Computing,
2016, 41(4): 428-452.
[27] 杜玉琴,侯福均,区间直觉语言 Frank 集结算子及其在
决策中的应用[J], 控制与决策, 2018, 33(1): 119-125.
(Du Y Q, Hou F J. Interval intuitionistic linguistic Frank
aggregation operators and their application in decision
making[J]. Control and Decision, 2018, 33(1): 119-125.)
[28] Qin J D, Liu X W. Frank aggregation operators for
triangular interval type-2 fuzzy set and its application in
multiple attribute group decision making[J]. J of Applied
Mathematics, 2014, 3: 1-24.
[29] Klement E P, Mesiar R, Pap E. Triangular norms[J].
Trends in Logic, 2000, 13(1): 169-193.
[30] Beliakov G, Pradera A, Calvo T. Aggregation functions:
A guide for practitioners[M]. Berling: Springer-Verlag,
2007: 139-141.
[31] Klir G, Yuan B. Fuzzy sets and fuzzy logic[M]. New
Jersey: Prentice hall, 1995: 414-446.
(责任编辑:郑晓蕾)
... Casasnovas and Torrens (2003) proposed an axiomatic approach to the scalar cardinality of FTNs. Several researchers proposed aggregation operators using FTN and FTCN in different fuzzy environments to consolidate uncertain information, for instance, intuitionistic fuzzy environment (Zhang et al. 2015), Pythagorean fuzzy environment (Xing et al. 2018;Yi et al. 2018), hesitant fuzzy environment (Qin et al. 2016), interval-valued neutrosophic fuzzy environment (Zhou et al. 2019) and triangular interval type-2 fuzzy environment (Qin and Liu 2014). ...
... Under the q-ROF environment, we elaborate a strict comparison between our proposed Frank aggregation operators with the existing ones. We calculate the decisionmaking results of our proposed method using intuitionistic fuzzy Frank power average (IFFPA) operator (Zhang et al. 2015), intuitionistic fuzzy Frank power weighted average (IFFPWA) operator (Zhang et al. 2015), Pythagorean fuzzy Frank weighted averaging (PFFWA) operator (Yi et al. 2018), Pythagorean fuzzy Frank weighted geometric(PFFWG) operator (Yi et al. 2018), q-ROF weighted averaging (q-ROFWA) operator (Liu and Wang 2018), q-ROF weighted geometric (q-ROFWG) operator (Liu and Wang 2018). The outcomes of the comparison are illustrated in Table 9. ...
... Under the q-ROF environment, we elaborate a strict comparison between our proposed Frank aggregation operators with the existing ones. We calculate the decisionmaking results of our proposed method using intuitionistic fuzzy Frank power average (IFFPA) operator (Zhang et al. 2015), intuitionistic fuzzy Frank power weighted average (IFFPWA) operator (Zhang et al. 2015), Pythagorean fuzzy Frank weighted averaging (PFFWA) operator (Yi et al. 2018), Pythagorean fuzzy Frank weighted geometric(PFFWG) operator (Yi et al. 2018), q-ROF weighted averaging (q-ROFWA) operator (Liu and Wang 2018), q-ROF weighted geometric (q-ROFWG) operator (Liu and Wang 2018). The outcomes of the comparison are illustrated in Table 9. ...
Article
Full-text available
In decision-making problems, q-rung orthopair fuzzy sets are considered as a more effective tool than intuitionistic fuzzy sets and Pythagorean fuzzy sets. This article develops some aggregation operators based on Frank t-norm and t-conorm for fusing q-rung orthopair fuzzy (q-ROF) information. Then a multiple attribute decision-making (MADM) approach is introduced based on the proposed operators. The Frank operations of t-norm and t-conorm can have the advantage of good flexibility with the operational parameter. From that point of view, in this paper, we extend the ideas of Frank t-norm and t-conorm to the q-ROF environment and introduce some aggregation operators. Moreover, we illustrate the compatible properties of the proposed operators. Generally, the attribute weights are unknown in the MADM problems. The analytical hierarchy process and entropy methods are efficient tools to handle such MADM problems with unknown attribute weights. So, we present an MADM approach with unknown attribute weights under q-ROF environment using proposed operators. Then to elaborate the flexibility and validity of the proposed model, we discuss and solve a numerical problem concerned with a government project of choosing the best way of industrialization. Next, we show how the involvement of the parameters in our proposed model affects the decision-making results. Finally, to exhibit the superiority of our proposed methodology, the obtained results are compared with the existing ones.
... Peng and Yang [11] researched subtraction and division operations on PFN. In addition, two hot topics are focused on aggregation operators (AO) [12][13][14][15][16][17][18][19] and decision-making algorithms [10,[20][21][22][23][24][25][26]. ...
... Because of the shortcomings (division by zero problem [10], counterintuitive phenomena [13,15,18]) of the existing PF decision-making algorithms, DMs may have encountered obstacle for selecting a convincing or best solution. Hence, the goal of this article is to handle the two issues discussed above by developing a Pythagorean fuzzy MCDM approach, which not only avoid division by zero problem, but also can obtain best alternative without counterintuitive phenomenon. ...
... (iii) Novel score function for PFN with dynamically adjustable parameter is provided for managing comparison failure issue [10,11,13]. (iv) The developed algorithm-based multiparametric similarity measure and ideal point with some selected existing methods [10,13,15,18] is compared by big data industry problem. It has no counterintuitive issues [13,15,18] and division by zero issue [10]. ...
Article
Full-text available
Big data industry decision is supremely important for companies to boost the efficiency of leadership, which can vastly accelerate industrialized. With regard to big data industry decision assessment, the intrinsic problem involves enormous inexactness, fuzziness and ambiguity. Pythagorean fuzzy sets (PFSs), managing the uncertainness depicted in non-membership with membership, are a quite practical way to capture uncertainness. Firstly, the innovative Pythagorean fuzzy score function is given to dispose the comparison issue. Innovative distance measure and similarity measure for PFSs with three parameters are explored, along with corresponding proofs therewith. Later, objective weight is ascertained by deviation-based method. Also, combined weight is skillfully designed, which can tellingly imply both subjective preference and objective preference. In addition, an approach to settle Pythagorean fuzzy problem by multiparametric similarity measure is presented. The efficacy of developed algorithm is elaborated by a big data industry decision issue. Moreover, a comparison of the introduced algorithm with the selected existing methods has been built on the basis of the division by zero issue and counterintuitive phenomena for displaying its effectiveness.
... Sarkoci [81] provided an investigation on the domination of the classes of Frank and Hamacher -norms. Recently, several Frank aggregation operators [82,83,84,85,86,87,88,89,90,91] have been proposed using the Frank triangular norms for MCDM problems. ...
... Frank operations [79] based on Frank triangular norms are more flexible and robust as compared to the Algebraic operations and the Einstein operations due to the presence of an additional parameter. Applicability of Frank -norms can be seen in various practical decision-making problems [82,83,84,85,86,87,88,89,90,91]. 3. Muirhead mean [25] expresses the inter-relationship among several input values and also provides more flexibility and robustness during the aggregation process due to a vector parameter. ...
Article
Full-text available
Pythagorean cubic fuzzy sets (PCFSs) are the most convenient aid to depict ambiguous data in practical decision-making situations. The Muirhead mean (MM) possesses the ability to capture the interrelationship between the criteria. Further, the MM generalizes several significant operators, for instance, the Bonferroni mean (BM) and the Maclaurin symmetric mean (MSM), etc. The Frank triangular norms can offer significant adaptability and robustness due to the presence of an additional parameter as compared to other families of triangular norms. Therefore, Muirhead mean and Frank operations can be combined for obtaining suitable results during decision-making under Pythagorean cubic fuzzy information with interrelated criteria. The primary goal of this study is to introduce some new MM operators based on Frank operations, namely, PCFFMM and PCFFGMM. Then some of the properties of PCFFMM and PCFFGMM are described. Moreover, an MCDM approach is formulated using PCFFMM and PCFFGMM operators. Eventually, the projected MCDM approach is demonstrated with a numerical example. The phenomenon of interrelationships among criteria of real-life problems can be suitably handled by the PCFFMM and PCFFGMM. The overall ranking result obtained by utilizing the proposed operators PCFFMM or PCFFGMM is more appropriate as compared to the ones obtained by using the BM, MSM, Heronian mean (HM), Choquet integral (CI), etc. Moreover, PCFFMM or PCFFGMM can be preferred ahead of BM, MSM, HM, and CI due to its computation simplicity, accuracy, flexibility, and robustness during the aggregation process of multiple correlated input data.
... Peng and Yang [4] explored the subtraction and division operations for PFNs and initiated a Pythagorean fuzzy SIR (PF-SIR) algorithm to handle multi-criteria group decision-making (MCGDM) issue. Moreover, some extension models [5][6][7][8] [9][10][11][12][13][14][15][16][17][18][19][20][21][22], decisionmaking methods [3,[23][24][25][26][27][28][29][30][31]. ...
Article
Full-text available
Pythagorean fuzzy set (PFS), disposing the indeterminacy portrayed by membership and non-membership, is a more viable and effective means to seize indeterminacy. Due to the defects of existing Pythagorean fuzzy similarity measures or distance measures (cannot obey third or fourth axiom; have no power to differentiate positive and negative difference; have no power to deal the division by zero problem), the major key of this paper is to explore the novel Pythagorean fuzzy distance measure and similarity measure. Meanwhile, some interesting properties of distance measure and similarity measure are proved. Some counterintuitive examples are presented to state their availability of similarity measure among PFSs.
Article
Accumulation of q ˜ -rung picture fuzzy information plays an essential part in decision-making situations. q ˜ -rung picture fuzzy sets can handle the uncertain information more precisely and flexibly because of the presence of parameter q ˜ . Also the Frank t-norm and t-conorm operations perform suitably for the data accumulation with the operational parameter. In this paper, we introduce q ˜ -Rung picture fuzzy Frank weighted averaging operator and q ˜ -rung picture fuzzy Frank weighted geometric operator by extending q ˜ -rung orthopair fuzzy Frank arithmetic and geometric aggregation operators respectively. We establish an algorithm to address the tedious decision-making problems using these operators. Eventually, we discuss a multiple attribute decision-making problem to demonstrate the utility and efficacy of the proposed method. A comparison of existing methods is made to reveal the supremacy and benefits of our proposed method.
Article
Big Data allows analysing and assessing all human production types with its 5Vs, which are Volume, Velocity, Variety, Veracity and Value. Big Data is useful to improve decision-making to adjust it better to market demand, specifically selection of supplier that is an important link to optimize the logistic chain of enterprises. In this case, leadership or decider is ahead one serious complex problem, inexact and fuzzy. Pythagorean fuzzy set (PFS) is disposing the indeterminacy data by the membership and the nonmembership functions; it is a generalization of the intuitionist fuzzy set when the last set is limited. First, some results for PFSs are displaying in this study as particular cases and generalization of some binary operations. After, an improved score function of Pythagorean fuzzy number is proposed to avoid the comparison problem in practice. In addition, an existing approach exploring the combined alternatives weight to settle Pythagorean fuzzy issue by multi-parametric similarity measure is applied with the new proposed score function to selection of supplier issue with five serious criteria as a Big Data industry decision-making problem in economic environment. Finally, a comparison of the presented method with some existing approaches has been executed in the light of counterintuitive phenomena for validating its advantages.
Article
Full-text available
Aggregation operators under the Pythagorean fuzzy environment and their applications to decision making are discussed. The Quasi-weighted geometric (QWG) operator and the Quasi-ordered weighted geometric (QOWG) operator are defined, and their natures are studied. Then, a class of aggregation operators called Pythagorean fuzzy aggregation operators are proposed, including the Pythagorean fuzzy order weighted average (PFOWA) operator, the generalized Pythagorean fuzzy order weighted average (GPFOWA) operator, the Pythagorean fuzzy order weighted geometric (PFOWG) operator, the generalized Pythagorean fuzzy order weighted geometric (GPFOWG) operator, the Quasi Pythagorean fuzzy order weighted average (QPFOWA) operator and the Quasi Pythagorean fuzzy order weighted geometric (QPFOWG) operator. A method based on generalized Pythagorena fuzzy aggregation operators for decision making is presented, and an example is given to illustrate the feasibility of the proposed method. © 2016, Editorial Office of Control and Decision. All right reserved.
Article
Full-text available
Pythagorean fuzzy sets (PFSs), originally proposed by Yager, are a new tool to deal with vagueness with the square sum of the membership degree and the nonmembership degree equal to or less than 1, which have much stronger ability than Atanassov's intuitionistic fuzzy sets to model such uncertainty. In this paper, we modify the existing score function and accuracy function for Pythagorean fuzzy number to make it conform to PFSs. Associated with the given operational laws, we define some novel Pythagorean fuzzy weighted geometric/averaging operators for Pythagorean fuzzy information, which can neutrally treat the membership degree and the nonmembership degree, and investigate the relationships among these operators and those existing ones. At length, a practical example is provided to illustrate the developed operators and to make a comparative analysis.
Article
Full-text available
Interval-valued intuitionistic fuzzy numbers (IVIFNs), which contain three ranges: the membership degree range, the non-membership degree range, and the hesitancy degree range, are very suitable to be used for depicting uncertain or fuzzy information. In this paper, we study the aggregation techniques of IVIFNs with the help of Frank operations. We first extend the Frank t-conorm and t-norm to interval-valued intuitionistic fuzzy environments and introduce several new operations of IVIFNs, such as Frank sum, Frank product, Frank scalar multiplication, and Frank exponentiation, based on which we develop several new interval-valued intuitionistic fuzzy aggregation operators, including the interval-valued intuitionistic fuzzy Frank weighted averaging operator and the interval-valued intuitionistic fuzzy Frank weighted geometric operator. We further establish various properties of these operators, give some special cases of them, and analyze the relationships between these operators. Moreover, we apply these operators to develop an approach for dealing with multiple attribute group decision making with interval-valued intuitionistic fuzzy information. Finally, a numerical example is provided to illustrate the practicality and effectiveness of the developed operators and approach.
Article
Based on the interval intuitionistic linguistic variables and the Frank operator, the operational laws of the Frank operator under the interval intuitionistic linguistic circumstance are firstly defined. Then, several interval intuitionistic linguistic Frank aggregation operators are introduced, such as the interval intuitionistic linguistic Frank weighted average operator, the interval intuitionistic linguistic Frank weighted geometric operator, and the interval intuitionistic linguistic Frank generalized weighted average operator, etc. and some of their properties are discussed. Moreover, two approaches are introduced on the basis of the interval intuitionistic linguistic Frank aggregation operators, and an example is given to illustrate the feasibility and effectiveness of the proposed approaches. © 2018, Editorial Office of Control and Decision. All right reserved.
Article
Firstly, via comparing the features of intuitionistic fuzzy set(FS) and Pythagorean fuzzy set(PFS), three ranking methods for Pythagorean fuzzy numbers(PFNs) are analyzed, and some flaws of two ranking methods are pointed out. Then, it is illuminated that each PFN is characterized by four parameters, i.e., the membership degree, the nonmembership degree, the strength of commitment and the direction of strength. Simultaneously, the distance measures of PFSs and PFNs are investigated. The Hamming distance measure, the Euclidean distance measure and the Minkowski distance measure between PFSs and PFNs are proposed, and the desired properties are discussed. Finally, a multiple attribute decision making method in the Pythagorean fuzzy environment based on the proposed distance measures is presented. A numerical example is provided to illustrate the validity and applicability of the presented approach. © 2017, Editorial Office of Control and Decision. All right reserved.
Article
The interaction operations over Pythagorean fuzzy numbers are defined, including addition, scalar multiplication, multiplication and power operations. Then, the Pythagorean fuzzy interaction weighted averaging (PFIWA) operator, Pythagorean fuzzy interaction ordered weighted averaging (PFIOWA) operator, Pythagorean fuzzy interaction weighted geometric (PFIWG) operator, and Pythagorean fuzzy interaction ordered weighted geometric (PFIOWG) operator are proposed, and the mathematical expressions of these operators are obtained by derivation and some properties of these operators are investigated. A method based on Pythagorean fuzzy interaction aggregation operators for decision making is presented, and an illustrative example is given to illustrate the stability and effectiveness of the proposed method. © 2017, Editorial Office of Control and Decision. All right reserved.
Article
Article
Recently, the TODIM (an acronym in Portuguese for Interactive Multi-criteria Decision Making) approach, which can characterize the decision makers' psychological behaviours under risk, has been introduced to handle multi-criteria decision making (MCDM) problems. Moreover, Pythagorean fuzzy set is an effective tool for depicting uncertainty of the MCDM problems. In this paper, based on the prospect theory, we first extend the TODIM approach to solve the MCDM problems with Pythagorean fuzzy information. Then, we conduct simulation tests to analyze how the risk attitudes of the decision makers exert the influence on the results of MCDM under uncertainty. Finally, a case study on selecting the governor of Asian Infrastructure Investment Bank is made to show the applicability of the proposed approach.
Article
In this paper, we define the Choquet integral operator for Pythagorean fuzzy aggregation operators, such as Pythagorean fuzzy Choquet integral average (PFCIA) operator and Pythagorean fuzzy Choquet integral geometric (PFCIG) operator. The operators not only consider the importance of the elements or their ordered positions but also can reflect the correlations among the elements or their ordered positions. It is worth pointing out that most of the existing Pythagorean fuzzy aggregation operators are special cases of our operators. Meanwhile, some basic properties are discussed in detail. Later, we propose two approaches to multiple attribute group decision making with attributes involving dependent and independent by the PFCIA operator and multi-attributive border approximation area comparison (MABAC) in Pythagorean fuzzy environment. Finally, two illustrative examples have also been taken in the present study to verify the developed approaches and to demonstrate their practicality and effectiveness.
Article
In this paper, we investigate multiple attribute decision making (MADM) problems based on Frank triangular norms, in which the attribute values assume the form of hesitant fuzzy information. Firstly, some basic concepts of hesitant fuzzy set (HFS) and the Frank triangle norms are introduced. We develop some hesitant fuzzy aggregation operators based on Frank operations, such as hesitant fuzzy Frank weighted average (HFFWA) operator, hesitant fuzzy Frank ordered weighted averaging (HFFOWA) operator, hesitant fuzzy Frank hybrid averaging (HFFHA) operator, hesitant fuzzy Frank weighted geometric (HFFWG) operator, hesitant fuzzy Frank ordered weighted geometric (HFFOWG) operator, and hesitant fuzzy Frank hybrid geometric (HFFHG) operator. Some essential properties together with their special cases are discussed in detail. Next, a procedure of multiple attribute decision making based on the HFFHWA (or HFFHWG) operator is presented under hesitant fuzzy environment. Finally, a practical example that concerns the human resource selection is provided to illustrate the decision steps of the proposed method. The result demonstrates the practicality and effectiveness of the new method. A comparative analysis is also presented.