Abstract
In this paper an innovative method of ranking neutrosophic number based on the notions of value and ambiguity of a single-valued neutrosophic number is being developed. The method is based on the convex combination of value and ambiguity of truth-membership function with the sum of values and ambiguities of indeterminacy-membership and falsity-membership functions. This convex combination is also termed as an index of optimism. The index of optimism, \(\lambda =1\), is termed as optimistic decision-maker as it considers the value and the ambiguity of the truth-membership function, ignoring the contributions from indeterminacy-membership and falsity-membership functions. Similarly, the index of optimism, \(\lambda =0\), is termed as pessimistic decision-maker as it considers the values and the ambiguities of the indeterminacy-membership and falsity-membership functions, ignoring the contribution from truth-membership function. Further, the index of optimism, \(\lambda =0.5\), is termed as moderate decision-maker as it considers the values and the ambiguities of all the membership functions. The approach is a novel as it completely oath to follow the reasonable properties of a ranking method. It is worth to mention that the current approach consistently ranks the single-valued neutrosophic numbers as well as their corresponding images.
Similar content being viewed by others
References
Aal SIA, Ellatif MAA, Hassan MM (2018) Two ranking methods of single valued triangular neutrosophic numbers to rank and evaluate information systems quality. Neutrosophic Sets Syst 19:131–141
Atanassov KT (1989) More on intuitionistic fuzzy sets. Fuzzy Sets Syst 33(1):37–45
Atanassov KT (1999) Intuitionistic fuzzy sets. Springer, Berlin, pp 1–137
Atanassov KT (2000) Two theorems for intuitionistic fuzzy sets. Fuzzy Sets Syst 110(2):267–269
Biswas P, Pramanik S, Giri BC (2016) Value and ambiguity index based ranking method of single-valued trapezoidal neutrosophic numbers and its application to multi-attribute decision making. Neutrosophic Sets Syst 12:127–138
Chakraborty D, Jana DK, Roy TK (2015) Arithmetic operations on generalized intuitionistic fuzzy number and its applications to transportation problem. Opsearch 52:431–471
Chutia R (2017) Ranking of fuzzy numbers by using value and angle in the epsilon-deviation degree method. Appl Soft Comput 60:706–721
Chutia R (2020) Ranking of Z-numbers based on value and ambiguity at levels of decision making. Int J Intell Syst 36(1):313–331
Chutia R, Chutia B (2017) A new method of ranking parametric form of fuzzy numbers using value and ambiguity. Appl Soft Comput 52(Supplement C):1154–1168
Chutia R, Saikia S (2018) Ranking intuitionistic fuzzy numbers at levels of decision-making and its application. Expert Syst 35(5):e12292
Deli I (2018) Operators on single valued trapezoidal neutrosophic numbers and SVTN-group decision making. Neutrosophic Sets Syst 22:131–150
Deli I, Subas Y (2017) A ranking method of single valued neutrosophic numbers and its applications to multi-attribute decision making problems. Int J Mach Learn Cybern 8(4):1309–1322
Edalatpanah SA (2020a) Neutrosophic structured element. Expert Syst 37(5):e12542
Edalatpanah SA (2020b) Systems of neutrosophic linear equations. Neutrosophic Sets Syst 33(1):92–104
Edalatpanah SA (2020) Data envelopment analysis based on triangular neutrosophic numbers. CAAI Trans Intell Technol 5(2):94–98
Edalatpanah SA (2020) A direct model for triangular neutrosophic linear programming. Int J Neutrosophic Sci 1(1):19–28
Edalatpanah SA, Smarandache F (2019) Data envelopment analysis for simplified neutrosophic sets. Neutrosophic Sets Syst 29:215–226
Giri B, Molla MU, Biswas P (2018) Topsis method for MADM based on interval trapezoidal neutrosophic number. Neutrosophic Sets Syst 22:151–167
Jana C, Pal M, Karaaslan F, Wang JQ (2020) Trapezoidal neutrosophic aggregation operators and their application to the multi-attribute decision-making process. Sci Iran 27(3):1655–1673
Karaaslan F (2018a) Gaussian single-valued neutrosophic numbers and its application in multi-attribute decision making. Neutrosophic Sets Syst 22(1):101–117
Karaaslan F (2018b) Multi-criteria decision-making method based on similarity measures under single-valued neutrosophic refined and interval neutrosophic refined environments. Int J Intell Syst 33(5):928–952
Karaaslan F, Davvaz B (2018) Properties of single-valued neutrosophic graphs. J Intell Fuzzy Syst 34(1):57–79
Karaaslan F, Hayat K (2018) Some new operations on single-valued neutrosophic matrices and their applications in multi-criteria group decision making. Appl Intell 48(12):4594–4614
Karaaslan F, Hunu F (2020) Type-2 single-valued neutrosophic sets and their applications in multi-criteria group decision making based on topsis method. J Ambient Intell Humaniz Comput 11:4113–4132
Lee JG, Senel G, Lim PK, Kim J, Hur K (2020) Octahedron sets. Ann Fuzzy Math Inform 19(3):211–238
Mao X, Guoxi Z, Fallah M, Edalatpanah SA (2020) A neutrosophic-based approach in data envelopment analysis with undesirable outputs. Math Probl Eng
Nancy Garg H (2016) An improved score function for ranking neutrosophic sets and its application to decision-making process. Int J Uncertain Quantif 6(5):377–385
Peng JJ, Wang JQ, Zhang HY, Chen XH (2014) An outranking approach for multi-criteria decision-making problems with simplified neutrosophic sets. Appl Soft Comput 25:336–346
Smarandache F (1998) A unifying field in logics neutrosophy: neutrosophic probability, set and logic. American Research Press, Rehoboth
Smarandache F (1999) A unifying field in logics: Neutrosophic logic. In: Philosophy. American Research Press, Rehoboth, pp 1–141
Smarandache F (2006) Neutrosophic set - a generalization of the intuitionistic fuzzy set. In: 2006 IEEE International Conference on Granular Computing, pp 38–42
Ulucay V, Kilic A, Yildiz I, Sahin M (2018) A new approach for multi-attribute decision-making problems in bipolar neutrosophic sets. Neutrosophic Sets Systs 23(1):142–159
Ulucay V, Kilic A, Yildiz I, Sahin M (2019) An outranking approach for MCDM-problems with neutrosophic multi-sets. Neutrosophic Sets Syst 30(1):213–224
Wang X, Kerre EE (2001a) Reasonable properties for the ordering of fuzzy quantities (I). Fuzzy Sets Syst 118(3):375–385
Wang X, Kerre EE (2001b) Reasonable properties for the ordering of fuzzy quantities (II). Fuzzy Sets Syst 118(3):387–405
Wang H, Smarandache F, Zhang Y, Sunderraman R (2010) Single valued neutrosophic sets. Multispace Multistruct 4:410–413
Yang W, Cai L, Edalatpanah SA, Smarandache F (2020) Triangular single valued neutrosophic data envelopment analysis: application to hospital performance measurement. Symmetry 12(4):588
Zadeh LA (1965) Fuzzy sets. Inf Control 8(3):338–353
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Proofs of the theorems
Proofs of the theorems
1.1 Proof of the Theorem 3.2
The proof of the above statements are as follows.
-
1.
The proof of this statement is followed immediately.
-
2.
Consider the cases when \(\widetilde{a}\succ \widetilde{b}\) happens, that is,
$$\begin{aligned} \widetilde{a}\succ \widetilde{b}\,\text {happens for}\, {\left\{ \begin{array}{ll} &{}{\mathcal {R}}_{\lambda }(\widetilde{a},0,0,0)>{\mathcal {R}}_{\lambda }(\widetilde{b},0,0,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},0,0,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{b},0,0,\pm 1)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},0,\pm 1,0)>{\mathcal {R}}_{\lambda }(\widetilde{b},0,\pm 1,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},0,\pm 1,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{b},0,\pm 1,\pm 1)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,0,0)>{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,0,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,0,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,0,\pm 1)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,\pm 1,0)>{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,\pm 1,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,\pm 1,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,\pm 1,\pm 1) \end{array}\right. }. \end{aligned}$$Consider the cases when \(\widetilde{b}\succ \widetilde{c}\) happens, that is,
$$\begin{aligned} \widetilde{b}\succ \widetilde{c}\,\text {happens for}\, {\left\{ \begin{array}{ll} &{}{\mathcal {R}}_{\lambda }(\widetilde{b},0,0,0)>{\mathcal {R}}_{\lambda }(\widetilde{c},0,0,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},0,0,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{c},0,0,\pm 1)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},0,\pm 1,0)>{\mathcal {R}}_{\lambda }(\widetilde{c},0,\pm 1,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},0,\pm 1,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{c},0,\pm 1,\pm 1)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,0,0)>{\mathcal {R}}_{\lambda }(\widetilde{c},\pm 1,0,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,0,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{c},\pm 1,0,\pm 1)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,\pm 1,0)>{\mathcal {R}}_{\lambda }(\widetilde{c},\pm 1,\pm 1,0)\\ &{}{\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,\pm 1,\pm 1)>{\mathcal {R}}_{\lambda }(\widetilde{c},\pm 1,\pm 1,\pm 1) \end{array}\right. } \end{aligned}$$Now, to proof this property, it need to discuss these \(8\times 8\) cases, which will make the proof tedious. However, one can see the proof trivially, if following claims can be established. Claim 1: Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=0\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\). Then \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). The proof of this claim is as follows. Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=0\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\), then \({\mathcal {V}}(\mu _{\widetilde{b}})\ne {\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{b}})\ne {\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{b}})\ne {\mathcal {V}}(\nu _{\widetilde{c}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{c}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). Claim 2: Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=\pm 1\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\). Then \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). The proof of this claim is as follows. Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=\pm 1\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\), then \({\mathcal {V}}(\mu _{\widetilde{b}})={\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{b}})={\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{b}})={\mathcal {V}}(\nu _{\widetilde{c}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{c}})\). So, \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). Claim 3: Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=\pm 1\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\). then \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). The proof of this claim is as follows. Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=\pm 1\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\), then \({\mathcal {V}}(\mu _{\widetilde{b}})={\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{b}})={\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{b}})={\mathcal {V}}(\nu _{\widetilde{c}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{c}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). Claim 4: Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=0\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\). Then \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). The proof of this claim is as follows. Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=0\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\), then \({\mathcal {V}}(\mu _{\widetilde{b}})\ne {\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{b}})\ne {\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{b}})\ne {\mathcal {V}}(\nu _{\widetilde{c}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{c}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{c}\). From these four claims, it is trivial enough to show that if \(\widetilde{a}\succ \widetilde{b}\) and \(\widetilde{b}\succ \widetilde{c}\), then \(\widetilde{a}\succ \widetilde{c}\). Further, from the definition of \(\succeq\), it follows that transitivity also holds for the order relation \(\succeq\).
-
3.
This statement is followed immediately, as the order relations \(\succ\) and \(\sim\) particularly based on order relation > and \(=\) of real numbers.
-
4.
If \(\widetilde{a}=\widetilde{b}\), then \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)={\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\). Thus, the statement is followed.
1.2 Proof of the Theorem 3.5
The proof of this theorem, follows immediately if the invariance of \(\theta _i\) in ordering \(\widetilde{a}\), \(\widetilde{b}\) and \(\widetilde{a}+\widetilde{c}\), \(\widetilde{b}+\widetilde{c}\) can be established. Hence, a claim has to be made. The claim is as follows.
Claim : The value of \(\theta _i\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\) are invariant in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\). The proof of the claim follows from the following eight cases:
- Case 1::
-
Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 2::
-
Let \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 3::
-
Let \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 4::
-
Let \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})={\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 5::
-
Let \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 6::
-
Let \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 7::
-
Let \(\theta _1=\pm 1\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _1=\pm 1\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
- Case 8::
-
Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})= {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{c}})\). So, \(\theta _i=\pm 1\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{c}\).
The above eight cases suggest that \(\theta _1\) and \(\theta _2\) are invariant in ordering \(\widetilde{a}\), \(\widetilde{b}\) and \(\widetilde{a}+\widetilde{c}\), \(\widetilde{b}+\widetilde{c}\). Hence, the claim.
Now, by the Theorem 3.1 it follows that
and
Hence, if \(\widetilde{a}\succeq \widetilde{b}\), then it is obvious that \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\). This leads to the inequality \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)+{\mathcal {R}}_{\lambda }(\widetilde{c},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)+{\mathcal {R}}_{\lambda }(\widetilde{c},\theta _1,\theta _2,\theta _3)\), which evidently follows the inequality \({\mathcal {R}}_{\lambda }(\widetilde{a}+\widetilde{c},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b}+\widetilde{c},\theta _1,\theta _2,\theta _3)\). Thus, the result follows immediately.
1.3 Proof of the Theorem 3.9
Let \(\widetilde{a}\succeq \widetilde{b}\). Then \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\). Let \(k>0\). Then using the Proposition 3.4, it follows that
Thus, when \(\widetilde{a}\succeq \widetilde{b}\), it follows that \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\). Equivalently, it follows that \(k{\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)\ge k{\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\), which can be trivially expressed as \({\mathcal {R}}_{\lambda }(k\widetilde{a},\theta _1,\theta _2,\theta _3) \ge {\mathcal {R}}_{\lambda }(k\widetilde{b},\theta _1,\theta _2,\theta _3)\). So, the result, \(k\widetilde{a}\succeq k\widetilde{b}\), follows immediately.
Let \(k<0\), assume \(k=-m<0\), then the following cases arise.
- Case 1::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _i=0\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},0,0)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},0,0)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _i=0\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,0,0)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,0,0)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 2::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\pm 1\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},0,0,\pm 1)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},0,0,\pm 1)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\pm {\mathcal {A}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\pm {\mathcal {A}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\mp 1\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\mp {\mathcal {A}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\mp {\mathcal {A}}(\nu _{-m\widetilde{a}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,0,\mp 1)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,0,\mp 1)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 3::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=0\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},0,\pm 1,0)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},0,\pm 1,0)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\pm {\mathcal {A}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\pm {\mathcal {A}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _1=0\), \(\theta _2=\mp 1\) and \(\theta _3=0\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\mp {\mathcal {A}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\mp {\mathcal {A}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,\mp 1,0)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,\mp 1,0)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 4::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=\pm 1\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},0,\pm 1,\pm 1)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},0,\pm 1,\pm 1)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\pm {\mathcal {A}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\pm {\mathcal {A}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\pm {\mathcal {A}}(\nu _{\widetilde{a}}) \ge {\mathcal {V}}(\nu _{\widetilde{b}})\pm {\mathcal {A}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _1=0\), \(\theta _2=\mp 1\) and \(\theta _3=\mp 1\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\mp {\mathcal {A}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\mp {\mathcal {A}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\mp {\mathcal {A}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\mp {\mathcal {A}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,\mp 1,\mp 1)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,\mp 1,\mp 1)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 5::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=0\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,0,0)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,0,0)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\pm {\mathcal {A}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\pm {\mathcal {A}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _1=\mp 1\), \(\theta _2=0\) and \(\theta _3=0\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\mp {\mathcal {A}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\mp {\mathcal {A}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\mp 1,0,0)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},\mp 1,0,0)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 6::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=\pm 1\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,0,\pm 1)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,0,\pm 1)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\pm {\mathcal {A}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\pm {\mathcal {A}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\pm {\mathcal {A}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\pm {\mathcal {A}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _1=\mp 1\), \(\theta _2=0\) and \(\theta _3=\mp 1\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\mp {\mathcal {A}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\mp {\mathcal {A}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\mp {\mathcal {A}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\mp {\mathcal {A}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\mp 1,0,\mp 1)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},\mp 1,0,\mp 1)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 7::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _1=\pm 1\), \(\theta _2=\pm 1\) and \(\theta _3=0\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,\pm 1,0)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,\pm 1,0)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\pm {\mathcal {A}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\pm {\mathcal {A}}(\mu _{\widetilde{a}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\pm {\mathcal {A}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\pm {\mathcal {A}}(\rho _{\widetilde{a}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _1=\mp 1\), \(\theta _2=\mp 1\) and \(\theta _3=0\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\mp {\mathcal {A}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\mp {\mathcal {A}}(\mu _{-m\widetilde{a}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\mp {\mathcal {A}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\mp {\mathcal {A}}(\rho _{-m\widetilde{a}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\mp 1,\mp 1,0)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},\mp 1,\mp 1,0)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
- Case 8::
-
Let \(\widetilde{a}\succeq \widetilde{b}\) for \(\theta _i=\pm 1\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(\widetilde{a},\pm 1,\pm 1,\pm 1)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\pm 1,\pm 1,\pm 1)\). Now, as \(\widetilde{a}\succeq \widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\pm {\mathcal {A}}(\mu _{\widetilde{a}})\ge {\mathcal {V}}(\mu _{\widetilde{b}})\pm {\mathcal {A}}(\mu _{\widetilde{a}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\pm {\mathcal {A}}(\rho _{\widetilde{a}})\ge {\mathcal {V}}(\rho _{\widetilde{b}})\pm {\mathcal {A}}(\rho _{\widetilde{a}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\pm {\mathcal {A}}(\nu _{\widetilde{a}})\ge {\mathcal {V}}(\nu _{\widetilde{b}})\pm {\mathcal {A}}(\nu _{\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\). Thus, \(\theta _i=\mp 1\) in ordering \(-m\widetilde{a}\) and \(-m\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\mp {\mathcal {A}}(\mu _{-m\widetilde{a}})\le {\mathcal {V}}(\mu _{-m\widetilde{b}})\mp {\mathcal {A}}(\mu _{-m\widetilde{a}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\mp {\mathcal {A}}(\rho _{-m\widetilde{a}})\le {\mathcal {V}}(\rho _{-m\widetilde{b}})\mp {\mathcal {A}}(\rho _{-m\widetilde{a}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\mp {\mathcal {A}}(\nu _{-m\widetilde{a}})\le {\mathcal {V}}(\nu _{-m\widetilde{b}})\mp {\mathcal {A}}(\nu _{-m\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\mp 1,\mp 1,\mp 1)\le {\mathcal {R}}_{\lambda }(-m\widetilde{b},\mp 1,\mp 1,\mp 1)\). Hence, the result \(-m \widetilde{a}\preceq -m \widetilde{b}\) follows immediately.
1.4 Proof of the Theorem 3.10
Let \(k>0\) and \(k\widetilde{a}\succeq k\widetilde{b}\). Then \({\mathcal {R}}_{\lambda }(k\widetilde{a},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(k\widetilde{b},\theta _1,\theta _2,\theta _3)\). However, by Proposition 3.4, it follows that \(k{\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)\ge k{\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\). Thus, the result follows immediately. If \(k<0\), let \(k=-m<0\), then \(-m\widetilde{a}\succeq -m\widetilde{b}\) implies that \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},\theta _1,\theta _2,\theta _3)\). Now, eight cases arise.
- Case 1::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _i=0\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,0,0)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,0,0)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\)and \({\mathcal {V}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},0,0,0)\le {\mathcal {R}}_{\lambda }(\widetilde{b},0,0,0)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 2::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\pm 1\) then \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,0,\pm 1)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,0,\pm 1)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\pm {\mathcal {A}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\pm {\mathcal {A}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\mp 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\)and \({\mathcal {V}}(\nu _{\widetilde{a}})\mp {\mathcal {A}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\mp {\mathcal {A}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},0,0,\mp 1)\le {\mathcal {R}}_{\lambda }(\widetilde{b},0,0,\mp 1)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 3::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=0\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,\pm 1,0)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,\pm 1,0)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\pm {\mathcal {A}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _1=0\), \(\theta _2=\mp 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\mp {\mathcal {A}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},0,\mp 1,0)\le {\mathcal {R}}_{\lambda }(\widetilde{b},0,\mp 1,0)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 4::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=\pm 1\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ne {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},0,\pm 1,\pm 1)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},0,\pm 1,\pm 1)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\pm {\mathcal {A}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\pm {\mathcal {A}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _1=0\), \(\theta _2=\mp 1\) and \(\theta _3=\mp 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\mp {\mathcal {A}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\mp {\mathcal {A}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},0,\mp 1,\mp 1)\le {\mathcal {R}}_{\lambda }(\widetilde{b},0,\mp 1,\mp 1)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 5::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=0\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\pm 1,0,0)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},\pm 1,0,0)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\pm {\mathcal {A}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _1=\mp 1\), \(\theta _2=0\) and \(\theta _3=0\), in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\mp {\mathcal {A}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\)and \({\mathcal {V}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},\mp 1,0,0)\le {\mathcal {R}}_{\lambda }(\widetilde{b},\mp 1,0,0)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 6::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=\pm 1\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ne {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\pm 1,0,\pm 1)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},\pm 1,0,\pm 1)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\pm {\mathcal {A}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\pm {\mathcal {A}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _1=\mp 1\), \(\theta _2=0\) and \(\theta _3=\mp 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\mp {\mathcal {A}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\mp {\mathcal {A}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},\mp 1,0,\mp 1)\le {\mathcal {R}}_{\lambda }(\widetilde{b},\mp 1,0,\mp 1)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 7::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _1=\pm 1\), \(\theta _2=\pm 1\) and \(\theta _3=0\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ne {\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\pm 1,\pm 1,0)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},\pm 1,\pm 1,0)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _1=\mp 1\), \(\theta _2=\mp 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},\mp 1,\mp 1,0)\le {\mathcal {R}}_{\lambda }(\widetilde{b},\mp 1,\mp 1,0)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
- Case 8::
-
Let \(-m\widetilde{a}\succeq -m\widetilde{b}\) for \(\theta _i=\pm 1\). Then \({\mathcal {V}}(\mu _{-m\widetilde{a}})={\mathcal {V}}(\mu _{-m\widetilde{b}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})={\mathcal {V}}(\rho _{-m\widetilde{b}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})={\mathcal {V}}(\nu _{-m\widetilde{b}})\) and \({\mathcal {R}}_{\lambda }(-m\widetilde{a},\pm 1,\pm 1,\pm 1)\ge {\mathcal {R}}_{\lambda }(-m\widetilde{b},\pm 1,\pm 1,\pm 1)\). Now, as \(-m\widetilde{a}\succeq -m\widetilde{b}\) it follows that \({\mathcal {V}}(\mu _{-m\widetilde{a}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\ge {\mathcal {V}}(\mu _{-m\widetilde{b}})\pm {\mathcal {A}}(\mu _{-m\widetilde{a}})\), \({\mathcal {V}}(\rho _{-m\widetilde{a}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\ge {\mathcal {V}}(\rho _{-m\widetilde{b}})\pm {\mathcal {A}}(\rho _{-m\widetilde{a}})\) and \({\mathcal {V}}(\nu _{-m\widetilde{a}})\pm {\mathcal {A}}(\nu _{-m\widetilde{a}})\ge {\mathcal {V}}(\nu _{-m\widetilde{b}})\pm {\mathcal {A}}(\nu _{-m\widetilde{b}})\). Clearly, \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, \(\theta _i=\mp 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Further, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\le {\mathcal {V}}(\mu _{\widetilde{b}})\mp {\mathcal {A}}(\mu _{\widetilde{a}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\le {\mathcal {V}}(\rho _{\widetilde{b}})\mp {\mathcal {A}}(\rho _{\widetilde{a}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\mp {\mathcal {A}}(\nu _{\widetilde{a}})\le {\mathcal {V}}(\nu _{\widetilde{b}})\mp {\mathcal {A}}(\nu _{\widetilde{b}})\). So, \({\mathcal {R}}_{\lambda }(\widetilde{a},\mp 1,\mp 1,\mp 1)\le {\mathcal {R}}_{\lambda }(\widetilde{b},\mp 1,\mp 1,\mp 1)\). Hence, the result \(\widetilde{a}\preceq \widetilde{b}\) follows immediately.
1.5 Proof of the Theorem 3.13
The proof of this theorem, follows immediately if the invariance of \(\theta _1\) and \(\theta _2\) in ordering \(\widetilde{a}\), \(\widetilde{b}\) and \(\widetilde{a}-\widetilde{c}\), \(\widetilde{b}-\widetilde{c}\) can be established. Hence, a claim has to be made. The claim is as follows.
Claim : The value of \(\theta _1\) and \(\theta _2\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\) are invariant in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\). The proof of the claim follows from the following eight cases:
- Case 1::
-
Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 2::
-
Let \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _1=0\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 3::
-
Let \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 4::
-
Let \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})={\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _1=0\), \(\theta _2=\pm 1\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 5::
-
Let \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=0\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 6::
-
Let \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _1=\pm 1\), \(\theta _2=0\) and \(\theta _3=\pm 1\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 7::
-
Let \(\theta _1=\pm 1\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _1=\pm 1\), \(\theta _2=\pm 1\) and \(\theta _3=0\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
- Case 8::
-
Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then, \({\mathcal {V}}(\mu _{\widetilde{a}})= {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})= {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\mu _{\widetilde{b}-\widetilde{c}})\), \({\mathcal {V}}(\rho _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\rho _{\widetilde{b}-\widetilde{c}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}-\widetilde{c}})= {\mathcal {V}}(\nu _{\widetilde{b}-\widetilde{c}})\). So, \(\theta _i=\pm 1\) in ordering \(\widetilde{a}-\widetilde{c}\) and \(\widetilde{b}-\widetilde{c}\).
The above eight cases suggest that \(\theta _1\) and \(\theta _2\) are invariant in ordering \(\widetilde{a}\), \(\widetilde{b}\) and \(\widetilde{a}-\widetilde{c}\), \(\widetilde{b}-\widetilde{c}\). Hence, the claim.
Now, by the Theorem 3.1 it follows that it follows that
and
Then, if \(\widetilde{a}\succeq \widetilde{b}\), then it is obvious that \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)\). Eventually, it leads to the inequality \({\mathcal {R}}_{\lambda }(\widetilde{a},\theta _1,\theta _2,\theta _3)+{\mathcal {R}}_{\lambda }(-\widetilde{c},\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b},\theta _1,\theta _2,\theta _3)+{\mathcal {R}}_{\lambda }(-\widetilde{c},\theta _1,\theta _2,\theta _3)\). Thus, evidently it follows that \({\mathcal {R}}_{\lambda }(\widetilde{a}+(-\widetilde{c}),\theta _1,\theta _2,\theta _3)\ge {\mathcal {R}}_{\lambda }(\widetilde{b}+(-\widetilde{c}),\theta _1,\theta _2,\theta _3)\). So, the result follows immediately.
1.6 Proof of the Theorem 3.15
Consider the cases when \(\widetilde{a}\succ \widetilde{b}\) happens, that is,
Consider the cases when \(\widetilde{c}\succ \widetilde{d}\) happens, that is,
Now, to proof this property, it need to discuss these \(8\times 8\) cases, which will make the proof tedious. However, one can see the proof trivially, if following claims can be established.
Claim 1: Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=0\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\). Then \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\). The proof of this claim is as follows. Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=0\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\). Then \({\mathcal {V}}(\mu _{\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{d}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{d}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\).
Claim 2: Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=\pm 1\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\). Then \(\theta _i=\pm 1\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\). The proof of this claim is as follows. Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=\pm 1\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\), then \({\mathcal {V}}(\mu _{\widetilde{c}})={\mathcal {V}}(\mu _{\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{c}})={\mathcal {V}}(\rho _{\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{d}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})={\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})={\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{d}})\). So, \(\theta _i=\pm 1\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\).
Claim 3: Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=\pm 1\) in ordering \(\widetilde{b}\) and \(\widetilde{c}\). Then \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\). The proof of this claim is as follows. Let \(\theta _i=0\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})\ne {\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})\ne {\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})\ne {\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=\pm 1\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\), then \({\mathcal {V}}(\mu _{\widetilde{c}})={\mathcal {V}}(\mu _{\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{c}})={\mathcal {V}}(\rho _{\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{c}})={\mathcal {V}}(\nu _{\widetilde{d}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{d}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\).
Claim 4: Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\), and \(\theta _i=0\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\). Then \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\). The proof of this claim is as follows. Let \(\theta _i=\pm 1\) in ordering \(\widetilde{a}\) and \(\widetilde{b}\). Then \({\mathcal {V}}(\mu _{\widetilde{a}})={\mathcal {V}}(\mu _{\widetilde{b}})\), \({\mathcal {V}}(\rho _{\widetilde{a}})={\mathcal {V}}(\rho _{\widetilde{b}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}})={\mathcal {V}}(\nu _{\widetilde{b}})\). Similarly, if \(\theta _i=0\) in ordering \(\widetilde{c}\) and \(\widetilde{d}\), then \({\mathcal {V}}(\mu _{\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{d}})\). Thus, it follows that \({\mathcal {V}}(\mu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\mu _{\widetilde{b}+\widetilde{d}})\), \({\mathcal {V}}(\rho _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\rho _{\widetilde{b}+\widetilde{d}})\) and \({\mathcal {V}}(\nu _{\widetilde{a}+\widetilde{c}})\ne {\mathcal {V}}(\nu _{\widetilde{b}+\widetilde{d}})\). So, \(\theta _i=0\) in ordering \(\widetilde{a}+\widetilde{c}\) and \(\widetilde{b}+\widetilde{d}\).
From these four claims, it is trivial enough to show that if \(\widetilde{a}\succ \widetilde{b}\) and \(\widetilde{b}\succ \widetilde{c}\), then \(\widetilde{a}+\widetilde{c}\succ \widetilde{b}+\widetilde{d}\).
Rights and permissions
About this article
Cite this article
Chutia, R., Smarandache, F. Ranking of single-valued neutrosophic numbers through the index of optimism and its reasonable properties. Artif Intell Rev 55, 1489–1518 (2022). https://doi.org/10.1007/s10462-021-09981-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10462-021-09981-3