Search

• The orness function and the normal/inverse normal distribution shape parameter are not monotonically correlated. • The orness function and the exponential/inverse exponential distribution shape parameter are monotonically correlated. • An enhanced weight generation approach for ordered weighted averaging operators is proposed. In the present paper, we demonstrate that the degree of orness, which measures the attitudes of decision-makers, does not decrease strictly with the fractile values when either the normal probability density function or its inverse form is used to generate the weights of ordered weighted averaging operators. As for the weights of ordered weighted averaging operators generated from either the exponential distribution and its inverse form, we prove the strict monotonicity of the orness function with respect to the distribution shape parameter. To solve the drawbacks of the probability-density-function-based weight generation approach, the present paper uses the interweaving method to adjust the probability-density-function-based weighting vector of an ordered weighted averaging operator based on the premise that the distribution shape parameter is excluded. This enhanced approach retains the preferences of decision-makers to the utmost when they change their attitudes toward objects. Finally, the feasibility and effectiveness of this novel paradigm for generating the weights of ordered weighted averaging operators are demonstrated with its promising application in a system for aggregating movie ratings. [ABSTRACT FROM AUTHOR]

In this paper, we point out that two theorems in a previous paper by Wei et al. (2013) are incorrect by some counterexamples in detail, and present the modified theorems. [ABSTRACT FROM AUTHOR]

A great number of real world problems can be associated with multicriteria decision making. These problems are often characterized by a high degree of uncertainty. Interval-valued intuitionistic fuzzy sets are a generalized form of an ordinal fuzzy set to deal with this natural uncertainty. In this paper, an enhanced version of the ELECTRE method, called ELECTRE III, is extended under the interval-valued intuitionistic fuzzy environment. The advantages and strengths of ELECTRE III as a decision aid technique and interval-valued intuitionistic fuzzy sets as an uncertain framework make the proposed method a suitable choice in solving practical problems. The application of the proposed method is illustrated by the solution of an investment project selection problem. [ABSTRACT FROM AUTHOR]

Environmental Impact Significance Assessment (EISA) is usually modeled as a multi-criteria decision making process for determining the importance of project’s impacts over involved environment, considering subjective judgments provided in a qualitative and/or quantitative way. Classical EISA methods are not efficient in handling heterogeneous contexts since experts are forced to use numerical scales even for assessing subjective environmental indicators and they also obtain numerical outputs of low interpretability. In this paper is proposed a new approach for heterogeneous EISA based on the linguistic 2-tuple fusion model for dealing with heterogeneous information. It provides a flexible evaluation framework in which experts can supply their preferences using different information domains conform to the nature and uncertainty of criteria as well as their level of knowledge and experience. Moreover the approach applies a multi-step aggregation process to obtain interpretable significance values without loss of information. [ABSTRACT FROM AUTHOR]

In the complex multi-attribute large-group decision-making (CMALGDM) problems in interval-valued intuitionistic fuzzy (IVIF) environment, attributes of the alternatives are often stratified and correlated. This paper proposes a decision-making method for these problems based on partial least squares (PLS) path modelling, which not only fully exploits the decision information of decision makers (DMs), but also effectively addresses the relativity problem in the decision attributes and objectively assigned weights to the primary decision attributes (i.e., "latent variables for decision making"). The method can be outlined in three steps. First, a two-stage method is proposed to transform the interval-valued intuitionistic fuzzy number (IVIFN) samples into single-valued samples. In this step, an improved C-OWA operator is first given to transform the IVIFN samples into intuitionistic fuzzy number (IFN) samples, which makes the preference information of the DMs more objectively aggregated. Then a proposed membership-based method is applied to reduce the information loss and transform the IFN samples into single-valued samples. Second, the estimated values and weights of the "latent variables for decision-making" are obtained by means of the PLS path modelling algorithm. Finally, a multi-alternative sorting method is devised in accordance with the estimated values and weights. An example is provided to illustrate the proposed technique and evaluate its feasibility and validity. [ABSTRACT FROM AUTHOR]