TL;DR
This paper introduces a novel framework for multi-criteria group decision-making (MCGDM) problems using simplified neutrosophic sets (SNSs). By defining an "interval dependent degree," the approach effectively captures the inherent uncertainty and distribution of information within SNSs, leading to more robust and stable decision outcomes.
What Problem Does This Paper Address?
In complex decision-making scenarios, especially those involving multiple criteria and decision-makers (MCGDM), accurately representing and processing uncertain information is crucial. Traditional fuzzy sets and even intuitionistic fuzzy sets struggle to fully capture situations where information is incomplete, inconsistent, or where the degree of "not sure" is significant. Simplified Neutrosophic Sets (SNSs) offer a more expressive way to handle this uncertainty by including truth, indeterminacy, and falsity membership degrees. However, existing methods often treat these degrees as fixed values, potentially losing valuable information about their distribution and interdependencies. This paper addresses the need for a method that can not only handle the multi-dimensional uncertainty of SNSs but also account for the implied distribution of these values, leading to more reliable and stable decision-making.
Methods
The core contribution of this paper is the development of an "interval dependent degree" for simplified neutrosophic numbers (SNNs). This is achieved through a multi-step process:
- Interval Transformation Operator: An operator is defined to transform an SNN (represented by truth, indeterminacy, and falsity membership degrees) into an interval. This interval captures the range of possible truth-membership values, effectively converting the SNN into a more manageable interval representation.
- Simplified Dependent Function: This function measures how a specific value relates to an interval, with a maximum at the optimal value and a minimum at the worst. Various forms of this function are proposed to reflect different decision-maker psychologies (e.g., risk aversion).
- Distribution Function: This function describes how values are distributed within an interval, modeled as a probability density function. Different distributions (uniform, triangular, normal, etc.) can be used to represent various scenarios.
- Interval Dependent Function: By integrating the simplified dependent function and a chosen distribution function, the paper defines the interval dependent function. This function quantifies the "dependent degree" of an SNN within the standard interval [0,1], considering both the value's position and its distribution.
- MCGDM Framework: A step-by-step procedure is outlined for applying this interval dependent degree to MCGDM problems. This includes normalizing the decision matrix, performing the interval transformation, selecting appropriate dependent and distribution functions, calculating individual SNN dependent degrees, aggregating these into a comprehensive dependent degree for each alternative, and finally, ranking the alternatives.
Datasets
Not discussed in the paper. The paper uses a hypothetical example adapted from a previous study to illustrate its methodology.
Key Insights and Ideas
- Beyond Fixed Values: The paper moves beyond treating SNN membership degrees as fixed points, recognizing that their distribution within an interval carries crucial information. This is a significant step in handling uncertainty more comprehensively.
- Psychology-Aware Decision Making: The introduction of different forms for the simplified dependent function allows the model to implicitly capture decision-maker preferences and psychological biases (like risk aversion), making the decision process more nuanced.
- Stability and Uncertainty Analysis: By allowing for the selection of various distribution functions, the framework inherently supports uncertainty and stability analysis. This means decision-makers can assess how sensitive the final ranking is to different assumptions about data distribution.
- Concise and Integrated Approach: The proposed method integrates multiple concepts (interval transformation, dependent function, distribution function) into a unified framework, reducing the complexity and number of intermediate steps often found in other MCGDM methods.
Key Results
The paper demonstrates the feasibility and effectiveness of its proposed method through an illustrative example. The key findings include:
- Consistent Ranking: Despite variations in the simplified dependent function (linear vs. nonlinear with different parameters) and distribution functions (uniform, normal, triangular, etc.), the final ranking of alternatives remained consistent. This highlights the robustness and stability of the proposed approach.
- Comparative Advantage: A comparison with existing methods showed that the proposed approach achieves the same sorting results while offering advantages in terms of maintaining information integrity, computational simplicity, and providing uncertainty and stability analysis capabilities.
Limitations and Future Work
- Limitations: While the paper demonstrates robustness across different distribution functions, it acknowledges that the framework's applicability to "complete uncertainty situations" might require further investigation.
- Future Work: The authors suggest extending the proposed method to other types of neutrosophic sets, such as interval neutrosophic sets, multi-valued neutrosophic sets, and complex neutrosophic sets. They also propose exploring applications in non-traditional areas like game theory and investigating its non-linear dynamics description capability.
Summary updated