TL;DR
This paper introduces new aggregation operators, ILWIOWA and ILGWIOWA, designed to handle complex decision-making scenarios involving "intuitionistic linguistic" (IL) information. These operators are applied to the critical problem of selecting low-carbon suppliers, offering a more flexible and robust approach than existing methods by better incorporating expert uncertainty and preferences.
What Problem Does This Paper Address?
The paper tackles the challenge of selecting suitable low-carbon suppliers, a crucial step in building sustainable supply chains. This selection process is inherently complex, involving multiple, often conflicting, criteria and significant uncertainty. Traditional decision-making methods often struggle with this uncertainty, as they typically require precise numerical data. While intuitionistic fuzzy sets (IFS) offer a way to handle vagueness, decision-makers may find it difficult to provide exact membership and non-membership degrees. This is where "intuitionistic linguistic sets" (ILS) come in, allowing for the use of linguistic terms (like "good," "fair," "poor") combined with intuitionistic fuzzy concepts. However, existing IL-based decision-making methods have limitations in how they aggregate information, particularly in how they utilize "order-inducing variables" which represent expert confidence or other important properties. This paper aims to bridge this gap by developing new aggregation operators that more effectively leverage these order-inducing variables within an IL environment.
Methods
The core of this paper lies in the development of novel aggregation operators for intuitionistic linguistic (IL) information.
- Intuitionistic Linguistic Weighted Induced Ordered Weighted Averaging (ILWIOWA): This operator extends the Weighted Induced Ordered Weighted Averaging (WIOWA) operator to the IL environment. The key innovation is that it not only uses order-inducing variables to determine the order of arguments but also uses them to modify the associated weights. This allows for a more nuanced aggregation that captures variations in expert confidence or other relevant properties.
- Intuitionistic Linguistic Generalized Weighted Induced Ordered Weighted Averaging (ILGWIOWA): This operator is a generalization of ILWIOWA. It introduces a parameter ($\lambda$) that allows for a wider range of aggregation behaviors, effectively encompassing other types of aggregation operators (e.g., geometric, quadratic, harmonic averaging) as special cases. This provides decision-makers with greater flexibility in choosing an aggregation scheme that best suits their preferences.
The paper also outlines a general Multiple Attribute Decision Making (MADM) framework that utilizes these new operators. The steps involve:
- Forming an IL evaluation matrix from expert assessments.
- Calculating order-inducing variables (e.g., confidence levels).
- Aggregating the evaluations for each alternative using the ILGWIOWA operator.
- Ranking the alternatives based on the aggregated results.
Datasets
Not discussed in the paper. The paper uses a hypothetical example for demonstration.
Key Insights and Ideas
- Bridging the Gap Between Linguistic Terms and Uncertainty: The paper effectively combines the intuitive nature of linguistic terms with the mathematical rigor of intuitionistic fuzzy sets through the concept of Intuitionistic Linguistic Sets (ILS). This is crucial for real-world decision-making where precise numerical data is often unavailable or difficult to obtain.
- Leveraging Order-Inducing Variables for Richer Aggregation: The ILWIOWA and ILGWIOWA operators go beyond simply ordering arguments. By using order-inducing variables to modulate weights, they capture subtle nuances in expert confidence or other relevant factors, leading to potentially more robust and informative aggregated results.
- Flexibility Through Generalization: The ILGWIOWA operator's parameter $\lambda$ offers significant flexibility. This allows decision-makers to tailor the aggregation process to their specific risk attitudes or preferences, making the method more adaptable to diverse real-world scenarios.
- Practical Application to a Critical Problem: Applying these novel operators to low-carbon supplier selection demonstrates their practical utility in a domain with significant environmental and economic implications.
Key Results
The paper presents a numerical example for low-carbon supplier selection.
- Using the ILGWIOWA operator (with $\lambda=1$), the most desirable alternative was identified as A4.
- A comparison with the ILIOWA operator (which only uses order-inducing variables for ordering, not weighting) resulted in a different ranking (A2 being most desirable). This highlights the advantage of the ILWIOWA/ILGWIOWA approach in capturing more information.
- The paper also shows how varying the parameter $\lambda$ in the ILGWIOWA operator can lead to different rankings of alternatives, emphasizing the operator's flexibility. For instance, with $\lambda \to 0$ (geometric averaging), A1 became the most desirable alternative, while with $\lambda=2$ or $\lambda=3$, A4 remained the most desirable.
Limitations and Future Work
- Limitations: The paper acknowledges that the order-inducing variables are assumed to represent experts' belief levels. In real-world applications, these variables could represent various properties (e.g., preference, importance, consistency), and the paper's current framework might need adaptation for different interpretations. The paper also notes that the ILIOWA operator, a precursor to their work, has limitations in fully utilizing order-inducing variables.
- Future Work: The authors suggest extending their method by incorporating distance measures and exploring Pythagorean fuzzy sets. They also propose applying their tools to other decision-making problems, such as choice, ranking, or sorting, and considering different decisional problematics. Further research could also explore the impact of different types of order-inducing variables and their interpretations on the aggregation results.
Summary updated