[Paper Review] Fuzzy Rankings: Properties and Applications

Jiří Mazurek|arXiv (Cornell University)|Mar 15, 2017

Multi-Criteria Decision Making15 references3 citations

TL;DR

This paper introduces fuzzy rankings as a generalization of crisp rankings to handle uncertainty in decision-making, where objects may be ranked with degrees of preference or equality. It proposes a formal framework for fuzzy rankings, analyzing properties like ordering, similarity, and indecisiveness, and demonstrates applications in group and multiple criteria decision making under uncertainty.

ABSTRACT

In practice, a ranking of objects with respect to given set of criteria is of considerable importance. However, due to lack of knowledge, information of time pressure, decision makers might not be able to provide a (crisp) ranking of objects from the top to the bottom. Instead, some objects might be ranked equally, or better than other objects only to some degree. In such cases, a generalization of crisp rankings to fuzzy rankings can be more useful. The aim of the article is to introduce the notion of a fuzzy ranking and to discuss its several properties, namely orderings, similarity and indecisiveness. The proposed approach can be used both for group decision making or multiple criteria decision making when uncertainty is involved.

Motivation & Objective

• Address the limitation of crisp rankings in real-world scenarios where decision-makers face uncertainty, time pressure, or incomplete information.
• Introduce a formal model of fuzzy rankings to represent partial or imprecise preferences among objects.
• Analyze key properties of fuzzy rankings, including ordering, similarity, and indecisiveness, to support robust decision analysis.
• Enable practical applications in group decision making and multiple criteria decision making under uncertainty.
• Provide a theoretical foundation for handling uncertainty in preference structures beyond traditional crisp rankings.

Proposed method

• Define fuzzy rankings as a generalization of crisp rankings using fuzzy sets to represent degrees of preference between objects.
• Formalize the concept of a fuzzy ranking as a fuzzy relation satisfying specific transitivity and reflexivity conditions.
• Introduce measures for similarity between fuzzy rankings to compare different decision outcomes.
• Propose an indecisiveness measure to quantify the level of ambiguity or uncertainty in a fuzzy ranking.
• Apply the framework to group decision making by aggregating individual fuzzy rankings into a collective fuzzy ranking.
• Use illustrative examples and figures to demonstrate the behavior and properties of fuzzy rankings in practical settings.

Experimental results

Research questions

• RQ1 How can crisp rankings be generalized to handle uncertainty and partial preferences in decision-making contexts?
• RQ2 What are the fundamental properties—such as ordering, similarity, and indecisiveness—that characterize fuzzy rankings?
• RQ3 How can fuzzy rankings be effectively used in group decision-making scenarios with conflicting or uncertain inputs?
• RQ4 What are the mathematical and computational properties that ensure consistency and interpretability of fuzzy rankings?
• RQ5 In what ways do fuzzy rankings improve upon traditional crisp rankings in multiple criteria decision making?

Key findings

• Fuzzy rankings provide a more realistic representation of preferences when decision-makers are uncertain or cannot rank all objects precisely.
• The proposed framework successfully captures partial orderings and equal rankings through fuzzy membership values.
• Similarity measures between fuzzy rankings allow for meaningful comparison and aggregation in group decision settings.
• The indecisiveness measure quantifies uncertainty levels, enabling decision-makers to assess the reliability of rankings.
• The approach is applicable to both individual and group decision-making processes under uncertainty.
• Theoretical analysis and illustrative examples confirm the consistency and interpretability of the fuzzy ranking model.