<link rel="stylesheet" href="styles.cf390c6bfbe7555f.css">

Publication:
Advanced fuzzy set extensions and novel decision making models

Loading...
Thumbnail Image

Date

Institution Authors

Journal Title

Journal ISSN

Volume Title

Publisher

ITU Graduate School

Research Projects

Organizational Units

Journal Issue

Abstract

Uncertainty is an inherent part of almost every decision-making process, especially in complex systems where information is incomplete, vague, or imprecise. Traditional mathematical models, which are largely based on binary logic and crisp data, often fall short in representing the subtleties and ambiguities that characterize real-world situations. In response to these limitations, Zadeh's introduction of fuzzy set theory in 1965 marked a foundational shift in how imprecision and vagueness are mathematically formalized. Fuzzy sets, by allowing partial membership, opened up new possibilities for modeling uncertainty in a more flexible and human-like manner. Since then, numerous extensions of fuzzy sets have been proposed, such as intuitionistic fuzzy sets, Pythagorean fuzzy sets, and q-rung orthopair fuzzy sets to better handle the increasing complexity and multidimensionality of uncertainty. These extensions offer more expressive frameworks that capture hesitation, disagreement, and detailed preference structures more effectively than classical models. However, despite these advancements, significant challenges remain. Existing models often fail to adequately address the levels of uncertainty and the nature of complex judgments. Moreover, the development of decision-making methods that are fully compatible with the latest fuzzy extensions is still limited, leaving a gap between theoretical progress and practical application. Conventional fuzzy set frameworks, while effective in many contexts, often fall short in addressing continuous uncertainty, decomposability, or the increasing complexity of preference representation in MCDM tasks. This study presents a comprehensive and original contribution to the field of decision-making under uncertainty by developing novel fuzzy set extensions and proposing innovative multi-criteria decision-making (MCDM) methodologies. Responding to these methodological and conceptual gaps, this thesis proposes a series of novel fuzzy set extensions alongside new or adapted decision-making models, aiming to enhance both theoretical modeling and practical decision support. Motivated by the inherent vagueness, ambiguity, and complexity of real-world decision environments, the thesis handles key limitations in the representation and analysis of uncertainty, both from a theoretical and an applied perspective. The first chapter introduces the Continuous Intuitionistic Fuzzy Sets (CINFUSs), an innovative extension of Atanassov's intuitionistic fuzzy sets. Unlike traditional models with discrete or static membership assignments, CINFUSs allow the membership and non-membership degrees to vary continuously within the intuitionistic constraint. This continuous formulation significantly enriches the expressive capacity of the fuzzy modeling framework by better accommodating the fluid nature of human judgment and uncertainty. In addition to offering a mathematically grounded structure, CINFUSs open the door to more flexible aggregation, comparison, and distance measures, which are critical in fine-tuned decision analysis. Building upon this conceptual advancement, the Chapter 2 proposes the Decomposed Pythagorean Fuzzy Sets (DPFSs), which introduce a structural decomposition within the Pythagorean fuzzy framework. The DPFS model separates membership and non-membership components into optimistic and pessimistic to allow independent treatment and interpretation of each dimension. This decomposition enables more granular modeling of decision-makers' preferences and perceptions of uncertainty, particularly valuable in settings that require explanation, transparency, or sensitivity analysis. By disentangling the contributing factors of vagueness, the model not only facilitates interpretability but also improves the quality of aggregation and ranking mechanisms in MCDM processes. The Chapter 3 presents the Continuous Pythagorean Fuzzy Sets (CPFUSs), which synthesizes the core idea of continuity (as introduced in CINFUSs) with the geometric and tolerant properties of the Pythagorean fuzzy set. The CPFUS framework is designed to reflect smooth variations in uncertainty while maintaining the mathematical relationship required by the Pythagorean condition. As a result, CPFUSs serve as a robust modeling environment for contexts that demand both expressive precision and a higher tolerance for hesitancy and ambiguity. This thesis enhances the theoretical diversity of fuzzy sets and offers a novel alternative to existing models in complex decision-making scenarios. Focusing on methodological development from theoretical modelling, the Chapter 4 introduces a new MCDM method named IF-DEVADA (Intuitionistic Fuzzy DEcision-making based on VAlue and Distance Analysis). This model is developed to address the limitations of static and single-perspective decision-making frameworks. IF-DEVADA integrates multiple distance-based evaluations under intuitionistic fuzzy conditions and accounts for dynamic and heterogeneous decision environments. It is particularly suitable for problems involving evolving preferences, multiple time periods, and group-based decisions. The method is validated through real-world applications that demonstrate its adaptability and analytical depth, marking it as a significant contribution to dynamic MCDM research. The Chapter 5 applies and adapts the well-established TOPSIS method within the context of q-rung orthopair fuzzy sets (q-ROFS), one of the most flexible and recent extensions of fuzzy sets. The proposed q-ROFS-TOPSIS model extends traditional MCDM analysis by accommodating a higher capacity for uncertainty and contradiction in the input data, making it suitable for complex systems with incomplete or inconsistent information. By integrating q-ROFS into a classical yet powerful decision-making model, this study not only demonstrates the versatility of q-rung environments but also enhances the practical applicability of fuzzy sets in modern MCDM frameworks. In its entirety, the thesis does not merely pursue technical novelty but strives to build a cohesive and future-oriented framework for modeling and managing uncertainty. Each chapter contributes a unique dimension, ranging from mathematical formulation to decision model construction, toward advancing the capabilities of both fuzzy set theory and decision-making science. The theoretical developments enhance how uncertainty is expressed, decomposed, and interpreted, while the methodological novelty provides robust tools for practical analysis across diverse application domains, including sustainability, policy prioritization, and strategic planning. By combining theoretical rigor with real-world applicability, this dissertation not only addresses current limitations in the literature but also lays the groundwork for future research in fuzzy decision-making. Together, the five studies form a consistent and forward-looking framework that enhances how uncertainty is represented, interpreted, and used in analytical decision-making. The proposed models are designed to be extensible and adaptable, encouraging further exploration across areas where there is uncertainty, complexity, and multiple criteria.

Description

Thesis (Ph.D.) -- Istanbul Technical University, Graduate School, 2025

Journal or Series

ISSN

ISBN

Rights

Keywords

endüstri mühendisliği, industrial engineering, cok ölçütlü karar verme, multiple criteria decision making, fuzzy karar verme, fuzzy decision making, fuzzy kümeler, fuzzy sets

Citation

Collections

Endorsement

Review

Supplemented By

Referenced By

Related Patent

Related Goal

Google Scholar
Scholar'da Ara ↗
Bu yayında DOI yok — Altmetric/Dimensions/PlumX/BIP! rozetleri DOI gerektirir.