Journal of Fuzzy Sets and Fuzzy Logic Design (e-ISSN: 3049-0227)

Type-2 Fuzzy Logic Systems for Robust Medical Diagnosis Under Deep Clinical Uncertainty: A Comparative Study with Type-1 Models

2026-01-21T10:21:15+00:00

Medical diagnosis as a problem of uncertainty is caused by subjectivity in the description of symptoms, noisy clinical measurements and non-uniformity in the interpretation by experts. The widely studied imprecision has traditionally been modelled using fuzzy logic systems, although conventional Type-1 fuzzy systems assume strictly defined membership functions, which do not accommodate high levels of uncertainty in clinical environments. The paper is indicative of a comparative analysis of Type-1 and Type-2 fuzzy logic framework in medical diagnostic decision support under the presence of profound clinical uncertainty in a systematic fashion. The two models use an integrated rule base with the aim that a clinically motivated diagnostic problem is formulated in a manner that both models are given a fair comparison. It has a Type-2 system, the representation of membership functions in the form of interval-valued representations directly involves uncertainty, providing an explicit representation of clinician disagreement and population heterogeneity. Massive experiments are performed on a carefully designed synthetic data that emulates diagnostic uncertainty within the real world, an example of which includes measurement randomness, missing data, and perturbations to membership. The results have shown that Type-2 fuzzy logic system is more sensitive, specific, and larger AUC than Type-1 counterpart when the uncertainty is larger. The findings have provided practical and theoretical evidence on the application of Type-2 fuzzy logic to provide sound medical diagnosis under deep and protracted uncertainty.

An Investigative Study of Fuzzy Cognitive Maps for Strategic Farm Management

2026-02-06T08:45:56+00:00

Strategic farm management is increasingly challenged by the growing complexity of socio-economic dynamics, environmental uncertainties, climate variability, and the rapid integration of precision agriculture technologies. In this context, effective decision-support tools are essential for managing interconnected agricultural systems. This review article investigates the application of Fuzzy Cognitive Maps (FCMs) as a robust and flexible modeling approach for strategic decision-making in agriculture. FCMs uniquely bridge the gap between qualitative expert knowledge and quantitative computational modeling by representing complex causal relationships among system variables in an intuitive and transparent manner. Through a structured analysis of recent literature, this study identifies key applications of FCMs in areas such as crop yield prediction, sustainability and resource-use assessment, risk and resilience evaluation, and agricultural policy analysis. The findings indicate that FCMs offer superior interpretability and explainability compared to traditional “black-box” machine learning models, making them particularly suitable for participatory farm management and stakeholder-driven planning. Nevertheless, challenges remain, including sensitivity to weight assignment, scalability, and the need for systematic methods to achieve expert consensus. Overall, this review highlights current advances in FCM-based agricultural research and outlines future directions for developing more resilient, adaptive, and data-informed farm management systems.

Fuzzy Logic-Based Medical Diagnosis System for Handling Uncertainty in Healthcare Data

2026-04-17T05:23:16+00:00

Medical diagnosis has always been characterised by its inherent uncertainty due to the lack of precision associated with various symptoms, the differences that exist between patients, and the incompleteness of clinical information available. Traditional decision-making models based on crisp/logical reasoning do not adequately account for such a large amount of uncertainty. In contrast to the use of traditional decision-making models, fuzzy models represent the best method for dealing with vagueness by incorporating linguistic variables (e.g., 'very low' or 'moderate', etc.) and approximate reasoning. This paper focuses on a specific fuzzy logic-based system designed for use within a medical diagnosis application. The system uses fuzzy logic inference mechanisms to map patient inputs (e.g., symptom data, medical history) to the level of severity of the disease (e.g., mild, moderate, severe), and is comprised of the following components: fuzzification; construction of a rule base; model for inference; and defuzzification. The mathematical development, modeling of membership functions, and analysis of this system are presented in detail. The findings from experimentation suggest that the accuracy, interpretability, and robustness of this fuzzy logic-based system are much greater than those of more traditional methods. Therefore, the results of this work point to the efficacy of fuzzy models in supporting real-world healthcare decision-making systems.

Smart Traffic Signal Control System Based on Fuzzy Logic Control

2026-04-17T06:09:55+00:00

Traffic congestion at urban intersections is rapidly increasing due to the continuous growth in vehicle population and urbanization, creating serious challenges for existing traffic management systems. Conventional traffic signal controllers generally operate on fixed-time cycles that do not consider real-time traffic conditions, resulting in inefficient traffic flow, increased vehicle waiting time, unnecessary fuel consumption, and higher environmental pollution. To address these limitations, this study proposes a fuzzy-logic-based adaptive traffic signal control system that dynamically adjusts signal timing based on multiple traffic parameters, including queue length, time of day, night traffic conditions, and the presence of emergency vehicles. The system also integrates a no-vehicle detection mechanism, which monitors vehicle activity at an intersection and automatically switches the signal after 10 seconds of inactivity, thereby eliminating unnecessary red-light waiting when no vehicles are present. A fuzzy inference system (FIS) is designed and simulated using the MATLAB Fuzzy Logic Toolbox to model human-like decision-making for signal control under uncertain and varying traffic conditions. The proposed system enhances flexibility, responsiveness, and efficiency compared to traditional fixed-time approaches. Simulation results demonstrate a significant improvement in traffic flow efficiency, reduction in average waiting time, and better utilization of intersection capacity, making the system suitable for implementation in intelligent transportation systems and smart city environments.

A Quantum-inspired Fuzzy Soft Computing and Machine Learning Framework for Adaptive and Intelligent Decision Making under Uncertainty

2026-04-17T06:43:27+00:00

The growing complexity and uncertainty of real-world decision-making systems require sophisticated computational systems that surpass the constraints of classical artificial intelligence. As a new quantum-inspired soft computing and machine learning architecture, this study proposes a novel quantum-inspired next-generation intelligent decision-making based on the principles of fuzzy logic, neural networks, evolutionary computation, and quantum-inspired optimization methods. The suggested framework leverages the concepts of probabilistic superposition and parallelism in quantum computing, but does not require quantum hardware to make learning more efficient, explore the solution space, and be resilient to uncertainty. The architecture is a combination of fuzzy inference systems to deal with imprecision, machine learning models to deal with adaptive pattern recognition and quantum-inspired algorithms to deal with global optimization and feature selection. The framework also includes interpretable AI processes so that the decisions made by the framework can be transparent and explainable. Using experimental analysis of various fields, such as healthcare, diagnostic and financial risk analysis and smart city management, accuracy, convergence, and reliability in decision making are found to be better than conventional soft computing and machine learning methods. The findings present possibilities of quantum-inspired hybrid intelligent systems in dealing with complex, uncertain and dynamic environments. The study presents a single paradigm of combining quantum-inspired computation with soft computing and machine learning, which will enable the realization of scalable, interpretable, and high-performance intelligent decision-making systems in future applications of AI.