Unmanned Underwater Vehicles (UUVs) are increasingly important in coastal ecological protection and marine resource utilization, where autonomous obstacle-avoidance trajectory planning is essential to ensure reliable operation. To handle the complex and dynamic nearshore environments, this study develops an enhanced Dynamic Window Approach (DWA), incorporating fuzzy logic reasoning. A rapid angular velocity variation evaluation function is introduced to improve the smoothness of planned trajectories, while a safety coefficient combined with a fuzzy inference-based dynamic weight adjustment strategy enables adaptive parameter allocation in response to varying obstacle distributions. These improvements elevate the environmental adaptability of UUVs and reduce the occurrence of local-optimal solutions. Extensive MATLAB simulation trials conducted under diverse dynamic scenarios verify that the method provides strong reliability and performance stability. Furthermore, results obtained from realistic underwater experiments show that the refined algorithm yields superior obstacle avoidance success, longer safe-navigation distances, smoother planned paths, and improved energy utilization when compared with the conventional DWA framework.

This article deals with the novel approach of solving a village supply chain model (hut inventory) studied in all rural areas. First of all, we develop a crisp mathematical model considering a parabolic demand function in the entire inventory process during the day time of approximately 10 h. At the peak of the business hours, the demand of the commodities becomes high and then it began to fall down and finally becomes zero, if some of the excess amount exist after the end of the day then it would be used for lost sale/exhaust under special rebate. The objective of the research is to find the actual optimal order quantity, optimal unit selling price and the optimal cycle time so as to maximise the overall system profit as well as the reduction of system cost. The unit selling price is set through the negotiation of the purchaser and the seller followed by a bargaining function. The demand rate of the customers and the purchasing price of the selling items by the retailer (seller) per unit item are assumed to the fuzzy stochastic variables. To solve the model, the concept of duality has been developed based on α -cuts and its dual β -cuts of fuzzy numbers. The numerical result shows that the discount model has the capability to achieve the higher per capita profit return 58.42% (cycle time 6.76 h, order quantity 251.19 kg) for lower investment of the retailer, but it attains maximum 65.94 % (cycle time 7.95 h, order quantity 312.96 kg) per capita profit return with respect to the all-items exhaust and that for partial item’s sale the per capita profit return reaches to 43.30 % (cycle time 10.50 h, order quantity 431.46 kg) all the time. However, the comparative analysis and graphical illustrations have been discussed to show the actual novelty of the proposed approach.

In order to reduce interference of the complexity and uncertainty of the operating environment on the electric energy metering box in accurately perceiving its operating status, this study proposes an intelligent method for accurate perception of the operating status of the electric energy metering box based on fuzzy reasoning. First, establish an indicator system that covers the factors influencing the operation status of intelligent energy metering boxes and collect indicator data. Then, establish an adaptive network based fuzzy inference system (ANFIS). Using indicator data as input information, the input data are fuzzified using a membership function, defined based on minimum ambiguity and transformed into a fuzzy set. Afterward, using the fuzzy IF-THEN rule, the fuzzy set of the input data is mapped to the fuzzy set of the output result. With the help of the five-layer structure of the ANFIS network, precise perception of the operating status of intelligent energy metering boxes is achieved through steps such as fuzzy processing, rule inference, sequential normalization, defuzzification, and output calculation. The experimental results show that this method can obtain relevant data based on the constructed indicator system and accurately perceive different operating states. Compared with existing methods, this method has a lower fluctuation rate of state perception delay, better adaptability, and can achieve accurate perception under different load conditions.

Most automobile repair and maintenance apps utilize the rule base and case base reasoning methodologies in their implementations. These two methodologies have their strengths and limitations, some of these limitations can be overcome by the Convolutional Neural Networks (CNN), a specialized subset of deep learning that has excelled in image analysis due to its ability to learn hierarchical representations (Brito, 2023). CNN extracts the pixel value of an image and create a feature map that it uses for processing through learning. In this study, 998 images of frequently occurring vehicle faults were captured from mechanic workshops operated within and by Akwa Ibom state Transport company. These images were subjected to 3 lightweight CNN models after pre-processing. The aim was to detect and classify faults in these damaged parts. Results obtained show that the 3 models demonstrated strong performance across the three key evaluation metrics: accuracy, precision, and recall with an accuracy of 92%, Precision of 91% and Recall of 90%. It is recommended that an integration of case-based reasoning, Fuzzy logic reasoning and CNN be undertaken to improve on the results and a high resolution camera be used to capture the images of the damaged parts for a better input to the CNN model.

A multi-objective genetic–Mamdani fuzzy reasoning framework for bias-resilient small business credit scoring

Unmanned Aerial Vehicle (UAV) swarms operating in contested environments face a critical “semantic gap” between raw, high-velocity network traffic and high-level mission security constraints, compounded by the risk of privacy leakage during collaborative learning. Existing deep learning (DL)-based Network Intrusion Detection Systems (NIDSs) suffer from opacity, prohibitive resource consumption, and vulnerability to gradient leakage attacks in federated settings, while traditional rule-based systems fail to handle encrypted payloads and evolving attack patterns. To bridge this gap, we present NeSySwarm-IDS (Neuro-Symbolic Swarm Intrusion Detection System), an end-to-end differentiable neuro-symbolic framework that simultaneously achieves high accuracy, strong privacy guarantees, and built-in interpretability under resource constraints. NeSySwarm-IDS integrates an extremely lightweight 1D convolutional neural network with a differentiable Łukasiewicz fuzzy logic reasoner incorporating attack-specific rules. By aggregating only low-dimensional logic rule weights with calibrated differential privacy noise, we drastically reduce communication overhead while providing (ϵ,δ)-DP guarantees with negligible utility loss. Extensive experiments on the UAV-NIDD dataset and our self-collected dataset demonstrate that NeSySwarm-IDS achieves near-perfect detection accuracy, significantly outperforming traditional machine learning baselines despite using limited training data. A detailed case study on GPS spoofing confirms the interpretability of our approach, providing axiomatic explanations suitable for autonomous mission verification. These results establish that end-to-end neuro-symbolic learning can effectively bridge the semantic gap in UAV swarm security while ensuring privacy and interpretability, offering a practical pathway for deploying trustworthy AI in contested environments.

The booming development of online education has made online classrooms an important component of the education field. In-depth analysis of students' learning behavior in online teaching can help teachers optimize teaching strategies and provide personalized learning support for students. Therefore, to carry out an in-depth analysis of students' learning behavior, this study collects data from online teaching platforms and preprocesses it. Subsequently, this study constructs a multi-perspective fuzzy reasoning model covering three dimensions: curriculum, individual, and class, to comprehensively consider students' learning performance from different levels. This model processes uncertain information in learning behavior data through fuzzy sets and fuzzy rules, achieving a multidimensional evaluation of learning performance. An improved XGBoost algorithm is designed to classify students' comment emotions. This improved algorithm optimizes the hyperparameters of the XGBoost algorithm by improving the grey wolf optimization algorithm. The algorithm enhances the accuracy of emotion classification and further explores the emotional tendencies and attitude feedback behind their learning behavior. The results showed that from a curriculum perspective, the completion rate of course tasks 3 weeks before the exam was basically above 45%, which was much higher than the completion rate three weeks after the task was released (both lower than 18%). These results indicated that students were more inclined to complete tasks before the deadline and had obvious procrastination. The maximum accuracy of the improved classification algorithm was 98.78%, which was 8.57%, 7.55%, 6.38%, and 6.01% higher than the comparison model, and its average time consumption was 58 ms. The recall rates on negative, positive, and neutral emotions were 98.35%, 97.69%, and 98.02%. The research model can effectively analyze students' online learning behavior and enable early identification of at-risk students, facilitating personalized teaching and precise intervention in online education.

Abstract This study introduces a parallel hybrid control architecture that integrates an interval type-2 fuzzy system (IT2FS) with a recurrent nonlinear autoregressive network with exogenous inputs (NARX) to enhance the line-following capability of autonomous guided vehicles (AGVs) under uncertainty and operational disturbances. In the proposed framework, the IT2FS explicitly models and compensates for system uncertainties, while the recurrent NARX structure contributes dynamic learning and improved temporal representation. Unlike conventional series fuzzy-neural hybrids, the proposed parallel fusion executes uncertainty-aware fuzzy reasoning and recurrent dynamic compensation concurrently, enabling robust real-time AGV control on a PLC platform. The parallel combination of the two components increases control flexibility, enabling faster adaptation and more stable behavior as operating conditions vary. The controller is developed, trained, and evaluated in a real-time environment, demonstrating accurate trajectory tracking and rapid responses to path variations. The results highlight the potential of type-2 fuzzy-NARX hybrid architectures for intelligent mobile-robot control, particularly in industrial autonomous navigation tasks.

Abstract Landfill leachate poses a persistent threat to water quality due to its complex and fluctuating composition. However, conventional monitoring methods often fail to capture rapid changes that can lead to environmental and regulatory risks. This study introduces a field-ready Internet of Things (IoT) system that integrates industrial-grade probes, a Raspberry Pi edge gateway, and a cloud-based backend to provide continuous and traceable leachate monitoring. A fuzzy inference engine anchored on regulatory standards and site-specific statistics translates multi-parameter measurements into Normal, Warning, and Critical risk levels. To enhance reliability, a dual-stage hysteresis mechanism stabilizes alarm states by combining asymmetric value thresholds with a short persistence window, thereby reducing false toggling under noisy conditions. Evaluation during a 21-day landfill pilot demonstrated 100% data delivery, a median latency of 1.00 s, availability of 99.24%, and rapid recovery with queued data replayed within seconds after induced outages. Risk labels generated at the edge and recomputed at the server matched for all samples, ensuring both timely alerts and auditability. The results confirm that combining resilient IoT telemetry with interpretable fuzzy risk reasoning can provide practical decision support for landfill operators while sustaining compliance reporting under intermittent connectivity.

Due to rapid technological advances, evolving legal regulations, and economic uncertainty, organizations today face increasing challenges in managing financial risks. It is common for traditional techniques to struggle to handle vast amounts of real-time data, respond to changing market conditions, and manage the inherent unpredictability of financial data. The purpose of this study is to overcome these obstacles by (i) creating a hybrid financial risk management framework that combines artificial intelligence (AI) with fuzzy logic, (ii) increasing the accuracy and robustness of financial risk predictions through the use of the Random Forest (RF) algorithm, and (iii) increasing the interpretability and transparency of decision-making through the use of fuzzy reasoning. By utilizing random forests, the proposed Adaptive Financial Risk Management System (AFRMS) constructs numerous decision trees and employs a voting mechanism to achieve robust risk classification. The incorporation of fuzzy logic enables the management of ambiguity and facilitates human-like reasoning, leading to risk assessments that are both accurate and interpretable. The system has been proven to deliver excellent performance in recognizing potential hazards and providing advanced decision support through extensive experimental validation on large financial datasets. As a result, risk management is facilitated by better-informed decision-making and stakeholder confidence is increased.

The accuracy and diversity of information flows have become major characteristics of the modern age Big Data. it may not be noted that information about people, organizations, and events that are deemed critical bottlenecks within the deployment of trustworthy machine learning solutions. Despite the exponential nature that the field of data has followed, the quality of this kind of data is often low, particularly represented by extreme cases of Class Imbalance and Noise. Conventional hard computing paradigm based on Aristotelian logic and sharp decision boundaries, often present catastrophicFailure Modes when faced with so many data pathologies. Generally, traditional classifiers are prone to making biased predictions towards the majority class and at the same time overfitting to noisy cases that contaminate the manifold of the feature set. This research paper offers a comprehensive, expert-level examination of the methodologies of Soft Computing (SC)—including Fuzzy Logic, Artificial Neural Networks, and Evolutionary Algorithms—as a solid substitute in dealing with the uncertainty of real-world data. We will critically analyze the theoretical foundations of SC, explaining how the tolerance for imprecision and incomplete truth makes possible the design of decision boundaries robust to overlapping class distributions and the presence of noise within the labels. Moreover, we will introduce a novel Hybrid Soft Computing Framework, namely the REF-DB (Robust Evolutionary-Fuzzy Data Balancing), which jointly exploits the ability to efficiently handle label noise of Fuzzy Logic Filtering and the global optimization potential of Evolutionary Under sampling. This paper will review state-of-the-art techniques operating at multiple tiers: from FSVM, Genetic Fuzzy Systems, to the latest trends that involve LLMs for semantic data augmentation and Quantum Soft Computing for high-dimensional feature mapping. Our work presents a thorough comparison experimental evidence drawn from several benchmark datasets, such as KEEL and UCI. We show through a proper comparative analysis that hybrid SC approaches guarantee maximum values of robust metrics, like the G-Mean and AUC, outperforming hard computing baselines formed by singular approaches. The study concludes with a forward-looking discussion on the integration of Neuro-Symbolic AI and Quantum Machine Learning, positing that the future of robust data analytics lies in the fusion of evolutionary adaptability and fuzzy reasoning.

Online fault diagnosis is crucial for improving the reliability of safety–critical industrial systems. Fault diagnosis becomes increasingly complex if the time-variations, fuzziness, and uncertain causal relationships related to the various internal factors in the system are present. In view of the time-varying working conditions of industrial systems, as well as the fuzzy uncertainty of fault data and knowledge, a C-DUCG (Cubic dynamic uncertain causality graph) approach is proposed for fault diagnosis. Such a solution is intended to help develop a generative cubic causality graph modeling scheme and a FUTURE (fuzzy temporal causality reasoning) algorithm, both of which facilitate the representation and reasoning about complex fault situations (involving temporal causalities and uncertain evidence). In C-DUCG, the strategies of causality simplification and EELA (event-oriented early logical absorption) are proposed to mitigate the complexities of modeling and reasoning. Comparative experiments and a sequence of fault diagnosis tests on a nuclear power plant (NPP) simulator validate the efficiency, recall, accuracy, and interpretability of C-DUCG in large-scale dynamic systems. Experiments reveal that the proposed algorithm achieves 0.62 and 0.999 in terms of recall rate AC@1 and AC@k, respectively, with the average reasoning time being 10 ms, and the average time spent in NPP fault diagnosis tests is 9.42 ms.

Introduction The optimization of machining process decision-making remains a major challenge in intelligent manufacturing due to the uncertainty of process information, incompleteness of rule bases, and the tendency of traditional algorithms to converge to local optima. Therefore, enhancing the adaptability and robustness of decision-making systems is a crucial task for achieving efficient and reliable computer numerical control (CNC) process planning. Methods This study proposes a hybrid decision-making approach that integrates fuzzy theory with support vector machines (SVM) to address uncertainty and incomplete knowledge representation in CNC turning. An Analytic Hierarchy Process (AHP) is used to determine the relative importance of influencing factors, and trapezoidal membership functions are designed to determine the credibility of fuzzy reasoning rules. When the credibility value falls below a defined threshold, a linear-kernel SVM model is activated to provide alternative decisions which formed a fuzzy-SVM collaborative reasoning mechanism. Results Experimental validation demonstrates that the proposed hybrid fuzzy-SVM collaborative method achieves remarkable classification accuracy on the test dataset. The system maintains stable performance even under low-credibility or incomplete rule conditions. The SVM module effectively compensates for the limitations of the fuzzy reasoning process, thereby improving the robustness of decision inference compared to single-model approaches. Conclusion The proposed fuzzy-SVM collaborative reasoning framework enhances the adaptability, stability, and interpretability of CNC machining process decision-making. These findings offer a practical and scalable solution for intelligent process planning in complex and uncertain manufacturing environments.

Kinship verification aims to determine whether two individuals share a familial relationship based on facial information. Cross-gender relationships (i.e., Father-Daughter and Mother-Son) continue to face formidable challenges due to the diversity and uncertainty of genetic inheritance. Existing studies primarily focus on extracting robust features and measuring similarity, with limited attention given to the fuzziness of gender differences. To address this issue, this paper proposes a kinship verification framework based on a fuzzy neural network, which adaptively extracts gender-independent kinship features and handles relationship fuzziness to improve cross-gender verification performance. Specifically, the Swin Transformer, which has demonstrated excellent performance in facial analysis, is employed to extract initial features. A fuzzy neural network is then designed to disentangle gender and kinship features, with a gender recognition task introduced to further enhance this disentanglement and improve the gender independence of kinship features. Subsequently, a multi-metric fuzzy reasoning module is adopted to integrate kinship features, extract latent kinship cues, and leverage a contrastive loss function to effectively mine potential negative sample information, thereby significantly enhancing the model's robustness. Experimental results on three publicly available datasets demonstrate that the proposed method achieves state-of-the-art performance.

This study examines and compares Mamdani, Sugeno, and ANFIS fuzzy reasoning models for evaluating student academic performance using a bench marking framework where the same inputs, membership functions, and evaluation metrics were applied across the models.

EEG-based deception detection remains challenging due to three critical limitations: high inter-subject variability, which restricts generalization, the black-box nature of deep learning models that undermines forensic interpretability, and substantial computational overhead arising from high-dimensional multi-channel EEG data. Although recent state-of-the-art approaches report accuracies of 82–88%, they fail to provide the transparency required for legal and forensic admissibility. To address these limitations, this study aims to develop an accurate, computationally efficient, and explainable EEG-based deception detection framework suitable for real-world forensic applications. The primary contribution of this work is a novel hybrid neuro-fuzzy architecture that jointly integrates intelligent channel selection, complementary deep feature learning, and transparent fuzzy reasoning, enabling high performance without sacrificing interpretability. The proposed framework follows a five-stage pipeline: (1) intelligent channel selection using Type-2 fuzzy inference with ANFIS-based ranking and multi-objective evolutionary optimization (MOEA/D), reducing EEG dimensionality from 64 to 14 channels (78.1% reduction); (2) dual-path deep learning that combines EEGNet for spatial–temporal feature extraction with InceptionTime-Light for multi-scale temporal representations; (3) a fuzzy attention mechanism to generate interpretable feature importance weights; (4) an ANFIS-based classifier employing Takagi–Sugeno fuzzy rules for transparent decision-making; and (5) triple-level interpretability through channel importance visualization, attention-weighted features, and extractable linguistic rules. The framework is evaluated on two benchmark datasets, such as LieWaves (27 subjects, 5-channel EEG) and the Concealed Information Test (CIT) dataset (79 subjects, 16-channel EEG). Experimental results demonstrate superior performance, achieving 93.8% accuracy on LieWaves and 92.7% on the CIT dataset, representing an improvement of 5.3 % points over the previous best-performing methods, while maintaining balanced sensitivity (92.4%) and specificity (95.2%). In conclusion, this work establishes that neuro-fuzzy integration can simultaneously achieve high classification accuracy, computational efficiency, and forensic-grade explainability, thereby advancing the practical deployment of EEG-based deception detection systems in real-world forensic applications.

Purpose This study aims to propose a hybrid vibration-based fault detection approach to identify early-stage pitting in bevel gearboxes, aiming to enhance machine reliability and prevent catastrophic failure. Design/methodology/approach To obtain vibration signals under both healthy and defective conditions, an experimental test rig was designed and constructed. The Discrete Wavelet Transform (DWT) was used to break down the signals up to the fourth level using a Daubechies-4 mother wavelet. The J-48 decision tree approach was used to pick significant statistical features, which were then categorized using an adaptive neuro-fuzzy inference system (ANFIS) model that combines neural learning and fuzzy logic reasoning. Findings The recommended DWT–ANFIS hybrid approach successfully distinguished between normal, slight, moderate and severe pitting conditions with classification accuracy of 98.01%. Research limitations/implications The study focuses on bevel gearboxes; further research may explore their application to other gearbox types and more complex industrial conditions. Practical implications The model supports real-time condition monitoring of gearboxes, enabling proactive maintenance and minimizing downtime. Social implications By preventing machinery failure, the approach contributes to workplace safety, energy efficiency and sustainable industrial operations. Originality/value To diagnose gearbox faults early, the paper presents an effective hybrid methodology that combines wavelet-based feature extraction, decision-tree-based feature selection and ANFIS classification. This study is innovative because it combines DWT, decision tree-based feature selection and ANFIS classification in a hybrid manner to detect pitting in bevel gearboxes early on, with exceptional accuracy and real-time monitoring capabilities. Peer review The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-08-2025-0383/

Retraction Note: Q-method optimization of tunnel surrounding rock classification by fuzzy reasoning model and support vector machine

In response to the limitations of existing network security alert screening methods in handling high-noise and incomplete data, this paper proposes a multi-level alert screening framework based on DBSCAN density clustering and RETE rule reasoning. The proposed method achieves adaptive analysis and precise screening of alert data by constructing a multi-stage processing pipeline that integrates density clustering, fuzzy reasoning, and dynamic neural networks. Key innovations include: employing the DBSCAN algorithm to perform unsupervised clustering and noise identification of alert data; introducing an improved RETE rule reasoning mechanism that supports weighted fuzzy matching to enhance fault tolerance for incomplete alert streams; and designing a BP neural network with dynamically adjustable structure to achieve accurate alert classification. Experimental results demonstrate that the proposed method achieves significant performance advantages on multiple real-world and benchmark datasets, with a true positive rate of 96.6%, a noise rate controlled within 18.7%, and CPU utilization below 1%, substantially outperforming existing mainstream solutions and exhibiting high practical application value.

The rapid expansion of Industrial Internet of Things (IIoT) infrastructures has introduced a complex, heterogeneous ecosystem that is increasingly vulnerable to diverse cyberattacks. Traditional machine learning- and deep learning-based intrusion detection systems provide strong predictive performance but lack interpretability, rely on static models, and fail to adapt to evolving attack patterns. This study proposes a Dynamic Neuro-Fuzzy Vulnerability Detection System (DNF-VDS) that integrates fuzzy rule-based reasoning with neural network-driven parameter optimization. The system supports automatic rule generation, continuous rule updating, and membership function adaptation, enabling interpretable and adaptive vulnerability detection suited for dynamic IIoT environments. Using the Edge-IIoTset dataset, the proposed system achieves 95% accuracy, 93.5% F1-score, and a false positive rate of 5%, outperforming baseline models, including Bidirectional Long Short-Term Memory, Support Vector Machine, Random Forest, and Decision Tree. Statistical significance testing confirms that improvements are non-random (p < 0.05). Unlike prior neuro-fuzzy approaches, DNF-VDS demonstrates quantifiable interpretability through rule analysis and activation profiling while maintaining scalability for industrial deployment. These findings establish DNF-VDS as a transparent, adaptive, and high-performing solution for IIoT vulnerability detection. Future work will explore real-time deployment and distributed learning across industrial nodes.