A minimum adjustment consensus model integrating trust development and dynamic outlier detection for large-scale group decision-making

Assessing anti-insurgency and religious violence mitigation policies using interval-valued Fermatean fuzzy hybrid group decision-making

A granular consensus-reaching process using blockchain-based mechanisms to foster trust relationships in fuzzy group decision-making

Mega construction projects (MCPs) face significant challenges in risk management due to their inherent complexity, particularly under uncertain environments. Existing risk management approaches often exhibit limitations when dealing with incomplete information and stakeholder divergence, making it difficult to comprehensively quantify risk uncertainty and its potential impacts. This study proposes a hybrid method that integrates gray fuzzy group decision-making (G-FGDM) with Monte Carlo simulation (MCS) to enhance risk identification and assessment in MCPs. First, a systematic framework of risk factors was developed through an extensive literature review. Second, the G-FGDM approach was employed to synthesize evaluations from multiple stakeholders, thereby quantifying the weights and fuzzy assessment values of various risk factors. Finally, MCS was used to simulate the probability distributions of different risk scenarios, enabling a dynamic analysis of the impact and uncertainty of risks on project success. Using a real-world MCP as a case study, the results indicate that technical and construction risks exert the highest impact, followed by financial and social risks, while policy risks demonstrate relatively stable effects. This study not only validates the effectiveness of the proposed hybrid method but also provides valuable decision-making support for improving risk management practices of MCPs in uncertain environments.

A two-stage three-way dynamic consensus approach for large-scale group decision-making in social networks with incomplete information and adjustment willingness

Identification and management of non-cooperative behaviors in large-scale group decision-making: Review, taxonomy and challenges from an LLM perspective

In response to the challenges of handling linguistic uncertainty in multi-criteria group decision-making (MCGDM), this paper introduces a novel decision framework based on linguistic q-rung orthopair fuzzy (Lq-ROF) sets. The motivations from the need to systematically address ambiguity and inconsistency in linguistic evaluations provided by decision-makers. To this end, the study develops three main methodological contributions. Firstly, a normalized bidirectional projection measure (NBDP) and its weighted extension (WNBDP) are proposed to resolve ranking inconsistencies in linguistic environments. Secondly, an axiomatized knowledge entropy measure for Lq-ROF information is established, enabling fine-grained differentiation among linguistic assessments and facilitating dynamic expert weighting. Thirdly, a non-linear programming model is formulated to objectively derive both attribute and expert weights by integrating bidirectional projection with generalized entropy principles. The proposed framework is rigorously evaluated through comparative studies against established methods, including aggregation-based techniques, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), Evaluation based on Distance from Average Solution (EDAS), and Complex Proportional Assessment (COPRAS). Results validate the robustness and theoretical advantages of the approach, confirming its effectiveness in quantifying linguistic uncertainty and delivering consistent decision support in complex MCGDM contexts.

The problems of consumer value in purchasing behavior are usually associated with their high degree of uncertainty, hesitation, and contradicting opinions among the decision-makers. Conventional fuzzy and intuitionistic fuzzy models cannot adequately represent such an achievement and directional uncertainty. Given this limitation, the study offers a novel hybrid multi-attribute group decision-making (MAGDM) model that integrates weighted aggregated sum product assessment (WASPAS) and criteria importance through intercriteria correlation (CRITIC) models into the circular spherical fuzzy set (CSFS) framework. The objective aspect of the proposed method is that it employs the CRITIC method to calculate the weight of the attributes based on the intensity of contrast, and the weighted product model is used to view the summation of the weighted attributes and product models. By doing so, the WASPAS method produces a strong alternative ranking. Representation of membership, abstinence, non-membership, and radius-based hesitation can be more expressive in the CSFS environment, which reflects the uncertainty of decision-makers in a more realistic way. The relevance and suitability of the proposed model are applied to a hypothetical case study of consumer value in buying behavior. The results indicate that "Simplifying the Buying Process" is the optimal strategy across all evaluation scenarios. Sensitivity and comparative analysis also prove the stability, robustness, and high-quality of the proposed CRITIC-WASPAS model in the framework of CSFS as compared to the current fuzzy decision-making methods. The proposed work offers feasible decision support to businesses and offers novel directions to advanced MAGDM applications in the face of complex uncertainty.

Large-scale group decision-making (LSGDM) using ordinal preference information provides an effective means of addressing the complexities of decision-making problems by streamlining the decision-making process and alleviating the burden on decision makers (DMs). Considering these benefits and current research limitations, this study introduces an integrated framework designed to facilitate ordinal consensus in LSGDM. Firstly, to enhance computational efficiency and address the shortcomings of traditional methodologies, an Improved Fuzzy C-Means for Applications with Noise (IFCMAN) algorithm is developed. Secondly, a new ordinal consensus measure grounded in the majority principle is introduced. Thirdly, recognising the interactive nature of consensus formation, cooperative game theory is employed to design a consensus-adjustment mechanism via a two-stage core-Nash bargaining approach. Furthermore, subgroup collusion is analysed, culminating in the characterisation of an optimal collusion strategy. Finally, the effectiveness and feasibility of the proposed method are illustrated through a pandemic-control case study, accompanied by relevant analyses. The results demonstrate that the proposed approach provides a robust and practical framework for enhancing consensus building in LSGDM with ordinal preferences.

An enhancing framework with an emphasis on decision balance in ensemble regression.

A consensus method based on reinforcement learning for group decision-making

Consensus reaching process based on three-way clustering and conflict detection in large-scale group decision-making

A comprehensive approach employing interval-valued q-rung orthopair fuzzy data for multi-attribute group decision-making

In complex decision-making environments, expert judgments are often uncertain and divergent, making aggregation into a single numerical value problematic. The classical Decision-Making Trial and Evaluation Laboratory (DEMATEL) addresses interdependencies among criteria but aggregates expert inputs using simple averaging, which neglects opinion diversity and creates an artificial sense of certainty. To address these limitations, we propose Group Interval-Based DEMATEL method Balancing Satisfaction and Uncertainty (GIBSU-DEMATEL), a novel framework that represents group judgments as intervals and optimizes their construction to simultaneously maximize expert satisfaction and minimize judgmental uncertainty. This dual-objective approach ensures that aggregated results better reflect the diversity of expert perspectives while maintaining analytical rigor. Unlike existing interval or fuzzy DEMATEL variants, which focus primarily on uncertainty, GIBSU-DEMATEL formalizes the trade-off between expert satisfaction and uncertainty through an optimization function, producing causal maps that are both interpretable and socially acceptable. The framework is validated through a case study on learning management system selection in higher education, demonstrating improved robustness, higher satisfaction, and transparent representation of divergent opinions compared with classical, interval and fuzzy DEMATEL. Sensitivity analysis further confirms the adaptability and reliability of the method across decision scenarios.

Primacy effect-based dynamic feedback mechanism considering communication sequence for multilevel infiltrative large-scale group decision-making

Three-dimensional dual hesitant fuzzy constructions of underlying operations, aggregation operators, decision frameworks and their applications of multiple attribute group decision-making

The rapid advances in artificial intelligence (AI) and machine learning have transformed computer technology. Now, computers can learn and adapt autonomously, make sound judgments, and create environments in which humans and AI work in harmony. Analytical frameworks that can process data with uncertainty, imprecision, and multiple perspectives are essential for the success of human-AI collaboration. In this connection, linguistic Z-number (LZN) evolution models play a crucial role in selecting tools for human-AI collaboration because they can better capture uncertainty and fuzziness. Given this, we extend multicriteria group decision-making (MCGDM) by incorporating LZNs. The structure of MCGDM under LZNs, the consensus-building process, estimation of experts' weights, and the ranking of options are the main challenges. In the experts' primary opinion-based MCGDM, we believe that expert participation and the opportunity to revise their opinions with original linguistic terms (LTs) would be more practical. However, in existing studies, experts have no scope to revise their opinions based on the original LTs. Most studies on consensus-building either do so automatically or advise experts to revise their opinions based on virtual LTs. But either automatically or virtually, LTs cannot be physically represented by words, which is a significant drawback. Given these facts, we propose an interactive, strategy-based consensus-building in which we advise experts to revise their opinions within the original LTs. To estimate experts' weights, we developed an integrated subjective and objective method. By simultaneously assigning subjective weights and opinion-based relative-relevance indices to the experts, we derive the adjusted subjective expert weights. The objective weights of experts are derived from the criterion-based entropy method. The ranking of options, we proposed the prospect and regret theory-based TOPSIS ranking approach to rank the options. Finally, a case study on the optimal selection of an AI tool for human-AI collaboration validates the proposed method. Through a comparative analysis with alternative methodologies, we demonstrate the feasibility, stability, and superiority of the proposed model.

In the rapidly changing world of banking systems, intelligent banking has emerged as a revolutionary force in the rapidly growing financial technology world. Smart banking provides a seamless, personalized experience, leveraging intelligent systems that apply technologies such as artificial intelligence (AI), blockchain, and big data to deliver mobile financial services. However, the risks of such innovations can also be confusing, such as cyber threats, data breaches, and vulnerabilities in work, and these aspects require strict assessment and control. This article presents a systematic multi-criteria group decision-making (MCGDM) model for risk assessment in intelligent banking, which incorporates the Circular Intuitionistic Fuzzy Bonferroni Mean (CIFBM) to assess the state of expert judgments and effectively manage uncertainty and ambiguity. Moreover, important theorems and properties, such as idempotency, monotonicity, and boundedness, have ensured its efficiency and reliability in risk assessment. A case study shows that the proposed approach has real-world applicability and can be used to prioritize risks, make decisions, and improve the security, efficiency, and flexibility of banking operations. This work provides a new, practical framework for addressing emerging risks in intelligent banking systems. It makes a valuable contribution to researchers and practitioners in the field of financial technology by combining rigorous theoretical development with practical implementation.

These days, our lives are governed by a large number of decisions. They are subject to consensus or disagreement among the players involved in the decision-making process. So it’s important to make a decision that meets everyone’s aspirations. Several decision-support methods have therefore been developed, but they all have their shortcomings. The EVAMIX method is one such aggregation method that has good properties but is used in the context of a single decision-maker. The aim of our work is to extend the EVAMIX method to group decision-making. To this end, we have proposed a collective aggregation method based on the EVAMIX method, incorporating the Gini mean and the hybrid operator of Zimmermann and Zysno. Finally, we conducted numerical experiments, which produced some interesting results.

The evaluation of artificial intelligence-empowered vocational education curriculum reform effects constitutes a multi-attribute group decision-making problem. Recent studies have applied advanced methods, including the Exponential TODIM (Tomada de Decisão Interativa Multicritério) and MABAC (Multi-Attributive Border Approximation Area Comparison) frameworks, to address such complex decision-making tasks. Given the pervasive uncertainty in performance data, probabilistic linguistic term sets provide a reliable mechanism for representing and processing imprecise information. This study develops an integrated probabilistic linguistic Exponential TODIM-MABAC (PL-ExpTODIM-MABAC) model to address multi-attribute group decision-making problems within a probabilistic linguistic environment. The MEREC (Method based on the Removal Effects of Criteria) method is employed to derive objective attribute weights from probabilistic linguistic term sets data and determine the relative importance of each evaluation criterion, while a detailed case study validates the practicality and effectiveness of the proposed approach.