Journal of Multi-Criteria Decision Analysis最新文献

The necessary and sufficient conditions for achieving optimality in multiobjective fractional control problems with multiple integrals are derived and verified in this study. These problems are analysed using fractional calculus, particularly the Riemann–Liouville integral, which generalises traditional integer-order integrals to non-integer orders, allowing for more flexible modelling of real-world systems. Under the assumption of quasiinvexity, we present adequate conditions for the efficiency of feasible solutions.

Although Stochastic Multicriteria Acceptability Analysis (SMAA) has been widely applied in real-world decision problems, limited research has examined the structural conditions that lead to rank disagreement between deterministic and stochastic model outputs. This paper addresses that gap through a simulation-based analysis of 50 randomly generated decision problems. First, one-hot encoded vectors were developed to compare the deterministic top-ranked alternatives with their SMAA rank acceptability distributions to evaluate rank divergence. Descriptive statistics showed that cases with disagreement had a substantially higher mean Jensen–Shannon Distance (JSD) (0.79) in comparison to non-divergent cases (0.43). Moreover, scatterplot analysis revealed that divergent cases typically have high JSD values (≥ 0.6), low rank-1 acceptability (≤ 0.2), and high rank expectation (≥ 4). Second, statistical techniques were used to compare differences between structural features, i.e., criteria, alternatives, minimum and maximum criteria. Furthermore, the Criteria Balance Score (CBS) was developed to quantify criteria type imbalance, where values of 0 show perfect balance and scores close to 1 demonstrate disparity. Results showed that divergent cases included decision problems with statistically significant larger model complexity, i.e., number of criteria, and criteria type min/max balance, which was an unexpected finding. Third, threshold-based analyses revealed that 62.5% of divergent cases included decision structures with 10 or more criteria, and that 75% of diverging cases with CBS below 0.20 had a min/max criteria type difference of 0 or 1. Finally, consistency in divergence patterns was independently explored within four multicriteria decision analysis models. Findings suggest that divergence is largely a function of decision space characteristics, rather than idiosyncrasies of individual models. Together, these findings provide real-world decision makers, analysts, and researchers with practical, evidence-based thresholds for instances when deterministic results may not be robust. By identifying these structural warnings in advance, decision makers can increase stakeholder trust and reliability in the decision-making process.

This paper investigates the optimality conditions for a class of nonconvex multitime fractional multiobjective variational problems. By using the parametric approach, we propose two novel inequalities: the weak multitime fractional vector variational-like inequality problem (WMFVVLIP) and the multitime fractional vector variational-like inequality problem (MFVVLIP). To address these problems, we establish an equivalence between the efficient solutions of the original problems and the solutions of the introduced inequalities. Furthermore, we apply the KKM lemma to demonstrate the existence of solutions to the (MFVVLIP). In addition, a numerical example is provided to illustrate the applicability of the proposed methodology and to demonstrate the effectiveness of the derived inequalities and the corresponding efficient solutions.

In multi-criteria decision-making (MCDM), structuring the problem by defining relevant alternatives and criteria is a critical prerequisite for effective analysis. In this paper, this foundational phase is addressed within the multifaceted context influencing teacher performance, which represents a crucial task for effective decision-making in education. However, traditional approaches often struggle to capture the complex interaction between the different features distributed over the teacher's living environment, work setting and emotional state, which represent a large and complex set of potential decision criteria. These features, represented as criteria, are essential for a comprehensive understanding of a teacher's context, represented as alternatives. The proposed approach introduces a formal, ontology-driven approach to this problem structuring task. We investigate the impact of feature selection on representing the multidimensional context of teachers (the alternatives), both individually and collectively. We propose a novel, unsupervised feature selection approach based on feature variance, which leverages a teacher context ontology to identify the most salient criteria (features) for subsequent analysis. By employing an importance-based threshold, the approach efficiently eliminates features with minimal explanatory power, leading to a more parsimonious and interpretable representation. Additionally, the proposed approach demonstrates superior performance according to the selected context in several key areas, providing a consistent, reliable set of representing features across different variations of data. Moreover, the proposed approach generates interpretable structures, such as lattices, to facilitate informed decision-making.

Transparency in computing is an important precondition to ensure the trust of users. One concrete way of delivering transparency is to provide explanations of computing results. To this end, we introduce a method for explaining the results of various linear and hierarchical multi-criteria decision-making (MCDM) techniques such as the weighted sum model (WSM) and the analytic hierarchy process (AHP). The two key ideas are (A) to maintain a fine-grained representation of the values manipulated by these techniques and (B) to derive explanations from these representations through merging, filtering, and aggregating operations. An explanation in our model presents a high-level comparison of two alternatives in an MCDM problem, presumably an optimal and a non-optimal one, illuminating why one alternative was preferred over the other. We show the usefulness of our techniques by generating explanations for two well-known examples from the MCDM literature. Finally, we show their efficacy by performing computational experiments.

In the domain of context-aware recommender systems, understanding and leveraging feature interactions is crucial for enhancing recommendation quality. Feature interactions delve into the complex interdependencies among user characteristics, item attributes, and contextual factors like time and location. Traditional models often struggle to effectively combine these diverse features, potentially leading to suboptimal recommendations. To tackle this issue, we propose enhancing context-aware recommender systems through deep feature interaction learning. Our model, which combines BiLSTM and Hybrid Attention mechanisms, offers a sophisticated architecture designed to exploit deep feature interactions effectively. This approach ensures that our system captures essential contextual dynamics, thereby improving the effectiveness of the recommendation process. Experimental results across multiple datasets validate the efficacy of our approach, showing significant improvements in key metrics such as $��\mathrm{AUC}�$ and $��\text{LogLoss}�$ compared to traditional and contemporary models. These achievements underscore our model's ability to deliver nuanced and adaptively tailored recommendations, marking a valuable contribution to the field of recommender systems.

Multiobjective optimization problems involve several conflicting objective functions to be optimised simultaneously and solutions to these problems represent different trade-offs. When applying interactive methods, a decision maker with domain expertise provides one's preference information over several iterations, according to which new solutions are computed until finding a solution with the most preferred trade-offs. Publications on interactive multiobjective optimization methods mainly focus on the optimisation algorithm, and little attention is paid to their implementations, not to mention the development of user interfaces that enable interaction with the decision maker. User interfaces involve icons but there are no studies about icons for the specific functionalities of multiobjective optimization methods. Icons convey meaning effectively to users interacting with technology. With these small pictorial representations, information on system functionalities is communicated quickly. However, the immediacy of icon recognition can also lead to misunderstandings and difficulties in using the system if they are not designed properly. Semantic distance in icon design indicates the closeness of the pictorial representation to its intended functionality and, thus, functions as the main principle in designing effective icons. An empirical study ($��N�=�38�$) was conducted to examine the semantic distances of icons for interactive multiobjective optimization methods implemented in an open-source software framework. The study addressed the main functionalities. According to our main findings, we suggest an icon set for the considered functionalities, to enable fluent interaction with decision makers and other involved parties utilising interactive multiobjective optimization methods via user interfaces.

In this study, by using the concepts of (strictly) strongly convexity and preconvexity, associated with controlled multiple integral type functionals, and a mean value type theorem, we formulate some connections between new classes of generalised Minty (weak) variational inequalities of vector-type and the corresponding multiple-objective extremization problems. The considered classes of variational models are motivated by their applications in real-world modelling problems. The presence of control variables and controlled multiple integrals are the main tools in establishing the new outcomes.

In this paper, a new concept of generalised convexity and a new class of exact exponential penalty method, namely the concept of $��\left(�p�,�r�\right)�$-E-invexity and $��l_1�$ exact exponential penalty E-function method, respectively are introduced for (not necessarily) differentiable vector optimization problem in which functions are E-differentiable. The conditions governing the equivalence between sets of (weak) efficient solutions of the original constrained E-differentiable vector optimization problem and of its associated unconstrained exponential penalised vector optimization problem are studied. Examples are given to illustrate the obtained results.