Real-world Vehicle Routing Problems (VRPs) involve multiple operational attributes such as capacity limits, time windows, backhauls, open routes, and route duration constraints. Most existing methods are tailored to specific variants, limiting adaptability and reuse. This paper introduces G-UniRouting:$A$Graph-Based Unified Neural Model for Solving Multi-Attribute Vehicle Routing Problems. Our approach integrates a Graph Neural Network (GNN) encoder that captures spatial and relational structure with a constraint-aware decoder that dynamically enforces feasibility through dynamic masking and distance-aware attention mechanisms. Our proposed model is an end-to-end framework trained using the REINFORCE algorithm with a rollout baseline to minimize routing costs. G-UniRouting is evaluated on 24 distinct VRP variants within a unified framework and employs multi-start inference to enhance solution quality. Experimental results show that our unified model generalizes across diverse constraints and consistently outperforms or rivals specialized baselines, offering a scalable and flexible solution to complex, multi-attribute routing problems.

The key factor for the success of Case Based Reasoning (CBR) systems is the quality of their case bases as well as the time spent in case retrieval process which is mainly depending on case bases' size. Indeed, the speed of the retrieval process is seriously decreasing when the case base becomes so heavy. To vouch for case bases' quality, a maintenance process must be provided. Hence, a field for Case Base Maintenance (CBM) emerges. However, a lot of works in CBM field suffers from some limitations and they generally reduce case base's competence during maintenance, especially when cases involving imprecise or uncertain information. To deal with these problems, we propose, in this paper, a new CBM approach named ECTD, Evidential Clustering and case Types Detection for case base maintenance, which is able to manage imperfection in cases by using belief function theory. The key idea of ECTD approach is to use machine learning technique, more accurately the evidential c-means (ECM). Then, it divides cases relative to the different partitions of clusters into four types so that we can subsequently perform the case base maintenance.

This paper presents a method based on fusion technique applied to signatures obtained from external stray flux to detect an inter turn short circuits in induction machines. This technique uses the belief functions framework to represent and merge the information about short circuits obtained from sensors placed around the machine to be diagnosed. The influence of the sensors positions around the machine to detect faults is studied. This fusion technique leads to a new diagnosis method, which only uses the information captured from the stray magnetic field around the machine, having then the advantage of being non-invasive. Six external flux sensors placed on a belt fixed around the machine provide information used for the diagnostic technique. These signatures are obtained by experimental tests using a rewound induction machine that allows one to create inter-turn short circuit faults with different severity levels.

Link prediction is a key research area in social network analysis that enables to understand how social networks evolve over time. It involves predicting the links that may appear in the future based on a snapshot of the social network. Various techniques addressing this problem exist but most of them deal with it under a certain framework. Yet, complete information about the social network of interest is frequently not available as knowledge about the nodes and edges may be partial and incomplete, hence any analysis approach must handle uncertainty in the prediction task. In this paper, we examine the link prediction problem in uncertain social networks by adopting the theory of belief functions. Firstly, a new graph-based model for social networks that encapsulates the uncertainties in the links' structures is proposed. Secondly, we use the assets of the belief function theory for combining pieces of evidence induced from different sources and decision making to propose a novel approach for predicting future links through information fusion of the neighboring nodes. The performance of the new method is validated on a real world social network graph of Facebook friendships.

This paper investigates a multi-criteria decision making method in an uncertain environment, where the uncertainty is represented using the belief function framework. In this context, we suggest a novel methodology that tackles the challenge of introducing uncertainty in the expert evaluations. Therefore, the Analytic Hierarchy Process with qualitative belief function framework is adopted to get numeric representation of qualitative assessment. In this work, we will also focus on two AHP extensions under qualitative AHP. Besides, we intend to describe some comparisons on the standard AHP and the presented models to judge their accuracy. We use also a simulation approach to compare the results of the different models based on different matrices dimensions.

In the belief function theory, several measures of uncertainty have been introduced. One of their possible use is unreliable source discounting before the fusion stage. Two different measures of uncertainty exist which are the intrinsic and extrinsic ones. The intrinsic measure makes it possible to assess the source's confusion whereas the extrinsic one measures the contradiction between sources. In this paper, we associate both measures in order to estimate the global reliability of a source. This method, named Generic Discounting Approach (GDA), is proposed in two different versions: Weighted GDA and Exponent GDA. Those reliability measures are integrated into a classifier. The method was tested, against to some pioneer approaches, on several UCI datasets as well as on an urban image classification problem and showed very encouraging results.

In modeling Multi-Criteria Decision Making (MCDM) problem, we usually assume that the decision maker is able to elicitate his preferences with precision and without difficulty. However, in many situations, the expert is unable to provide his assessment with certainty or he is unwilling to quantify his preferences. To deal with such situations, a new MCDM model under uncertainty is introduced. In fact, we focus here on the problem of modeling expert opinions despite the presence of incompleteness and uncertainty in their preference assessments. Besides, our proposed solution suggests to model these preferences qualitatively rather than exact numbers. Therefore, we propose to incorporate belief preference relations into a MCDM method. The expert assessments are then formulated as a belief function problem since this theory is considered as a useful tool to model expert judgments.

Decision making by considering multiple information sources could provide interesting results. For that reason, fusion formalisms were a major concern in the belief function community. In this context, the Belief function theory allows information fusion thanks to its combinations tools that it integrates. Nevertheless, belief function theory highlights a limit in the merging of contradictory (conflictual) sources. Many authors tackled this problem offering contributions in this field. Unfortunately, no proposed operator has distinguished by its adequacy regardless the type of handled sources. In this paper, we demonstrate the limits of some referenced works and we diagnostic the issues origin. We propose a conflict management approach based on an extra-information that guides the treatment. We also integrate a generic associative base borrowed from the data mining domain in order to apply the adequate conflict management.

This paper investigates the problem of preference modeling under multi-criteria decision making methods in which some assessments cannot be provided in the pair-wise comparison process. Therefore, we introduce a new method that will be able to elicitate preferences in an uncertain environment, where the expert may express incomplete and incomparable ones. Indeed, we suggest to transform these qualitative assessments into quantitative information based on belief function framework. Then, in order to illustrate our approach, we propose to compare our method to the existing approaches.

In this paper, a method for diagnosis of AC machines using the spectrum of the near magnetic field is presented. The method is associated to a fusion process based on belief functions which analyze the measurements. In previous works, it has been shown that it is possible to detect the inter-turns short circuit in the stator windings of electrical machines using a noninvasive method. It is based on the analysis of the variation of sensitive harmonics when the load varies, and eliminates the main drawback presented by other diagnostic methods which use the comparison with a healthy state assumed known. Several measurements around the machine are necessary to increase the probability of the fault detection because the fault position relatively to the sensor can strongly influence the results. So in this paper it is proposed to exploit conjointly the whole measurements in order to obtain a more robust and reliable diagnostic and to increase the probability of detecting the fault. The merging of the different estimations being realized through the belief functions framework, this approach is tested on real measurements. Experimental tests are performed on a special rewound induction machine in order to validate the theoretical approach.

This paper introduces a system for exchanging and managing imperfect information about events in vehicular networks (VANET). Using belief functions, this model is developed through an application using smartphones.

In this paper, we present an analysis of different approaches relative to the correction of belief functions based on the results given by a confusion matrix. Three different mechanisms based on discountings are detailed. These methods have the objective to assess the discounting rates to be assigned to a source of information. These discounting rates allow to correct raw data, based on learnt decisions given by the confusion matrix. These corrections differ according to the use of classical or contextual or distance using discountings. An illustrative example is presented to emphasize the interest and also to show the differences between these adjustments. We carry experimentations on real databases to analyze and interpret these adjustment approaches.

The Dempster-Shafer theory, or evidence theory, is used in different fields such as data fusion, regression or classification. Within the framework of this theory, uncertain and imprecise data are represented using belief functions. Data fusion operators as well as the decision rule of this theory were largely developed and formalized. The aim of the paper is to present modeling methods of knowledge for the initialization of belief functions. Moreover, an experimental comparison of these different modeling methods on real data extracted from images of dermatological lesions is presented.