Portail des publications scientifiques IMT Mines Alès
Not a member yet
5198 research outputs found
Sort by
Assisting the early development stages of privacy-aware software: the PRIAM tooled metamodel for GDPR
No full textPart of a special issue on Regulatory compliance in software engineeringInternational audienceContext:As software systems are more tailored to users, personal data is collected and exploited more than ever before. This situation raises the issue of user privacy protection. Conforming to personal data protection regulations, such as the European General Data Protection Regulation (GDPR), has thus become a legal obligation for application providers. However, there are no widely adopted proposals to formalize, implement, and assess compliance with the personal data privacy protection required by GDPR.Objective:In order to help application developers in the early stages of the development process, our overarching objective is to propose a tooled software engineering approach to integrate personal data protection capabilities, thus contributing to the development-by-design of privacy-aware software aligned with GDPR requirements.Method:We developed a method called PRIAM (PRIvacy Assessment Method) that goes beyond a conceptual description of the regulation by incorporating concrete, actionable software artifacts. This article presents the cornerstone of this method – PRIAM metamodel – along with its companion artifacts.Results:PRIAM metamodel captures the main concepts of GDPR and is then supported by a domain-specific language, user stories, and a dedicated database schema. The comprehensiveness and relevance of PRIAM metamodel have been qualitatively evaluated by GDPR experts through a questionnaire. Complementarily, an AI-based evaluation has been conducted, using some Large Language Models (LLMs), opening perspectives for fast, iterative evaluations of metamodels that formalize regulation texts. Besides, the practicality and usefulness of PRIAM metamodel and all its companion artifacts are highlighted through the running example of a Sport center management application, where privacy enforcement features, tailored to the specific personal data of the application, are generated and integrated.Conclusion:These two elements assert the viability of our proposal as a practical solution for assisting the development of privacy-aware applications that are compliant with GDPR requirements, thanks to customizable sets of actual development artifacts, systematically derived from a validated comprehensive formalization of the regulation articles
Effect of alignment of h-BN platelets in LDPE disks on the thermal conductivity: Microstructure and modeling
No full textInternational audienceIn recent years, hexagonal boron nitride (h-BN) has gained significant attention due to its high thermal conductivity and low electrical conductivity, making it an attractive option for effective heat dissipation in electronic devices. This article specifically explores the impact of unidirectional alignment of h-BN platelets on the thermal conductivity of a low-density polyethylene matrix composite (LDPE/h-BN). The composite materials were fabricated by layering thin films, and various parameters, such as platelet size and film thickness, were systematically investigated. Additionally, dynamic mechanical behavior analysis (DMA) was conducted to assess the influence of incorporating h-BN particles on the dynamic mechanical properties of the composites. The study also involved modeling to enhance our understanding of the correlation between particle orientation and thermal conductivity. Regarding the difference between both h-BN and the use of different dies to elaborate the films, in-plane TC are relatively different because the orientation of the platelets in the final disk is highly influenced by both parameters. This alignment led to a remarkably high thermal conductivity value of 4.25 W/(m·K) with 50 wt% of h-BN 003. The study underscored the critical roles played by particle size and film thickness in achieving optimal thermal conductivity. Notably, this study stands out by avoiding the use of solvents during the composite development process, which sets it apart from approaches generally developed
Fluorination to Convert the Surface of Lignocellulosic Materials from Hydrophilic to Hydrophobic
No full textThis article belongs to the Special Issue Superhydrophobic Surfaces: Wetting Phenomena and Preparation MethodsInternational audienceNatural fibers are increasingly used as sustainable, lightweight, and low-cost alternatives to glass fibers in polymer composites. However, their inherent hydrophilicity and surface polarity limit compatibility with non-polar polymer matrices. Both gas/solid and plasma fluorination modify only the surface of lignocellulosic materials. Mild conditions are mild, with reactivity governed by fluorine concentration, temperature, and material composition. Surface energy is typically assessed through contact-angle measurements and surface analytical techniques that quantify changes in hydrophobicity and chemical functionalities. In wood, fluorination proceeds preferentially in lignin-rich regions, making lignin a key component controlling reactivity and the spatial distribution of fluorinated groups. Natural fibers follow the same logic as for flax, which is a representative example of lignin content. Applications of fluorinated bio-based materials include improved moisture resistance, enhanced compatibility in composites, and functional surfaces with tailored wetting properties. Scalability depends on safety, cost, and process control, especially for direct fluorination. Durability of the treatment varies with depth of modification, and environmental considerations include the potential release of fluorinated species during use or disposal
SVOCs, VOCs and microbiological contamination in sports facilities, what are people exposed to?
No full textInternational audienceThe indoor air quality of sports facilities was studied in France for different types of sports rooms (fitness, dojos,motor skills, physical activity). A total of 53 VOCs, 44 SVOCs and 7 microbiological parameters were assessed inair or settled dust. In air, levels of VOCs and SVOCs concentrations in air were like those found in schools andhomes, and SARS-CoV-2 was not identified. High concentrations of total bacteria were observed in two sportfacilities with values exceeding the indoor air guideline of European directive and the World Health Organization.In indoor dust, SVOCs ranged from <0.05 μg/g to 274 μg/g, and five SVOCs showed median concentrationshigher than in homes, daycare centers, or schools: 2-ethylhexyl diphenyl phosphate (EHDPP), triphenyl phosphate(TPP), benzophenone, 4-tert-butylphenol (4tBP), and 4-tert-octylphenol (4tOP). Considering the toxicity ofthe compounds quantified, a risk assessment seems necessary for the health of athletes and children in gyms, butexposure parameters for sports activities (dust ingestion, inhalation rates, and skin contact) are not available atthis time
AI-enhanced RUL prediction of PEMFCs under dynamic operating conditions using XGBoost-based HI extraction and hybrid transformer-GRU model
No full textInternational audienceProton Exchange Membrane Fuel Cells (PEMFCs) are critical for zero-emission energy systems, particularly in electro-hydrogen generators (GEH2). Accurate Remaining Useful Life (RUL) prediction is crucial for ensuring operational reliability and enabling predictive maintenance. However, dynamic operating conditions present a significant challenge for existing prognostic approaches, particularly in extracting robust Health Indicators (HIs). Conventional HIs, often based on voltage or power, are highly sensitive to mission profiles and fail to generalize in real-world conditions. To address this limitation, we propose a novel data-driven approach based on XGBoost regression to extract a degradation-specific HI directly from raw voltage measurements. This method effectively filters out transient fluctuations caused by varying power demands, isolating the true degradation trend without requiring complex preprocessing or domain expertise. Leveraging the extracted HI, we introduce a hybrid deep learning model that combines Transformer networks and Gated Recurrent Units (GRUs) to capture temporal dependencies and provide accurate RUL predictions under dynamic conditions. Explainable AI techniques are integrated to interpret the model’s predictions and analyze the influence of operational variables on fuel cell degradation. The proposed framework is validated on a real-world industrial dataset from four PEMFC stacks operating in GEH2 systems. Experimental results demonstrate superior accuracy, robustness, and generalizability compared to state-of-the-art methods, highlighting the potential of this scalable and interpretable approach for predictive maintenance in complex industrial environments
Enhanced Temporal Convolutional Network Based Approach for Degrada-tion Prediction of Reverse Osmosis Systems
No full textInternational audienceReverse Osmosis (RO) degradation underscores the importance of predictive capabilities to develop optimal maintenance strategies that minimize losses. In this study, we develop a Temporal Convolutional Network (TCN) model to predict the RO system states using the primary indicator for RO analysis: the fluctuations in differential pressure across the RO vessel. Specifically, data from a real desalination plant for the period 2015 to 2020 are used. The dataset encompasses 14 RO train operations, including routine operations, significant maintenance events, temporary shutdowns, and element replacements. The proposed approach uses temporal convolutional operations to capture the dynamic pressure behavior at both ends of the membrane, enabling faster, more accurate anomaly detection. A key challenge in applying deep learning to this domain is the heavy reliance on real-world operational data. The approach involves a strong data preprocessing strategy that reveals subtle relationships between operating time and pressure dynamics. Accurate prediction of membrane degradation also ena-bles preventive and recovery actions, which reduce maintenance expenses. The proposed method is evaluated against con-ventional models, including LSTM, CNN-LSTM, and GRU, using data from the real desalination plant. Experimental results demonstrate that the proposed model achieves the lowest predic¬tion error and shows strong potential for deployment in practical desalination operations
Advancing Holonic Systems in the Era of Artificial Intelligence and Digital Twins for Trustworthy and Effective Human-Centric Intelligent Systems
No full textInternational audienceHolonic systems provide a powerful paradigm for modeling complex, distributed, and autonomous systems. However, their deployment in large-scale, heterogeneous environments raises challenges related to coordination, semantic alignment, bias migration, ethical reasoning, and real-time adaptation. This paper extends classical holonic architectures by introducing five complementary holon dimensions: effective, understanding, unbiased, vigilant, and ethical holons. Each dimension targets a specific capability required for resilient and socially responsible system behavior. These capabilities range from sustainability-aware decision-making and contextual understanding to bias mitigation and situation awareness. To enable coherent operation across autonomous and heterogeneous holons, this paper proposes to integrate federated interoperability as a foundational architectural principle. A dedicated federated interoperability holon (FIH) is proposed to dynamically manage technical and semantic interoperability. It also supports on-the-fly ontology alignment through short-lived ontologies as well as orchestrates coordination among specialized holons without centralized control. The framework is analyzed through a holon-based simulation approach, where holons act as autonomous simulation entities and are instantiated as High Level Architecture (HLA) federates. The applicability of the proposed framework is illustrated through a rural mobility use case focused on transportation systems. In this scenario, holonic coordination supports decision-making regarding multimodal integration, routing strategies, and the identification of mobility deserts. Overall, this work advances holonic system design by proposing an integration of specialized holon roles with federated interoperability mechanisms. It provides a scalable foundation for adaptive and socially aligned systems in domains such as mobility, smart territories, and enterprise ecosystems
How our homes shape the way we move
No full textInternational audienceBackgroundHuman movement within the home remains poorly understood despite growing interest in smart environments and aging-in-place technologies. Investigating walking behavior within domestic spaces, which are subject to specific constraints and obstacles, may reveal significant differences from outdoor walking and provide a deeper understanding of human-environment interactions in indoor settings.MethodsWe present a four-year longitudinal study of natural indoor walking behavior, conducted using proximity-sensitive flooring in a smart apartment located in Montpellier, in southern France. Eight participants, organized into pairs, were monitored to observe their natural indoor walking behavior.ResultsThe data, processed through algorithmic analysis, enabled the characterization of walking behavior indoors. By analyzing 2.5 million steps and over 346,000 individual walking trajectories, we show that indoor locomotion is fragmented, variable, and strongly constrained by the environment
Temporally Consistent and Controllable Video Generation of 2D Cine CMR via Latent Space Motion Modeling
No full textInternational audienceCine cardiac magnetic resonance is the gold standard for assessing cardiac function, but the scarcity of public datasets limits the development of advanced data-driven models. To address this limitation, we propose a generative method for synthesizing temporally coherent and anatomically consistent cardiac sequences. Our text-to-video framework decouples cardiac spatial structure from temporal motion. First, a fine-tuned diffusion model synthesizes an initial frame from a clinical text prompt, controlling anatomical features. Then, a latent flow model conditioned on a cardiac phase embedding generates the complete cardiac motion, ensuring spatial consistency and temporal control. Our model generates anatomically and pathologically diverse sequences with high temporal coherence and strong fidelity to input prompts, achieving a FID of 31.68 for image realism and a CLIP score of 31.04 for text-image alignment. These experimental results highlight its potential to produce high-fidelity, on-demand medical data, offering a scalable solution to data scarcity
Allocation dynamique de ressources : comparaison d'approches de recherche opérationnelle et d'apprentissage par renforcement
No full textNational audienceDe nombreux secteurs s'intéressent à la gestion de files d'attente physiques ou digitales. L'optimisation de ces files d'attente est par exemple particulièrement cruciale dans les hôpitaux, où la gestion des services d'urgence est un enjeu majeur. Nous exposons dans cette note le positionnement de nos travaux sur la gestion de files d'attente et l'allocation de ressources. Ceux-ci, résolument appliqués, sont réalisés en collaboration avec la société ESII, qui apporte ses données ainsi que son expertise en matière d'optimisation de files d'attente.Nous nous intéressons spécifiquement au problème d'allocation d'agents physiques à des usagers arrivant dynamiquement et stochastiquement sur un site. Nous nous appuyons pour cela sur des approches d'optimisation issues de la Recherche Opérationnelle (RO) et de l'Intelligence Artificielle (IA). L'objectif est de proposer la meilleure allocation - au sens de la minimisation du temps passé par les usagers dans le système - en un temps de calcul quasi-instantané