Université de technologie de Troyes open archive
Not a member yet
10722 research outputs found
Sort by
Access Control Modeling and Validation for Ethereum Smart Contracts
No full textInternational audienceABSTRACT Smart contracts are self‐executing programs that operate on a blockchain network. They are designed to automate transaction execution without the need for intermediaries. Once deployed in the blockchain network, smart contracts cannot be altered. However, this immutability can also lead to security risks. If a smart contract contains a vulnerability upon deployment, it cannot be corrected, leaving the code vulnerable. Therefore, incorporating security considerations during the design phase of smart contract development is crucial. Access control is a key security concept that must be integrated into the design of smart contracts to prevent unauthorized access to critical functions or data. In this paper, we introduce a Model‐Driven Architecture (MDA) approach to design access control for smart contracts, and we validate, using Ethereum, the proposed approach using Smart Contract Security Verification Standard (SCSVS) rules
Energy-efficient optimization of multi-echelon inventory systems
No full textInternational audienceEfficient energy management in the cold supply chain is crucial for reducing costs and environmental impact. This study presents an integrated distribution inventory system that focuses on energy considerations throughout the supply chain. The model incorporates energy usage from production, warehousing, and transportation processes into the average total cost of the system, providing a comprehensive analysis of energy cost components. By considering various factors such as production rate, ordering policy, warehouse filling level, and truck types, the model offers insights into the energy efficiency of the system. The model is formulated as a mixed-integer nonlinear programming (MINLP) problem. To solve this problem, a heuristic algorithm is proposed, aiming to optimize the total cost, including energy costs, while providing near-optimal decision variables. A study based on a real-world company serves as a practical illustration of the model’s effectiveness. A comparison between the integrated inventory system and a non-inventory system reveals significant reductions in energy consumption for warehousing (23.85%) and the overall system costs (2.74%). After testing four groups of datasets, the proposed heuristic algorithm outperforms the LINGO solver in terms of cost minimization (for the first two groups) and computational time, validating its efficiency. Sensitivity analyses are performed to assess the impact of key parameters such as energy unit costs, distance, transportation speed, and demand on energy costs and system performance. These analyses provide valuable insights for decision-makers, supporting informed decision-making and the identification of practical strategies for optimizing energy usage
Multimodal Prompt Distillation for Continual Learning: Continuous Error Correction
No full textInternational audienceMultimodal continual learning is critical for dynamic scenarios, as it enables systems to incrementally learn across diverse modalities. However, significant challenges persist, including cross-modal representation bias and catastrophic forgetting. To address these issues, this paper proposes a continual learning framework based on Multimodal Prompt Distillation, which enhances the cross-modal alignment capability of the CLIP model through a dual-path adaptive mechanism. In the visual branch, a task-specific prompt pool dynamically calibrates image features, while in the textual branch, context optimization enhances visual-language collaboration. A distillation-based correction prompt mechanism is introduced, where feature distillation preserves representation stability, feature scale and shift adjust features, and correction prompts address misalignments. Finally, feature concatenation training strengthens memory retention. Experimental results demonstrate that the proposed method achieves superior performance across multiple benchmark datasets, including ImageNet-R and CIFAR-100. Extensive ablation studies further validate the effectiveness of both the multimodal prompt collaboration and the distillation strategies
An adaptive statistical model of nuclear rods temperature for the detection of total and instantaneous blockage
No full textInternational audienceThe monitoring of critical systems is of the utmost importance, especially when undetected malfunctions could lead to major accidents. This paper focuses on the temperature monitoring of fuel rod assemblies within nuclear power plants, with the goal of detecting total and instantaneous blockages as reliably and quickly as possible. First, we address the modelling of the temperature of the whole fuel rod assembly altogether. We propose a linear parametric model that is adaptive, incorporating previous temperature measurements to enhance its accuracy. This approach allows us to distinguish between regular, non-anomalous temperatures and the anomalous thermal event due to a blockage. The proposed sequential, or online, detection scheme is reliable, as the false alarm rate and detection power are analytically bounded. The model and subsequent statistical test are generic, making the methodology applicable to a wide range of nuclear cores. Numerical experiments, using real temperature measurements from the Superphénix power station, demonstrate the accuracy of the proposed model and the relevance of the detection procedure
Traces, Breadcrumbs, and Patina: Exploring and Designing with Traces of Activity
No full textInternational audienceWhile CSCW increasingly acknowledges the essential role of policies in technology adoption, more research should be conducted on their role in ongoing digital transformation projects. Current retrospective accounts lack information on how the actors interpret and mobilize policies to conduct such projects and achieve their agendas. Our study aims to take part in the effort to address this gap through a qualitative inquiry into an ongoing patient records digitization project within a general hospital service. We emphasize how each actor, from administrative staff to physicians, accounts for different policies to motivate their actions and decisions on the framework of this digitization project
Organisation, pérennisation et résilience : la gestion des connaissances au sein d’unécosystème à haute fiabilité
No full textThis thesis explores knowledge management in an inter-organizational collaborationcontext, focusing on a high-reliability organization (HRO) operating in a regulated safetyenvironment, referred to here as Alpha. Conducted within the framework of an industrialproject with Alpha, the study stems from the need to structure its knowledge management andsustain the knowledge shared with external partners. The research gradually developed aroundorganizational issues specific to knowledge management in this empirical context, whilebuilding on an original theoretical framework. The thesis draws on the literature onorganizational resilience, knowledge management, and knowledge ecosystems to addresschallenges related to sustaining knowledge in an HRO context. It examines how knowledgemanagement enables Alpha to develop resilience capacities and contribute to the emergenceof a knowledge ecosystem with its stakeholders. A qualitative methodology was adopted, basedon case studies within Alpha and its ecosystem.The findings, presented in three articles, show that for HROs, the challenges lie less insimply structuring knowledge and more in building resilience capacities. Developing a robustknowledge management system proves essential for strengthening organizational resilience andsustaining shared knowledge. Ultimately, this thesis contributes to a better understanding ofinter-organizational knowledge management and resilience mechanisms within an HROcontext.Cette thèse explore la gestion des connaissances dans un contexte de collaborationinter-organisationnelle, en se concentrant sur une organisation à haute fiabilité (HRO) évoluantdans un environnement de sûreté régulée, désignée sous le pseudonyme Alpha. Cela dans lecadre d’un projet industriel avec Alpha qui est né de la nécessité de structurer sa gestion desconnaissances et de pérenniser les connaissances partagées avec ses partenaires externes. Larecherche s’est progressivement construite autour de problématiques organisationnellesspécifiques à la gestion des connaissances dans ce contexte empirique, tout en s’appuyant surun cadre théorique original. La thèse mobilise les littératures sur la résilience organisationnelle,la gestion des connaissances et les écosystèmes de connaissances pour aborder les enjeux liés àla pérennisation des connaissances dans un contexte d’HRO. La recherche explore commentla gestion des connaissances permet à Alpha de développer des capacités de résilience etcontribuer à l’émergence d’un écosystème de connaissances, avec ses parties prenantes. Pource faire nous adoptons une méthodologie qualitative, reposant sur des études de cas au seind’Alpha et de son écosystème.Les résultats, présentés en trois articles, montrent que, pour les HRO les défis résidentmoins dans la simple structuration des connaissances que dans la construction de leurscapacités de résilience. En effet, le développement d’un système de gestion des connaissancesrobuste s’avère crucial pour renforcer la résilience organisationnelle et pérenniser lesconnaissances partagées. In fine, cette thèse contribue à une meilleure compréhension desmécanismes de gestion des connaissances inter-organisationnelles et de résilience dans uncontexte de HRO
Broadband integrated directional coupler for heralded single-photon characterization
No full textInternational audienceThe generation of pure single-photon states and their characterization are fundamental in quantum information processing applications like quantum computing. In this work, we design integrated optical directional couplers on a silicon nitride platform for heralded single-photon characterization. We assume the generation of such photons in silicon nitride waveguides through the spontaneous four-wave mixing process. We show that manipulating the waveguide geometry and dielectric contrast makes it possible to obtain nearly wavelength-independent directional coupler designs that perform as 50:50 beamsplitters over a broad bandwidth around the central design wavelength. Assuming two independent but identical heralded single-photon sources, we compare the visibility obtained by simulating the Hong-Ou-Mandel interference for three different 50:50 beamsplitters: an ideal one (perfectly wavelength-independent), an engineered one (nearly wavelength-independent), and an unengineered one (significantly wavelength-dependent). The purity obtained with the engineered beamsplitter closely matches the purity obtained with the ideal one, for which the interference visibility is a direct measure of the quantum state purity. This finding implies that the proposed engineered beamsplitter is suitable for characterizing heralded single-photon states, provided its bandwidth is 23 nm
FDTD-engineered Pd/PMMA nanocomposites for tunable plasmonic hot spots and dual SERS–catalytic functionality
No full textInternational audienceUnlocking the dual power of plasmonics and catalysis, palladium nanoparticles (PdNPs) embedded in poly(methyl methacrylate) (PMMA) offer a versatile platform for next-generation sensing and reaction technologies.Yet, their localized surface plasmon resonance (LSPR) tunability remains less explored compared to traditionalgold and silver systems. In this study, we present a comprehensive design framework using three-dimensionalfinite-difference time-domain (3D FDTD) simulations to optimize Pd/PMMA nanocomposites by systematicallyvarying nanoparticle shape (spheres, cubes, nanorods), size (30–180 nm), and interparticle spacing (down to 5nm). The simulations reveal broad LSPR tunability across 307–700 nm and near-field enhancements up to 18-fold. Experimental validation using 4-mercaptopyridine (4-MPy) as a SERS probe confirms the presence ofplasmonic “hot spots,” delivering enhancement factors of ~105. These findings establish Pd/PMMA nano-composites as chemically resilient, tunable materials for high-performance SERS sensing and plasmon-assistedcatalysis in harsh or dynamic environments
Directional light scattering in Mie-resonant Si particles with ultra-thin plasmonic shells
No full textInternational audienceWe present the synthesis and characterization of Au-decorated Si core-shells as candidate meta-atoms. We found a damped magnetic dipole (MD) for smaller Si cores (100 – 130 nm) and an enhanced MD for larger cores (150 – 200 nm). Continuous plasmonic shells of ~12 nm are needed to significantly improve forward scattering.Subwavelength-sized Si particles interact strongly with visible (vis) and near-infrared (NIR) light to produce strong electric and magnetic resonances. These can combine to produce interesting optical effects, such as pure forward scattering. A requirement for this to occur efficiently is that the electric dipole (ED) and magnetic dipole (MD) modes must be of similar amplitude and phase. At wavelengths where this occurs, the particles act analogously to the forward-propagating point sources of light used in Huygens’ constructions. This directional scattered light has a range of potential applications in the creation of metamaterials.We have investigated dielectric@metal core-shell architectures comprised of both resonant cores and resonant shells as candidate particles in which the spectral overlap of the electric and magnetic dipoles might be controlled to create strong directional scattering. There are currently two reports of the synthesis and characterization of Au shells around Si cores, both thicker than desirable. [1,2] Chaâbani et al. presented Si@Au particles, with non-uniform shells composed of Au particle diameters between 10 and 25 nm, which presented enhanced electric field and Fano-resonances due to the coupling of the Mie modes of the Si core and the localized surface plasmon resonance (LSPR) of the Au shell. [1] Sugimoto et al. similarly presented Si@Au particles with a rough ~25 nm Au shell. [2] In both of these reports of Si@Au core-shell particles, the experimental data could not be accurately fit by simulations due to non-spherical cores and inhomogeneous shells. Ultrathin and homogeneous coatings have not yet been achieved around spherical Si particles.In this study, we present a two-step aqueous approach to prepare Si@Au core-shell particles with controllable shell thickness below 10 nm. The Au nanoparticle (AuNP) density around the Si particles can be increased by performing a second functionalization/deposition step. We studied the electromagnetic response of the particles using single-particle scatter spectroscopy and electron energy loss spectroscopy (EELS). Our results were compared with reference Si spheres and SiO2@Au core-shell particles, to allow us to establish the contribution from the Au decoration to the optical response of the hybrid particles. To further elucidate the nature of the electromagnetic response of the particles, these observations were supported by T-matrix simulations which replicated our experimental findings, and showed the importance of controlling the shell/core dimensions and the need for a continuous shell to maximize forward scattering. We found that continuous plasmonic shells of ~12 nm thickness are needed to significantly improve forward scattering intensity.References1.Chaâbani, W, J Proust, S Ouellet, A Movsesyan, J Béal, R Bachelot, T Xu, A L Baudrion, PA Adam, D Boudreau, A Chehaidar, and J Plain, “Si@Au core–shell nanostructures: Toward a new platform for controlling optical properties at the nanoscale.” J Phys Chem C, Vol. 125, 20606. 2021. DOI: 10.1021/acs.jpcc.1c061822.Sugimoto, H., T Hinamoto, Y Kazuoka, A Assadillayev, S Raza, and M Fujii. “Mode hybridization in silicon core–gold shell nanosphere.” Small Vol. 18, 2204890, 2022. DOI: 10.1002/smll.20220489