The medIAte project arises in response to the growing deployment of AI-based decision-support systems in industrial settings. These systems are designed to assist or automate part of the operational decision-making process and are commonly seen as a mean to boost productivity and enhance the reliability of operations. However, their specific effects on human factors—particularly the cognitive, emotional, and behavioural dimensions of operators—remain largely unexplored. Although such systems can alleviate cognitive workload and facilitate decision making under certain conditions, there is a concern that excessive automation may diminish workers’ sense of autonomy, motivation, and overall engagement, ultimately impacting their well-being and performance. In contrast, maintaining a high degree of operator autonomy and decision-making power might foster positive psychosocial outcomes. In this context, the medIAte project is designed to examine how varying levels of AI decision support affect these psychological dimensions and worker performance in a production environment. To address these issues, two main hypotheses guide the research. One hypothesis posits that a high level of automation in the decision-making process could negatively affect operators by reducing their motivation, engagement, and performance. The alternative hypothesis suggests that granting operators a high degree of autonomy will have a positive impact on these same variables. To test these hypotheses, the project employs an iterative and rigorous experimental approach that unfolds in four phases. First, an online study involving 500 participants will evaluate their attitudes, perceptions, behaviours in response to different levels of AI integration (ranging from no AI support to partial or full automation). Next, a laboratory experiment with 100 novice operators will delve deeper using a multi-method approach that includes perceptual, physiological, and observational measurements. This is followed by a replication of the experiment with 30 experienced operators to enhance ecological validity by considering professional experience. Finally, a longitudinal study will be conducted over a six-month period in factory conditions, allowing us to observe the medium-term effects of AI integration through repeated measurements. The findings from these studies are expected to generate robust empirical evidence that will lead to several international peer-reviewed publications in high-impact journals. In practical terms, the project will provide concrete recommendations for industrial decision-makers and system designers on how to implement AI in a way that enhances both productivity and operator well-being. Additionally, medIAte will contribute to the enrichment of academic and professional training programs at its partner institutions by providing insights on the operational use of AI-based decision-support systems. The project is an interdisciplinary collaboration among researchers from the Institut d’Administration des Entreprises de Lyon (iaelyon), the École Nationale Supérieure des Arts et Métiers (ENSAM Cluny), Polytechnique Montréal, and the Université du Québec à Montréal (UQAM), integrating perspectives from management sciences, engineering, and work psychology.

The objective of this academic proposal is to initiate a technological breakthrough by developing a new class of locally-resonant passive acoustic materials for stealth and discretion in underwater acoustics. These metamaterials are synthesized from polymer engineering and involve strong resonant multiple-scattering phenomena within the medium. The main focus of this project is the engineering of sound/noise control in marine environment for the military and civilian areas. The potentialities of these new materials are: increasing sound absorption levels; the possible reduction of thickness of anechoic or masking coatings; a good compatibility with the industrial constraints of manufacture and use. This proposal is a strongly multidisciplinary project between three CNRS laboratories from the Bordeaux campus (experts in wave-physics, soft-matter and microfluidics techniques) and a major industrial group specialized in naval defence. The academic partners have more than 7 years of joint research on the topic of metamaterials (design and manufacturing) and DCNS has recently had a CIFRE/DGA action with one of them. This long collaboration coupled with a geographical proximity and a complementarity of skills up to the industrial level, is a key point to meet the materials and acoustics challenges of the project. The materials challenge. These inclusion-type materials will incorporate sub-millimetric porous micro-resonators (made by emulsions or microfluidics) dispersed in an elastomer matrix adapted to the marine environment. Using "dense" and "resonant" inclusions must make it possible to address two major challenges for better performance of the boat-hull coverings: resistance to hydrostatic pressures during immersion; higher absorption properties due to the resonant multiple scattering. The wave physics challenge concerns the modeling and the experimental proof of the functions and characteristics sought for the synthesized subwavelength materials/structures. An important phase for ultrasonic characterization under mechanical loading of the laboratory samples will indicate the performance of the latter, in particular in terms of absorption. Contextualized experiments will be conducted to predict the anechoic/masking power of the laboratory materials, as well as acoustic measurements on metric panels placed in a pressurized tank. The industrial challenge seeks to take into account at the project outset, a number of manufacturing and use constraints that cannot be avoided by the industrial over the medium to long terms. This is why the soft-matter techniques that are easily-to-be-industrialized techniques, and the account for the hydrostatic pressure are two key elements at the heart of this exploratory-research project for naval engineering. The synoptic operational overview of PANAMA is as follows. 1. Definition of the resonant inclusion media (acoustic design) according to the targeted specifications (absorption level, frequency range, static/dynamic impedance, static loading). 2. Chemistry and synthesis of porous micro-resonators according to certain criteria: size, shape, calibration, controlled polydispersity, mass production. Incorporation of the objects in an elastomer matrix. 3. Acoustic experiments/tests (in laboratory: under loading in open air; in a conventional acoustic water-tank at atmospheric pressure; in a specialized laboratory: in a pressurized tank).

In Additive Manufacturing, Directed Energy Deposition (DED) is a promising technology that gains a growing interest in industry. An essential feature of this process its rapid fabrication capability, even for large-size parts. However, generating good material deposition trajectories remain a huge challenge that CAM software often fail to correctly deal with. The KAM4AM project aims at developing a software for DED manufacturing, based on the proven Artificial Intelligence technology of Reinforced Learning, to get a learning and adaptive CAM solution. A list of study cases from industry will help to collect the typologies of parts as well as technical and scientific issues related to DED technology. This data, combined with research cases, will enable to define the objectives and the functions of the learning environment that needs to be created. The main research challenges are (1) to design a problem-independent reward system, based on expert rules of the DED domain, (2) to develop a phenomenological model of the DED process, fast enough for allowing the numerous iterations required for the learning process. A last step consists in a thorough test of the generated trajectories, followed by the integration of these trajectories into Esprit Additive software.

Heat pump systems provide cost-effective solutions for recovering heat from various sources for use in a variety of industrial, commercial, or residential applications. As energy costs continue to rise, it is imperative to save energy and improve overall energy efficiency. Improving the performance and reliability of heat pumps, as well as their environmental impact, is an ongoing concern, and it is forced by the European Community. Most compressors in heat pump installations are centrifugal compressors, but for future large applications, axial-flow compressors might be solutions as well. Although compressors are a proven technology, the use of heavy organic vapors as working fluids increases the impact of blade roughness and non-ideal compressibility effects. As a result, low-fidelity numerical tools used in the design process can lead to large discrepancies in predicted performance, while high-fidelity models are prohibitively costly for industrial use. High-fidelity numerical simulations and experiments will be employed to improve our fundamental understanding of flows of complex fluids in compressors. Experiments in a compressor test and wind tunnel facility will be fused with high-fidelity simulations based on a compressible flow solver to learn data-driven models. The latter will be implemented and tested in low-fidelity codes against measurements. Axial-flow compressors, which are currently less used in the heat pump context, will also be studied experimentally and numerically. Utilizing a compressor cascade, secondary flow, shock wave boundary layer interactions, and roughness effects will be investigated in detail. Multifidelity optimization will be implemented to optimize compressor blades, in particular the secondary flows responsible for a significant proportion of losses. Axial and centrifugal compressors optimized to take advantage of the real gas effects associated with the use of organic vapor as a working fluid will then be tested in the CLOWT wind tunnel. The synergy between the numerical and experimental team has already been demonstrated in a previous ANR-DFG project Regal-ORC (13 articles in peer-reviewed journals and 20 contributions to international conference proceedings) and can unravel the physical mechanisms driving the energy efficiency of axial or centrifugal compressors, the key element in heat pump systems.

Thermal energy storage (TES) is a key element, for effective and efficient generation and utilization of heat where heat supply and heat demand do not match spatially and in time. Applications of TES systems are expected in several areas such as: solar heating and cooling of buildings; power generation using thermal conversion processes; seasonal storage in combination with district heating systems... The most common experimental ways to improve the heat transfer through heat storage materials, usually the so-called Phase Change Materials (PCMs), is by adding either extended surfaces or encapsulated phase change materials. In this project, our objective is to propose and analyze efficient and easy-to-implement algorithms for the simulation of the phase change process and therefore the optimization of the capsule location in order to increase the speed of the loading/unloading process. For this purpose, solutions using full order models and reduced order models (ROMs) will be performed. To achieve our objectives, it will be necessary : 1) to propose a solver to simulate the phase change in a medium loaded with capsules whose role is to enhance the transfer. A solver developed by I2M already exists to simulate the phase change without the capsules, it is therefore necessary to numerically model their presence and their interactions with the carrier PCM ; 2) to develop mathematical and algorithmic optimization tools to improve the performance and the behavior of the PCM ; 3) to build efficient and robust parametric ROMs in order to considerably reduce the computation time of the optimization method ; and finally 4) to validate the developed tools through benchmarks and comparisons with laboratory experiments.