Search

April 15th, 2019: Notre-Dame Cathedral in Paris was burning, the spire collapsed on the nave, vaults crumbled and most of the timber roof was gone. In the post-disaster context, the authenticity and the monitoring of the archaeological remains are crucial for their potential reuse during reconstruction. This paper analyzes the collapsed transverse arch from the nave of Notre-Dame as a case study of reconstruction, using the digital twin framework. We propose four facets for the digital twin experiment—physical anastylosis, reverse engineering, spatio-temporal tracking of assets, and operational research—that are described in detail, while being assembled to support a hybrid reconstruction hypothesis. The digital twin can realize the parallel unfolding of physical-native and digital-native processes, while acquiring and storing heterogeneous information as semantically structured data. The results demonstrate that the proposed modeling method facilitates the formalization and validation of the reconstruction problem and increases solutions performances. As result, we present a digital twin framework application ranging from acquisition to data processing that informs a successful hybrid reconstruction hypothesis.

International audience; The FeMo cofactor (FeMoco) of Mo nitrogenase is responsible for reducing dinitrogen to ammonia, but it requires the addition of 3-4 e-/H+ pairs before N2 even binds. A binding site at the Fe2/Fe3/Fe6/Fe7 face of the cofactor has long been suggested based on mutation studies, with Fe2 or Fe6 nowadays being primarily discussed as possibilities. However, the nature of N2 binding to the cofactor is enigmatic as the metal ions are coordinatively saturated in the resting state with no obvious binding site. Furthermore, the cofactor consists of high-spin Fe(II)/Fe(III) ions (antiferromagnetically coupled but also mixed-valence delocalized), which are not known to bind N2. This suggests that an Fe binding site with a different molecular and electronic structure than the resting state must be responsible for the experimentally known N2 binding in the E4 state of FeMoco. We have systematically studied N2 binding to Fe2 and Fe6 sites of FeMoco at the broken-symmetry QM/MM level as a function of the redox-, spin-, and protonation state of the cofactor. The local and global electronic structure changes to the cofactor taking place during redox events, protonation, Fe-S cleavage, hydride formation, and N2 coordination are systematically analyzed. Localized orbital and quasi-restricted orbital analysis via diamagnetic substitution is used to get insights into the local single Fe ion electronic structure in various states of FeMoco. A few factors emerge as essential to N2 binding in the calculations: spin-pairing of dxz and dyz orbitals of the N2-binding Fe ion, a coordination change at the N2-binding Fe ion aided by a hemilabile protonated sulfur, and finally hydride ligation. The results show that N2 binding to E0, E1, and E2 models is generally unfavorable, likely due to the high-energy cost of stabilizing the necessary spin paired electronic structure of the N2-binding Fe ion in a ligand environment that clearly favors high-spin states. The results for models of E4, however, suggest a feasible model for why N2 binding occurs experimentally in the E4 state. E4 models with two bridging hydrides between Fe2 and Fe6 show much more favorable N2 binding than other models. When two hydrides coordinate to the same Fe ion, an increased ligand-field splitting due to octahedral coordination at either Fe2 or Fe6 is found. This altered ligand field makes it easier for the Fe ion to acquire a spin-paired electronic structure with doubly occupied dxz and dyz orbitals that backbond to a terminal N2 ligand. Within this model for N2 binding, both Fe2 and Fe6 emerge as possible binding site scenarios. Due to steric effects involving the His195 residue, affecting both the N2 ligand and the terminal SH- group, distinguishing between Fe2 and Fe6 is difficult; furthermore, the binding depends on the protonation state of His195.

International audience; Despite the wide availability of large motion databases, and recent advances in exploiting them to create tailored contents, artists still spend tremendous efforts in editing and adapting existing character animations to new contexts. This motion editing problem is a well-addressed research topic. Yet only a few approaches have considered the influence of the camera angle, and the resulting visual features it yields, as a mean to control and warp a character animation. This paper proposes the design of viewpoint-dependent motion warping units that perform subtle updates on animations through the specification of visual motion features such as visibility, or spatial extent. We express the problem as a specific case of visual servoing, where the warping of a given character motion is regulated by a number of visual features to enforce. Results demonstrate the relevance of the approach for different motion editing tasks and its potential to empower virtual characters with attention-aware communication abilities.

International audience; This paper presents two methods to evaluate and draw distinction among general cooperative and aggressive strategic behavior in the iterated prisoner’s dilemma (IPD). The first method identifies classes of strategies, clusters them into subclasses based on their cooperative or aggressive disposition, and then comprehensively assess their behavior. The second method relies on applying transformative manipulations that soften or harden strategies of a given set and compare them with the results obtained with the initial set. Although examined classes of initial strategies are very different, the results are stable and convergent. While our findings agree with the classical analyses of the IPD, the results are qualified and precise. Furthermore, the reproducibility of prior results via new experimental methods offer experimental evidence. The proposed comparative strategy analysis method is a new instrument that can have broad implications for analyzing games beyond the prisoner’s dilemma.

This paper presents a computational approach for finding the optimal shapes of peristaltic pumps transporting rigid particles in Stokes flow. In particular, we consider shapes that minimize the rate of energy dissipation while pumping a prescribed volume of fluid, number of particles and/or distance traversed by the particles over a set time period. Our approach relies on a recently developed fast and accurate boundary integral solver for simulating multiphase flows through periodic geometries of arbitrary shapes. In order to fully capitalize on the dimensionality reduction feature of the boundary integral methods, shape sensitivities must ideally involve evaluating the physical variables on the particle or pump boundaries only. We show that this can indeed be accomplished owing to the linearity of Stokes flow. The forward problem solves for the particle motion in a slip-driven pipe flow while the adjoint problems in our construction solve quasi-static Dirichlet boundary value problems backwards in time, retracing the particle evolution. The shape sensitivities simply depend on the solution of one forward and one adjoint (for each shape functional) problems. We validate these analytic shape derivative formulas by comparing against finite-difference based gradients and present several examples showcasing optimal pump shapes under various constraints., Comment: 24 pages, 8 figures

International audience; To face market volatility, reconfigurable manufacturing systems (RMS) aim to efficiently and cost-effectively react to changes. We focus on one characteristic of RMS: the scalability (ability to adapt the volume of throughput). In the literature, the only few indicators for scalability are not always formally defined and usually only consider a partial view of scalability. Moreover, most of them are actually more suited for the configuration planning rather than for the design. However, the design of the RMS has a high impact on its scalability. We propose the first combinatorial definition of this problem and a new measure to fully assess the scalability of a system at the design phase. This measure, based on a multi-objective approach, can assess the scalability of single-product manufacturing systems, analysing all configurations that it can implement. We present numerical experiments to compare this indicator with a state-of-the-art scalability indicator and with some classical production line design indicators, and we show that future research should focus on scalability as a specific criterion to optimise during the design of an RMS. In addition, the results obtained allow us to infer some managerial insights on the best levers to use when performing a reconfiguration for scalability purpose.

International audience; Manufacturing systems are socio-technical systems, with explicit interactions between humans and technologies in shared workspaces. These shared workspaces could also be called hybrid collaborative manufacturing systems, which involve workers as well as technological equipment and combine the benefits of human workers and new Industry 4.0 technologies, such systems are particularly useful in a context requiring flexibility and adaptability. Furthermore, the new Industry 5.0 approach has the objective to shift toward more human-centric and resilient manufacturing systems. The key problems to solve in the design of collaborative manufacturing systems are the combinatorial assembly line balancing problem and the equipment selection problem. An efficient and sustainable line requires a cost-effective choice of equipment while improving the ergonomics and the safety of workers. Both decisions of balancing workload and the assignment of equipment impact the ergonomics of a collaborative system and present conflicting criteria. To this end, we propose a multi-objective approach, the objectives are the optimisation of the investment costs and the ergonomics with a fatigue and recovery criterion. We propose to linearise the fatigue and recovery to formulate a new Mixed Integer Linear Programming formulation. We developed an exact multi-objective solving algorithm based on the ϵ-constraint to obtain the trade-off between these objectives. We conducted numerical experiments with different instances from the literature with promising results for instances with up to 45 operations. Finally, we discuss insightful managerial conclusions and future research perspectives.

International audience; Long-Range Frequency Hopping Spread Spectrum (LR-FHSS) is a recent modulation designed for communications from low-power ground end-devices to Low-Earth Orbit (LEO) satellites. To decode a frame, an LR-FHSS gateway must receive at least one header replica and a large proportion of payload fragments. However, LR-FHSS headers will likely be lost when there are many concurrent transmissions. In this paper, we motivate the header loss problem with an analytical model, propose a linear programming model to extract headerless frames and design a cost-effective sliding window heuristic. Simulation results show that our approach exhibits near-optimal headerless detection and extraction results while ensuring a low computational cost. The proposed method is, therefore, suitable for future LR-FHSS gateways located onboard resource-constrained LEO satellites.

International audience; Social simulation as a scientific tool requires the development of credible behaviours for agents simulating the behaviour of human actors. One way to improve the credibility of the simulations obtained is to integrate cognitive, emotional and social dimensions into agents' decision-making. However, developing agents with these dimensions can be complicated for social science researchers who do not have expert knowledge of computer programming. The BEN architecture (Behavior with Emotions and Norms) integrates emotional and social dimensions into decision making based on a BDI architecture for an application in the field of social simulations. This modular architecture is based on a formalization of the cognitive, affective and social dimensions. BEN is implemented in the GAMA multi-agent modeling and simulation platform and used on an example case of nightclub evacuation.; La simulation sociale en tant qu'outil scientifique néces-site le développement de comportements crédibles pour les agents simulant le comportement d'acteurs humains. Une façon d'améliorer la crédibilité des simulations ob-tenues est d'intégrer des dimensions cognitive, affective et sociales dans la prise de décision des agents. Pour autant, développer des agents avec ces dimensions peut s'avérer compliqué si l'on veut rester accessible à des chercheurs qui ne sont pas experts en informatique. L'architecture BEN (Behavior with Emotions and Norms) intègre des dimensions affectives et sociales à une prise de décisions ba-sée sur une architecture BDI pour la réalisation de simulations sociales. Cette architecture modulaire s'appuie sur une formalisation des dimensions cognitives, affectives et sociales. BEN est implémenté dans la plateforme de modé-lisation et de simulation multi-agent GAMA et est ici illus-trée sur un cas d'évacuation d'urgence.

International audience; We propose a multi-physics numerical model for a self-healing ceramic matrix mini-composite under tensile load. Crack averaged PDEs are proposed for the transport of oxygen and of all the chemical species involved in the healing process and studied in the dimensionless form to perform the most appropriate discretization choices concerning time integration, and boundary conditions. Concerning the fibres’ degradation, a slow crack growth model explicitly dependent on the environmental parameters is calibrated using a particular exact solution and integrated numerically in the general case. The tow failure results from the statistical distribution of the fibres’ initial strength, the slow crack growth kinetics, and the load transfer following fibres breakage. The lifetime prediction capabilities of the model, as well as the effect of temperature, spatial variation of the statistical distribution of fibres strength, and applied load, are investigated highlighting the influence of the diffusion/reaction processes (healing) on the fibre breakage scenarios.

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers: Medtr4g107930: CP3; Medtr4g079770: CP4; Medtr5g022560: CP2; Medtr4g079470: CP5; TC106667: Actine; Medtr1g099310.1: PR8; Medtr4g120970.1/Medtr6g033450.1: PR10; Medtr5g010640.1: PR5.3; Medtr8g096910.1: PR5.6; Medtr5g088770.1: PHYTOCYSTATIN32; Medtr2g026040.1: PHYTOCYSTATIN5; Medtr4g085800: DNF2; Medtr3g079850: SymCRK; MtrunA17_Chr7g0239441: RSD. Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers: Medtr4g107930: CP3; Medtr4g079770: CP4; Medtr5g022560: CP2; Medtr4g079470: CP5; TC106667: Actine; Medtr1g099310.1: PR8; Medtr4g120970.1/Medtr6g033450.1: PR10; Medtr5g010640.1: PR5.3; Medtr8g096910.1: PR5.6; Medtr5g088770.1: PHYTOCYSTATIN32; Medtr2g026040.1: PHYTOCYSTATIN5; Medtr4g085800: DNF2; Medtr3g079850: SymCRK; MtrunA17_Chr7g0239441: RSD.; International audience; Medicago (Medicago truncatula) establishes a symbiosis with the rhizobia Sinorhizobium sp, resulting in the formation of nodules where the bacteria fix atmospheric nitrogen. The loss of immunity repression or early senescence activation compromises symbiont survival and leads to the formation of nonfunctional nodules (fix−). Despite many studies exploring an overlap between immunity and senescence responses outside the nodule context, the relationship between these processes in the nodule remains poorly understood. To investigate this phenomenon, we selected and characterized three Medicago mutants developing fix− nodules and showing senescence responses. Analysis of specific defense (PATHOGENESIS-RELATED PROTEIN) or senescence (CYSTEINE PROTEASE) marker expression demonstrated that senescence and immunity seem to be antagonistic in fix− nodules. The growth of senescence mutants on non-sterile (sand/perlite) substrate instead of sterile in vitro conditions decreased nodule senescence and enhanced defense, indicating that environment can affect the immunity/senescence balance. The application of wounding stress on wild-type (WT) fix+ nodules led to the death of intracellular rhizobia and associated with co-stimulation of defense and senescence markers, indicating that in fix+ nodules the relationship between the two processes switches from opposite to synergistic to control symbiont survival during response to the stress. Our data show that the immune response in stressed WT nodules is linked to the repression of DEFECTIVE IN NITROGEN FIXATION 2 (DNF2), Symbiotic CYSTEINE-RICH RECEPTOR-LIKE KINASE (SymCRK), and REGULATOR OF SYMBIOSOME DIFFERENTIATION (RSD), key genes involved in symbiotic immunity suppression. This study provides insight to understand the links between senescence and immunity in Medicago nodules. Analyses of Medicago mutants with non-functional nodules highlight the relationship and mechanisms controlling the establishment of the immune and senescence programs during nodule organogenesis.

We propose a thorough comparison of polynomial chaos expansion (PCE) for indicator functions of the form 1 c≤X for some threshold parameter c ∈ R and a random variable X associated with classical orthogonal polynomials. We provide tight global and localized L2 estimates for the resulting truncation of the PCE and numerical experiments support the tightness of the error estimates. We also compare the theoretical and numerical accuracy of PCE when extra quantile/probability transforms are applied, revealing different optimal choices according to the value of c in the center and the tails of the distribution of X.

International audience; Model merging conflicts occur when different stakeholders aim to integrate their contradicting changes that are applied concurrently to update software models. We conduct an extensive systematic mapping study on conflict management techniques and relevant collaboration attributes to the versioning and merging models from 2001 to the middle of 2021. This study follows the standard guidelines within the software engineering domain. We analyzed a total of 105 articles extracted from an initial pool of more than 1800 articles to infer a taxonomy for conflict management techniques. We use this taxonomy to classify existing approaches to understand characteristics, shortcomings, and challenges on conflict management techniques in merging models. It also provides a solid foundation for future work in this area. We show that syntactic conflicts are the most studied type and that the top three popular conflict detection techniques are constraint violation, change overlapping, and pattern matching. We observe the lack of a comprehensive state-of-the-art comparison between academic or industrial tools, as well as the need for real-world case studies. Finally, we show that recent trends have focused on online collaboration, where teams of stakeholders work on large-scale models.

International audience; The Internet of Things (IoT) aims for connecting Anything, Anywhere, Anytime (AAA). This assumption brings about several software engineering challenges that constitute a serious obstacle to its wider adoption. The main feature of the Internet of Things (IoT) is genericity w.r.t the variability of software and hardware technologies. Model-Driven Engineering (MDE) is a paradigm that advocates using models to address software engineering problems. It can help to meet the genericity of the IoT from a software engineering perspective. Existing MDE approaches for the IoT focus only on modeling the internal behavior of things but lack a comprehensive approach dedicated to network modeling. In the present paper, we introduce a network-oriented methodology based on MDE to unify the IoT's heterogeneous concepts. Fundamentally, we avoid the intrinsic heterogeneity of the IoT by separating the network's specification (i.e., the things, the communication scheme, and its constraints) from its concrete implementation (i.e., the low-level artifacts such as source code and documentation). Technically, the methodology relies on a model-based Domain-Specific Language (DSL) and a code generator. The former enables the modeling of the network's specification, and the latter provides a procedure to generate the low-level artifacts from this specification. Our results show that this methodology makes IoT's software engineering more rigorous, helps prevent bugs earlier, and saves time.

The determination of the number of mixture components (the order) of a finite mixture model has been an enduring problem in statistical inference. We prove that the closed testing principle leads to a sequential testing procedure (STP) that allows for confidence statements to be made regarding the order of a finite mixture model. We construct finite sample tests, via data splitting and data swapping, for use in the STP, and we prove that such tests are consistent against fixed alternatives. Simulation studies and real data examples are used to demonstrate the performance of the finite sample tests-based STP, yielding practical recommendations of their use as confidence estimators in combination with point estimates such as the Akaike information or Bayesian information criteria. In addition, we demonstrate that a modification of the STP yields a method that consistently selects the order of a finite mixture model, in the asymptotic sense. Our STP is not only applicable for order selection of finite mixture models, but is also useful for making confidence statements regarding any sequence of nested models.

International audience; This research reports on the recent development of black-box optimization methods based on single-step deep reinforcement learning (DRL) and their conceptual similarity to evolution strategy (ES) techniques. It formally introduces policy-based optimization (PBO), a policy-gradient method that relies on a policy network to describe the density function of its forthcoming evaluations, and uses covariance estimation to steer the policy improvement process in the right direction. The specifics of the PBO algorithm are detailed, and the connection to evolutionary strategies (especially covariance matrix adaptation evolutionary strategy) is discussed. Relevance is assessed by benchmarking PBO against classical ES techniques on analytic functions minimization problems, and by optimizing various parametric control laws intended for the Lorenz attractor. Given the scarce existing literature on the topic, this contribution definitely establishes PBO as a valid, versatile black-box optimization technique, and opens the way to multiple future improvements building on the inherent flexibility of the neural networks approach.

International audience; Thermal storages are components used in energy systems, such as district heating networks or thermal power plants, in order to decouple the supply of heat from its use. Usage rate of monophasic thermocline storages is highly dependent on the thermal gradient zone inside the fluid, also named thermocline. While thermal stratification results of a formation phase followed by a degradation phase, the early stages of thermocline establishment is primarily responsible for its thickness. CFD allow to consider the multiple physical phenomena involved during the thermocline formation, in particular the buoyancy effects. These effects are usually considered by selecting either a variable density with respect to the temperature or a constant one by using the commonly used (first-order) Boussinesq approximation. However, the former approach implies an increased computational cost, and the latter is only valid for an unclear validity range of temperature difference. Hence, this article suggests the use of a second-order Boussinesq approximation, coupled with a RANS turbulence approach, to better account for buoyancy effects in a turbulent water flow submitted to a large temperature differences. CFD results obtained with a quadratic Boussinesq approximation are similar to the one obtained with a variable density but with a computation time reduced by half. This approach is applied to the issue of reducing the thermocline thickness during its creation and the impact of linear flow rate ramps is assessed on both a uniform and initially stratified storage. On an initially cold tank, results show that the longer the ramp time, the thinner the thermocline. In contrast, on the initially stratified tank tested, a gradual injection shows no significant reduction of the thermocline thickness. This can be relevant when performing storage management enhancement.

Welding defect detection is still often performed by the human visual inspection or with non destructive tests. These quality inspections methods can be time consuming and can have an important error rate. In this paper, we propose an approach for the detection of welding faults through the detection of abnormal subsequences of the welding voltage signal. The approach is based on the One-Class SVM with distance substitution kernels. The One Class-SVM has been used in many works for the detection of abnormal subsequences. However, often after transforming the subsequences into a set of features. Nevertheless, finding the relevant features for anomaly detection may be challenging. Dealing with the raw subsequences in distance-based approaches, on the other hand, are known to be effective and can generalize well to different problems but they often suffer from high computational cost, which may restrict their application, especially with the need of real time predictions in an industrial context. We show in this work that the One-Class SVM can be successfully used directly with the raw subsequences. This is achieved by employing distance substitution kernels. These class of kernels has not yet gained widespread adoption for time series anomaly detection. The results show that the approach is both accurate and fast, which makes it more suitable for real-time welding monitoring. We further propose an approach for automatic diagnosis of welding defects.

International audience; BackgroundVirus Yellows (VY), a disease caused by several aphid-borne viruses, is a major threat to the global sugar beet production. Following the ban of neonicotinoid-based seed treatments against aphids in Europe, increased efforts are needed to monitor and forecast aphid population spread during the sugar beet growing season. In particular, predicting aphid flight seasonal activity could allow anticipation of the timing and intensity of crop colonisation and contribute to the proper implementation of management methods. Forecasts should be made early enough to assess risk, but can be updated as the season progresses to refine management. Based on a long-term suction-trap dataset gathered between 1978 to 2014, we built and evaluated a set of models to predict the flight activity features of the main VY vector, Myzus persicae, at any location in the French sugar beet production area (circa 4.105 ha). Flight onset dates, length of flight period and cumulative abundance of flying aphids were predicted using climatic and land-use predictors as well as geographical position.ResultsOur predictions outperformed current models published in the literature. The importance of the predictor variables varied according to the predicted flight feature but winter and early spring temperature always played a major role. Forecasts based on temperature were made more accurate by adding predictors related to aphid winter reservoirs. In addition, updating the model parameters to take advantage of new weather data acquired during the season improved the flight forecast.ConclusionOur models can be used as a tool for the mitigation in sugar beet crops.

National audience; Le marketing est un outil indispensable pour les entreprises qui souhaitent faire connaître un nouveau produit ou fidéliser des consommateurs. Néanmoins, cet outil est influencé par de nombreux facteurs qui rendent sa compréhension complexe et difficilement mesurable dans sa globalité. Nous proposons dans cet article un modèle centré individus qui nous permet de mesurer par exemple les effets des guerres des prix sur la fidélité des consommateurs, l'augmentation des ventes du produit en promotion, la baisse des ventes des concurrents ou les effets sur les consommateurs de la répétition des promotions. Ce modèle multi-agents, à la fois simple et puissant, offre la possibilité de quantifier ces phénomènes et constitue une avancée dans la compréhension et la calibration de nouvelles stratégies marketing.

National audience; Dans cet article, nous proposons une approche multi-agent pour résoudre le problème du covoiturage dynamique multi-saut. Dans notre système, les passagers et les conducteurs sont représentés comme des agents autonomes et rationnels en perpétuelle interaction pour satisfaire leurs propres objectifs comme leur temps d'attente ou leur temps de trajet par exemple. Dans la solution proposée, les agents conducteurs et passagers ont une perception modélisée dynamiquement en utilisant des R-Arbres. Nous modélisons leurs préférences en matière de détour et de trajet et montrons l'impact de celles-ci sur la résolution d'une instance de covoiturage dynamique. Les résultats présentés montrent que notre système permet de traiter dynamiquement des requêtes complexes de passagers tout en minimisant l'impact du partage de trajet pour les conducteurs, et ce, pour un large spectre de préférences et de comportements.

National audience; Simulation of multi-robots systems require the integration of both the robotic and network components. Leveraging the existing tools from each community seems evident, but merging two completely different simulators proves challenging. This paper presents the state of the art in co-simulators that tackle this problematic, and provide details on the challenges arising when one wants to create an efficient co-simulator.; La simulation de systèmes multi-robots nécessite l'intégration des composantes robotique et réseau au sein d'un même environnement de simulation. La réutilisation des outils existants dans les communautés robotique et réseau semble être une évidence, mais la fusion de deux simulateurs présente des défis structurels qu'il faut surmonter afin d'obtenir une simulation "réaliste" d'un système multi-robots. Dans cet état de l'art, nous présentons les co-simulateurs qui abordent cette problématique et détaillons les défis auxquels il faut répondre pour la conception d'un co-simulateur performant.

International audience; Nous étudions ici le problème de l'allocation de jobs concurrents, composés de tâches situées, sous-jacent au déploiement distribué du patron de conception MapReduce sur une grappe de serveurs. Afin de mettre en oeuvre notre stratégie multi-agents qui vise à minimiser le délai moyen de réalisation des jobs (flowtime), nous proposons une architecture composite d'agent qui permet la concurrence des négociations et des consommations. Nos expérimentations montrent que, lorsqu'elle est exécutée en continue lors du processus de consommation, notre stratégie de réallocation : (1) améliore le flowtime ; (2) ne pénalise pas la consommation ; (3) est robuste aux aléas d'exécution.

National audience; Les sites de questions-réponses permettent aux utilisateurs, qui partagent les mêmes intérêts, d'échanger des connaissances dans le cadre d'un processus d'apprentissage continu. En particulier, la plateforme Stack Overflow soutient des communautés de pratique où : (a) la majorité du contenu est produit par un petit nombre de membres très actifs ; (b) les réponses aux questions les plus récurrentes sont rapidement capitalisées. Nous proposons un modèle centré individus d'une communauté de pratique présentant une dynamique sociale conforme à ces faits stylisés et qui explique le phénomène de sécession des experts.

International audience; The development of a capnometry wristband is of great interest for monitoring patients at home. We consider a new architecture in which a non-dispersive infrared (NDIR) optical measurement is located close to the skin surface and is combined with an open chamber principle with a continuous circulation of air flow in the collection cell. We propose a model for the temporal dynamics of the carbon dioxide exchange between the blood and the gas channel inside the device. The transport of carbon dioxide is modeled by convection–diffusion equations. We consider four compartments: blood, skin, the measurement cell and the collection cell. We introduce the state-space equations and the associated transition matrix associated with a Markovian model. We define an augmented system by combining a first-order autoregressive model describing the supply of carbon dioxide concentration in the blood compartment and its inertial resistance to change. We propose to use a Kalman filter to estimate the carbon dioxide concentration in the blood vessels recursively over time and thus monitor arterial carbon dioxide blood pressure in real time. Four performance factors with respect to the dynamic quantification of the CO2 blood concentration are considered, and a simulation is carried out based on data from a previous clinical study. These demonstrate the feasibility of such a technological concept.

International audience; Eulalie is an operational, open-source, and free information system dedicated to electroacoustic music for the documentation and preservation of works. It is based on the Dorémus ontology, which we adapt to the speci c eld of electroacoustic music, and on its implementation in a database (Heurist, used together with Nakala to host the documents). The project was born in the context of Art Zoyd Studios, a center for musical creation. One of the missions of this center is to preserve the repertoire of the th-and st-century pieces involving electronics. The principles of interoperability and open data have guided the di erent choices made throughout the project. Thus, not only is it possible to make the content of the Art Zoyd Studios information database interact with other resources on music, but the Eulalie system itself can be used by other institutions facing similar problems.; Eulalie est un système documentaire opérationnel, distribué en open source et gratuit, dédié à la musique électroacoustique, à des fins de documentation et de conservation des œuvres. Il repose sur l’ontologie Dorémus, dont nous proposons une adaptation à ce champ musical spécifique, ainsi que sur son implémentation dans une base de données (Heurist, utilisé avec Nakala pour l’hébergement des documents). Le projet est né dans le contexte d’Art Zoyd Studios, centre de création musicale dont l’une des missions est la préservation et l’entrée au répertoire de pièces du xxe et du xxie siècles utilisant l’électronique. Les principes d’interopérabilité et d’ouverture des données ont guidé les différents choix effectués tout au long du projet. Ainsi, non seulement il est possible de faire dialoguer le contenu de la base documentaire d’Art Zoyd Studios avec d’autres ressources sur la musique, mais le système Eulalie lui-même peut être utilisé par d’autres institutions rencontrant des problèmes similaires.

International audience; Networks are now widely used to represent, quantify and model animal behaviour. These approaches have proved valuable in linking individual behaviours to emergent population level patterns and quantifying the implications of these population structures for wider ecological and evolutionary processes. However, there are diverse conceptual representations of network data and choosing the right tool to answer a particular question can be challenging. Here I provide an overview of different network representations, highlighting their potential applications in behavioural ecology and drawing attention to key resources to help with their implementation. My aim is to provide an accessible guide that helps behavioural ecologists take full advantage of the potential of the different ways in which their data can be used to generate social other networks.

The spatiotemporal resolution of Partial Differential Equations (PDEs) plays important roles in the mathematical description of the world's physical phenomena. In general, scientists and engineers solve PDEs numerically by the use of computationally demanding solvers. Recently, deep learning algorithms have emerged as a viable alternative for obtaining fast solutions for PDEs. Models are usually trained on synthetic data generated by solvers, stored on disk and read back for training. This paper advocates that relying on a traditional static dataset to train these models does not allow the full benefit of the solver to be used as a data generator. It proposes an open source online training framework for deep surrogate models. The framework implements several levels of parallelism focused on simultaneously generating numerical simulations and training deep neural networks. This approach suppresses the I/O and storage bottleneck associated with disk-loaded datasets, and opens the way to training on significantly larger datasets. Experiments compare the offline and online training of four surrogate models, including state-of-the-art architectures. Results indicate that exposing deep surrogate models to more dataset diversity, up to hundreds of GB, can increase model generalization capabilities. Fully connected neural networks, Fourier Neural Operator (FNO), and Message Passing PDE Solver prediction accuracy is improved by 68%, 16% and 7%, respectively.

International audience

International audience; Reconfigurable manufacturing systems (RMS) intend to bridge the gap between dedicated and flexible manufacturing systems. If the literature is mainly focused on the design step and tactical planning of such systems, few research projects have addressed scheduling at the operational level. While setup times may occur in flexible manufacturing systems, reconfiguration times considered in RMS may affect several resources at once, and hence require specific modelling and solving approaches to be considered. This paper first formalises the problem at hand through integer linear programming. An iterative search method is then provided to obtain solutions to larger-scale instances. Results obtained on generated instances show that managing even few possible configurations can yield significant improvements in solutions’ quality. Meanwhile, the extended search space implied by the increase in available configurations hinders the convergence to a good solution in a reasonable computation time, which suggests further investigations.

International audience; En raison de leur utilité et de leur commodité, les modèles pluie-débit sont largement utilisés dans la recherche et l'ingénierie. Les applications de ces modèles vont de l'estimation des risques d'inondation à la gestion des ressources en eau et des sécheresses. L’INRAE a développé une famille de modèles hydrologiques conceptuels au cours des 30 dernières années, avec pour objectif principal de concevoir des modèles aussi efficaces que possible en termes de simulation du débit, en s’assurant qu’ils sont applicables à une large diversité de bassins versants, tout en restant parcimonieux en termes de données d’entrée. Afin de rendre disponibles ces modèles hydrologiques, un package nommé airGR a été développé il y a 10 ans. Il propose également des outils complémentaires (module d’accumulation et de fonte de la neige, algorithme de calage, calcul de critères de performance, etc.). Ces dernières années, un ensemble de packages s’est développé autour d'airGR. Le package airGRteaching a ainsi été conçu pour des applications simples et nécessite des connaissances limitées en programmation, et il offre une interface graphique particulièrement utile à des fins éducatives. La package airGRdatasets fournit des jeux de données permettant notamment de réaliser des exercices proposés avec le package airGRteaching sur 19 bassins versants. Le package airGRiwrm permet l’intégration des influences anthropiques dans un modèle hydrologique semi-distribué spatialement (barrages, prélèvements pour l’agriculture, etc.) et les règles de gestion associées. Enfin, le package airGRdatassim permet d'assimiler des données d'observation pour contraindre les modèles GR, notamment pour améliorer les prévisions des crues. En plus de ces packages, la airGRgalaxy comprend deux applications web disponibles sur sunshine.inrae.fr : airGRmaps qui fournit des paramètres régionalisés en France pour les modèles journalier GR à partir de coordonnées géographiques, et une démo de l'interface d’airGRteaching. Cette communication vise à présenter l’ensemble de ces outils.

International audience; The recent rise of interest in distributed applications has highlighted the importance of effective information dissemination. The challenge lies in the fact that nodes in a distributed system are not necessarily synchronized, and may fail at any time. This has led to the emergence of randomized rumor spreading protocols, such as push and pull protocols, which have been studied extensively. The k-pull operation, which allows an uninformed node to ask for the rumor from a fixed number of other nodes in parallel, has been proposed to improve the pull algorithm's effectiveness. This paper presents and studies the performance of the k-pull operation in the presence of a certain fraction f of non-cooperative nodes. Our goal is to understand the impact of k on the propagation of the rumor despite the presence of a fraction f of non-collaborative nodes.

International audience; Better estimation of air change rate in naturally ventilated buildings is a key for supporting passive summer thermal comfort strategies. In window aired configurations, this estimation is challenging either with measurements or building simulation methods. In this study, we describe and apply the state space modeling methodology based on the carbon dioxide concentration (CO2) mass balance equation to a simple test case with numerically generated data. We show that a two-state CO2 / Air Change Rate model is suitable. As a modeling novelty, we demonstrate the benefit of improving the formulation with a variable diffusion term for the air change rate state equation. From the test case study findings, we emphasize that a lower interior-exterior CO2 difference results in weaker performance and we list some prospects for future work.

International audience; We consider the problem of learning time-consuming functions defined over unordered sets of vectors. Such functions arise frequently, in particular in the context of networks of devices whose number is not fixed and that interact with each other. A working example is the modeling of a wind farm. Unordered sets of vectors are a mix of integer and continuous input variables suitable for functions that are permutation-invariant. The time-consuming aspect of the functions is, classically, treated by approximating them with a Gaussian process.This study addresses the problem of defining valid and efficient covariance kernels over cloudsof points in the context of Gaussian process surrogate modeling. We review methods for defining such kernels. These kernels are compared on a set of analyticalfunctions inspired from different engineering problems, such as the design of experiments andthe modeling of wind farms production. The extrapolation properties of the kernels are testedon geometrically transformed clouds.We show that modeling 2D clouds of points as supports of discrete uniform distributions should be preferred to a Gaussian representation of the clouds. A detailed investigation of the good performance of MMD-based kernels illustrates how they adapt their hyperparameters to the geometrical properties of the studied functions.

International audience; Today’s requirement for improved electrical systems increase needs for accurate, fast and versatile models. MATLAB/SIMULINK is a good candidate in terms of average simulation for accuracy, versatility and speed, but for detailed simulation (i.e. designed with the SimScape toolbox) it can be very slow, especially when multiple power converters are simulated. This issue can be solved by real time simulators based on FPGAs like OP-5607 from Opal-RT. With these real-time simulators, complex systems can be simulated with a time step of a few hundreds of nano-seconds. This article presents a new methodology in simulation fields which let researchers to interface fast response power electronics converters and slower physicals parts. This new methodology allows designers to evaluate systems at multiple time scales: minutes, seconds and milliseconds. This article emphasizes today’s possibilities of simulating a system as complex as the propulsion line of a cruise ship during an entire 30 min trip. Finally, this kind of simulation allows engineers to increase the accuracy of modeling, even to work on realistic faulty conditions.

International audience; Assessment of geometry effects affecting shock-induced cavitation within a droplet is investigated for the first time. To do this, we use a thermodynamically well-posed multiphase numerical model accounting for phase compression and expansion, which relies on a finite pressure-relaxation rate formulation and which allow for heterogeneous nucleation. These geometry effects include the shape of the transmitted wave front, which is related to the shock speed to droplet sound speed ratio, and the droplet geometry (cylindrical versus spherical). Phenomenological differences between the column and the droplet configurations are presented. In addition, the critical Mach number for cavitation appearance is determined for both cases: between M = 1.8 and M = 2 for the column, and between M = 2 and M = 2.2 for the droplet. Based on the transmitted wavefront geometry, with Mach number varying from 1.6 to 6, two cavitation regimes have been identified and the transition has been characterised: an exponentially (M < 4.38) and a linearly (M > 4.38) increasing bubble-cloud volume. On more applied aspects, we also investigate the influence of the bubble cloud on the interface disruption and compare the results against the pure liquid droplet test case. A parallel with the technique of effervescent atomization is eventually presented.

In the Humanities, the emergence of digital methods has opened up research questions to quantitative analysis. This is why HTR technology is increasingly involved in humanities research projects following precursors such as the Himanis project. However, many research teams have limited resources, either financially or in terms of their expertise in artificial intelligence. It may therefore be difficult to integrate handwritten text recognition into their project pipeline if they need to train a model or to create data from scratch. The goal here is not to explain how to build or improve a new HTR engine, nor to find a way to automatically align a preexisting corpus with an image to quickly create ground truths for training. This paper aims to help humanists easily develop an HTR model for medieval manuscripts, create and gather training data by knowing the issues underlying their choices. The objective is also to show the importance of the constitution of consistent data as a prerequisite to allow their gathering and to train efficient HTR models. We will present an overview of our work and experiment in the CREMMALab project (2021-2022), showing first how we ensure the consistency of the data and then how we have developed a generic model for medieval French manuscripts from the 13 th to the 15 th century, ready to be shared (more than 94% accuracy) and/or fine-tuned by other projects.

International audience; In the present work we propose a reduced-order method to solve the state estimation problem when nonlinear parametric time-dependent systems are at hand. The method is based on the approximation of the set of system solutions by means of a Tensor Train format. The particular structure of Tensor Train makes it possible to set up both a variational and a sequential method. Several numerical experiments are proposed to assess the behaviour of the method.

International audience; Prediction of chaotic systems relies on a floating fusion of sensor data (observations) with a numerical model to decide on a good system trajectory and to compensate non-linear feedback effects. Ensemble-based data assimilation (DA) is a major method for this concern depending on propagating an ensemble of perturbed model realizations. In this paper we develop an elastic, online, fault-tolerant and modular framework called Melissa- DA for large-scale ensemble-based DA. Melissa-DA allows elastic addition or removal of compute resources for state propagation at runtime. Dynamic load balancing based on list scheduling ensures efficient execution. Online processing of the data produced by ensemble members enables to avoid the I/O bottleneck of file-based approaches. Our implementation embeds the PDAF parallel DA engine, enabling the use of various DA methods. Melissa-DA can support extra ensemble-based DA methods by implementing the transformation of member background states into analysis states. Experiments confirm the excellent scalability of Melissa-DA, propagating 16,384 members for a regional hydrological critical zone assimilation relying on the ParFlow model on a domain with about 4M grid cells. The same use case was ported to the PDAF state-of-the-art DA framework relying on a MPI approach. A comparison with Melissa-DA at 2,500 members on 20,000 cores shows our approach is about 50% faster per assimilation cycle.

Crop multi-model ensembles (MME) have proven to be effective in increasing the accuracy of simulations in modelling experiments. However, the ability of MME to capture crop responses to changes in sowing dates and densities has not yet been investigated. These management interventions are some of the main levers for adapting cropping systems to climate change. Here, we explore the performance of a MME of 29 wheat crop models to predict the effect of changing sowing dates and rates on yield and yield components, on two sites located in a high-yielding environment in New Zealand. The experiment was conducted for 6 years and provided 50 combinations of sowing date, sowing density and growing season. We show that the MME simulates seasonal growth of wheat well under standard sowing conditions, but fails under early sowing and high sowing rates. The comparison between observed and simulated in-season fraction of intercepted photosynthetically active radiation (FIPAR) for early sown wheat shows that the MME does not capture the decrease of crop above ground biomass during winter months due to senescence. Models need to better account for tiller competition for light, nutrients, and water during vegetative growth, and early tiller senescence and tiller mortality, which are exacerbated by early sowing, high sowing densities, and warmer winter temperatures., This study was a part of the Agricultural Model Intercomparison and Improvement Project (AgMIP) Wheat Phase 4 and was supported by the French National Research Institute for Agriculture, Food (INRAE) and the International Maize and Wheat Improvement Center (CIMMYT) through the International Wheat Yield Partnership (IWYP, grant IWYP115). SD and PM acknowledge support from the metaprogram Agriculture and forestry in the face of climate change: adaptation and mitigation (CLIMAE) of INRAE. YC and FT acknowledge support from the National Natural Science Foundation of China (No. 31761143006). RPR and GBM acknowledge financial support from BARISTA project (031B0811A) through ERA-NET SusCrop under EU-FACCE JPI. KCK was funded by the German Federal Ministry of Education and Research (BMBF) through the BonaRes project ’’I4S’’ (031B0513I). AS and TG acknowledge funding by the German Federal Ministry of Education and Research (BMBF) through the BonaRes Project “Soil3” (FKZ 031B0026A). KCK and JEO were supported by the Ministry of Education, Youth and Sports of Czech Republic through SustES—Adaption strategies for sustainable ecosystem services and food security under adverse environmental conditions (CZ.02.1.01/0.0/0.0/16_019/000797). FE acknowledges support by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC 2070 – 390732324”. TKDW was funded by the German Research Foundation (DFG, Grant Agreement SFB 1253/1 2017). MAS and NS at Rothamsted Research received grant-aided support from the Biotechnology and Biological Sciences Research Council (BBSRC) through Designing Future Wheat [BB/P016855/1] and Achieving Sustainable Agricultural Systems [NE/N018125/1] jointly funded with NERC. TP and FT are supported by the DivCSA project funded by the Academy of Finland (decision no. 316215).

Few real-world data are available in patients with advanced metastatic non-small cell lung cancer (NSCLC) treated with first-line immunotherapy, particularly in those with brain metastases at treatment initiation.This was a national, retrospective, multicenter study that consecutively included all patients with PD-L1-positive (tumor proportion score ≥ 50%) advanced NSCLC who initiated first-line treatment with pembrolizumab as a single agent between May 2017 (date of availability of pembrolizumab in this indication in France) to November 22, 2019 (approval of the pembrolizumab-chemotherapy combination). Data were collected from medical records with local response assessment.The cohort included 845 patients and 176 (20.8%) had brain metastases at diagnosis. There were no significant differences in outcomes for patients with and without brain metastases: 9.2 (95% CI 5.6-15) and 8 (95% CI 6.7-9.2, p = 0.3) months for median progression-free survival (PFS) and, 29.5 (95% CI 17.2-NA) and 22 (95% CI 17.8-27.1, p = 0.3) months for median overall survival (OS), respectively. Overall response rates were 47% and 45% in patients with and without cerebral metastases. In multivariate analysis, performance status 2-4 vs. 0-1 and neutrophil-to-lymphocyte ratio ≥ 4 vs. 4 were the main independent negative factors for OS; brain metastasis was not an independent factor for OS.In this large multicenter cohort, nearly 20% of patients initiating pembrolizumab therapy for advanced NSCLC had cerebral metastases. There was no significant difference in response rates, PFS and OS between patients with and without brain metastases.

International audience; Coffee is one of the most consumed beverages in the world and is crucial in the economy of many developing countries. The search to improve coffee quality comes from many fronts, as do the many ways to measure quality and the factors that affect it. Several techniques are used to measure the different metrics to assess coffee quality, across different types of coffee samples and species, and throughout the entire process from farm to cup. In this work, we conducted a systematic mapping study of 1,470 articles to identify the aspects of quality that are the most important in the scientific literature to evaluate coffee throughout the processing chain. The study revealed that cup quality and biochemical composition are the most researched quality attributes. The main objective of the reviewed studies is the correlation between different quality measurements. The most used techniques are the analytical chemistry methods. The most studied species is Coffea arabica. The most used sample presentation is green coffee. The postharvest stage is the most researched, in which quality control receives more attention. In the preharvest stage, management practices stand out. Finally, the most used type of research was the evaluation research.

International audience; HPC systems are composed of multiple tiers of storage, from the top high performance tier (high speed SSDs) to the bottom capacitive one (tapes). File placement in such architecture is managed through prefetchers (bottom-up) and eviction policies (top-down). Most state-of-the-art work focus on the former while using algorithm flavors of LRU, LFU and FIFO for the latter. LRU was for long considered the best choice. However, recent studies has shown that the simplicity of FIFO could make it more scalable than LRU because of metadata management, and thus more adequate in several cases. In this paper, we propose a new eviction policy based on predicted files lifetimes. It is comparable to FIFO in terms of metadata overhead and simplicity (thus scalability), while giving a hit ratio comparable to LRU (or even 10% better for some tested traces). We also propose a naive multi-tier heterogeneous storage simulator implementation to evaluate such policies.

International audience; We propose an interactive animation method for the ejection of gas and ashes mixtures in volcano eruption. Our novel, layered solution combines a coarse-grain, physically-based simulation of the ejection dynamics with a consistent, procedural animation of multi-resolution details. We show that this layered model can be used to capture the two main types of ejection, namely ascending plume columns composed of rapidly rising gas carrying ash which progressively entrains more air, and pyroclastic flows which descend the slopes of the volcano depositing ash, ultimately leading to smaller plumes along their way.We validate the large-scale consistency of our model through comparison with geoscience data, and discuss both real-time visualization and off-line, realistic rendering.

With the increasing computing and networking capabilities, IoT devices and edge gateways have become part of a larger IoT–edge–cloud computing continuum, where processing and storage tasks are distributed across the whole network hierarchy, not concentrated only in the cloud. At the same time, this also introduced continuous delivery practices to the development of software components for network-connected gateways and sensing/actuating nodes. These devices are placed on end users’ premises and are characterized by continuously changing cyber-physical contexts, forcing software developers to maintain multiple application versions and frequently redeploy them on a distributed fleet of devices with respect to their current contexts. Doing this correctly and efficiently goes beyond manual capabilities and requires an intelligent and reliable automated solution. This paper describes a model-based approach to automatically assigning multiple software deployment plans to hundreds of edge gateways and connected IoT devices implemented in collaboration with a smart healthcare application provider. From a platform-specific model of an existing edge computing platform, we extract a platform-independent model that describes a list of target devices and a pool of available deployment plans. Next, we use constraint solving to automatically assign deployment plans to devices at once with respect to their specific contexts. The result is transformed back into the platform-specific model and includes a suitable deployment plan for each device, which is then consumed by our engine to deploy software components not only on edge gateways but also on their downstream IoT devices with constrained resources and connectivity. We validate the approach with a fleet deployment prototype integrated into a DevOps toolchain used by the partner application provider. Initial experiments demonstrate the viability of the approach and its usefulness in supporting DevOps for edge and IoT software development.

This paper introduces the F3ORNITS non-iterative co-simulation algorithm in which F3 stands for the 3 flexible aspects of the method: flexible polynomial order representation of coupling variables, flexible time-stepper applying variable co-simulation step size rules on subsystems allowing it and flexible scheduler orchestrating the meeting times among the subsystems and capable of asynchronousness when subsystems constraints requires it. The motivation of the F3ORNITS method is to accept any kind of co-simulation model, including any kind of subsystem, regardless on their available capabilities. Indeed, one the major problems in industry is that the subsystems usually have constraints or lack of advanced capabilities making it impossible to implement most of the advanced co-simulation algorithms on them. The method makes it possible to preserve the dynamics of the coupling constraints when necessary as well as to avoid breaking C1 smoothness at communication times, and also to adapt the co-simulation step size in a way that is robust both to zero-crossing variables (contrary to classical relative error-based criteria) and to jumps. Two test cases are presented to illustrate the robustness of the F3ORNITS method as well as its higher accuracy than the non-iterative Jacobi coupling algorithm (the most commonly used method in industry) for a smaller number of co-simulation steps.

International audience; This paper addresses modeling and simulating pedestrian trajectories when interacting with an autonomous vehicle in a shared space. Most pedestrian-vehicle interaction models are not suitable for predicting individual trajectories. Data-driven models yield accurate predictions but lack generalizability to new scenarios, usually do not run in real time and produce results that are poorly explainable. Current expert models do not deal with the diversity of possible pedestrian interactions with the vehicle in a shared space and lack microscopic validation. We propose an expert pedestrian model that combines the social force model and a new decision model for anticipating pedestrian-vehicle interactions. The proposed model integrates different observed pedestrian behaviors, as well as the behaviors of the social groups of pedestrians, in diverse interaction scenarios with a car. We calibrate the model by fitting the parameters values on a training set. We validate the model and evaluate its predictive potential through qualitative and quantitative comparisons with ground truth trajectories. The proposed model reproduces observed behaviors that have not been replicated by the social force model and outperforms the social force model at predicting pedestrian behavior around the vehicle on the used dataset. The model generates explainable and real-time trajectory predictions. Additional evaluation on a new dataset shows that the model generalizes well to new scenarios and can be applied to an autonomous vehicle embedded prediction.

IFOSMONDI iterative algorithm for implicit co-simulation of coupled physical systems (introduced by the authors in july 2019 during the Simultech conference, p.176-186) enables us to solve the nonlinear coupling function while keeping the smoothness of interfaces without introducing a delay. Moreover, it automatically adapts the size of the steps between data exchanges among the systems according to the difficulty of the solving of the coupling constraint. The latter was solved by a fixed-point algorithm in the original implementation whereas this paper introduces the JFM version (standing for Jacobian-Free Methods). Most implementations of Newton-like methods require a jacobian matrix which can be difficult to compute in the co-simulation context, except in the case where the interfaces are represented by a Zero-Order-Hold (ZOH). As far as IFOSMONDI coupling algorithm uses Hermite interpolation for smoothness enhancement (up to Third-Order-Hold), we propose hereafter a new formulation of the non-linear coupling function including both the values and the time-derivatives of the coupling variables. This formulation is well designed for solving the coupling through jacobian-free Newton type methods. Consequently, successive function evaluations consist in multiple simulations of the systems on a co-simulation time step using rollback. The orchestrator-workers structure of the algorithm enables us to combine the PETSc framework on the orchestrator side for the non-linear Newton-type solvers with the parallel integrations of the systems on the workers side thanks to MPI processes. Different nonlinear methods will be compared to one another and to the original fixed-point implementation on a newly proposed 2-systems academic test-case (mass-spring-damper type) with direct feedthrough on both sides.