HAL-EMSE
Not a member yet
9132 research outputs found
Sort by
Some Thoughts on the Reliability of Reconfigurable Manufacturing Systems
No full textInternational audienceReliability of manufacturing systems, i.e. the analysis of their failures, is a key component to ensure the productivity and competitiveness of industrial companies. In an increasingly volatile global environment, the ability to control internal factors of uncertainty, such as system components breakdowns, is crucial. The structure of Reconfigurable Manufacturing Systems (RMS) should help in this matter, and this advantage is often mentioned, but few papers actually examine this question in depth. In this article, we discuss the link between scalability, one of the main features of RMS, and reliability and we present a new analytical method to assess their reliability. This method offers an interesting alternative to existing approaches since it is suitable for the design phase, when few data are available, and it could lead to the development of more computationally tractable procedures for reliability assessment. A didactic example illustrates its use and the conclusions that can be drawn from its application as well as future research directions
A New Digital Twins Technique for Minimizing Setup Times in Reconfigurable Stations of a Mixed Model Flow Lines
No full textInternational audienceThe integration of reconfigurable manufacturing systems (RMS) with digital twin (DT) technology presents both opportunities and complexities. Since many reconfigurations in mix-model flow lines are related to product variety, and thus product arrival from upstream station contains the information for most of the reconfiguration requirements, we propose a novel scheme for the use of workstation DTs to best handle the frequent reconfigurations. We present the concept of two alternating DT replicas for reconfigurable stations. One of these DTs (referred to as the current twin) is focused on processing the current product, while the second twin (the incoming twin) is focused on the incoming product and its process requirements. This mechanism allows the incoming twin to prepare for changes and start the reconfiguration as early as possible. Some reconfigurations may be done before the workpiece arrival, and a plan for minimizing the reconfigurations time would be ready for the switching moment between incoming and outgoing products. The hand-off process of the workpiece, as it switches stations, is reflected by real-time digital hand-off process at the station where the incoming twin and current twin interchange roles. This is a key to the improved efficiency of the process which is also the paper’s major contribution to enhancing the flexibility and responsiveness of mixed model flow lines. The paper concludes with a discussion of future research directions, including the integration of artificial intelligence for predictive reconfiguration and the extension of the digital twin concept to entire production networks
Task 5: Environmental and economic impacts of newly developed processes and materials
No full textNational audienc
An Iterative Framework for Agile Transformation: Lessons from Programme Reform
No full textInternational audienceIn rapidly evolving environments, enterprises increasingly adopt agile transformation strategies to navigate internal and external changes and maintain competitiveness. However, the absence of clear guidance often complicates these initiatives. This paper addresses this gap by integrating a theoretically grounded framework with complementary methods to tackle a case study in engineering education. The case study demonstrates how a refined version of the framework can effectively structure a programme reform project. The findings from this research contribute to both theory and practice by offering insights into how to operationalise agile transformation
Changement climatique : comment territorialiser la responsabilité de la métropole de lyon dans le dépassement des limites planétaires ?
No full textMillenaire 3 La prospective de la Métropole de LyonEn émettant massivement des gaz à effet de serre, l’humanité a progressivement dépassé les seuils qui permettent de garantir la stabilité du système climatique.Un tel dépassement rend possible l’atteinte de points de bascule susceptibles d’entraîner une sortie potentiellement catastrophique du régime climatique de l’Holocène — cette période climatique qui a débuté il y a plus de 10 000 ans et durant laquelle sont nées les civilisations.Peut-on mesurer la responsabilité d’un territoire dans ce dépassement ?Et quelles leçons en tirer ?Cet article présente les résultats d’un exercice mené en collaboration avec la Métropole de Lyon par des chercheurs du laboratoire Environnement Ville Société de l’École des Mines de Saint-Étienne
Alpha-case promotes fatigue cracks initiation from the surface in heat treated Ti-6Al-4V fabricated by Laser Powder Bed Fusion
No full textInternational audienceThis research investigates the effect of the formation of an oxygen-stabilised titanium alpha layer – called alpha-case at the surface – on the fatigue properties of Ti-6Al-4V (Ti64) alloy components produced by Laser Powder Bed Fusion (L-PBF). Three post processing heat treatments with different controlled atmospheres were carried out on samples with as-built surfaces to evaluate how differences in alpha-case layer thickness and hardness affect the material’s susceptibility to surface embrittlement and its overall fatigue performance. The investigation includes bulk and subsurface microstructural analysis, surface characterisation by X-ray computed tomography (XCT), and fatigue testing. Key findings show that alpha-case layers can reduce the fatigue resistance of L-PBF fabricated Ti64. The presence of a 70±3 µm thick alpha-case layer was found to promote crack initiation. This is emphasised by a higher density of initiated cracks, thus leading to a reduction in fatigue life. Conversely, thinner alpha-case layers were found to have a reduced impact on the fatigue performance, highlighting the critical role of post processing heat treatments in modulating the fatigue resistance of the material. The use of XCT to characterise the surfaces of the specimens in 3D confirms that fatigue cracks primarily initiate at surface notches, highlighting the predominance of as-built surfaces over microstructure in determining the fatigue resistance of L-PBF Ti64 components
Modeling the Oxidation Kinetics of Ti-6Al-4 V Spherical Powder at High Temperatures
No full textInternational audienceThe aim of the present work is to establish a kinetic law for the oxidation of Ti-6Al-4V (Ti64) spherical powder at high temperatures miming the possible oxidation of such powder within a laser powder bed fusion process. The oxidation experiments were followed by isothermal and isobaric thermogravimetry between 700 and 750 °C, under a controlled partial pressure of O2in the range of 0.1 to 0.75 atm. Beside the duplex structure of the oxide layers formed, namely an inner layer composed mainly of TiO2and an outer one composed of Al2O3, it was found that the oxidation rate is limited by one rate-determining step occurring in a single reaction zone: the Al2O3 layer. The study of how the growth rate varies with the partial pressure of O2 highlighted that the rate-determining step is the diffusion of interstitial oxygen as a dumbbell in this Al2O3 layer. Based on physico-geometrical description of the reaction, a complete reaction rate equation is then proposed by taking into account the spherical geometry and the dimensions of the Ti64 particles as well as a dependence of the reaction rate with temperature and partial pressure of O2. The rate law is very satisfactorily confronted to the experimental data
Aggregated model for anomaly detection: a statistical learning approach
No full textInternational audienceThe core of this research is a novel mathematical model for anomaly detection indiscrete manufacturing processes. Designed to identify significant deviations in production lead timewithin a production line, the model is constructed using statistical learning techniques and aggregatesmultiple individual classification models. It relies on the principles of information theory, particularlyShannon’s Entropy, to quantify uncertainty and enhance the precision of anomaly identification.Consequently, the model demonstrates high performance in distinguishing outliers from anomalies,making it applicable to real-world production data. The model is extensible and can be integrated intoproduction line control systems for real-time anomaly detection, aligning with Industry 4.0 principles,including smart manufacturing and data-driven decision-making.Le cœur de cette recherche est un modèle mathématique novateur pour la détection desanomalies dans les processus de fabrication discrète. Conçu pour identifier des écarts significatifs dansles délais de production au sein d'une ligne de production, le modèle est construit en utilisant destechniques d'apprentissage statistique et agrège plusieurs modèles individuels de classification. Ils'appuie sur les principes de la théorie de l'information, notamment l'Entropie de Shannon, pourquantifier l'incertitude et améliorer la précision de l'identification des anomalies. En conséquence, lemodèle démontre des performances élevées dans la distinction entre les valeurs aberrantes et lesanomalies, ce qui le rend applicable aux données de production réelles. Le modèle est extensible et peutêtre intégré dans des systèmes de pilotage des lignes de production pour une détection des anomalies entemps réel, s'inscrivant dans les principes de l'Industrie 4.0, notamment la fabrication intelligente et laprise de décision pilotée par les donnée
Collocation-based kriging with applications to the prediction of perfect flows
No full textNational audienc
Using scenario-based planning approaches to solve scheduling problems in healthcare
No full textNational audienceThis poster, presented at the HCERES* audit of the LIMOS laboratory in 2025, summarizes several contributions of the St-Etienne applied mathematics team of the LIMOS made between 2023 and 2025. * HCERES : Haut Conseil de l'Evaluation de la Recherche et de l'Enseignement Supérieu