Portail HAL IMT Mines Albi
Not a member yet
5440 research outputs found
Sort by
Integrating Uncertainty into a Supply Chain Network for Adaptive S&OP Process
No full textPart 6: Uncertainty and Collaboration in Supply ChainInternational audienceInstability is the new normal for supply chain networks. In this context, the Adaptive Sales and Operations Planning (AS&OP) process proposed by the Demand-Driven Adaptive-Enterprise model aims to improve the way these supply chain networks manage volatile, uncertain, complex and ambiguous environments. However, the process, as originally defined, reliesmainly on a deterministic maximum likelihood approach, thus limiting its scope. The present research work examines existing strategies for integrating uncertainty into the strategic planning and ultimately proposes an original Decision Support System that integratesuncertainty via interval data and scenario-based planning. An illustrative case study validates this proposal and discusses its limitations
Étude et définition d’une approche outillée de modélisation des situations de crise et des situations à risques dédiées à l’entraînement en environnement immersifs
No full textThis thesis explores the development of a tool-based approach to virtual reality training dedicated to crisis management. The main objective of this research is to facilitate the teacher's experience. The teacher is considered as the scriptwriter during training in an interactive virtual environment. Thanks to the complementary contribution of new training, it is possible to increase the quality of the response of stakeholders to any type of crisis. The context of this study is to integrate the various components of a crisis into a structure based on model engineering. The aim is to design a practical tool for trainers. This tool is used to model, design and simulate crisis situations in an immersive way, within imagined or existing environments. The methodology adopted is based on exploring cascading effects, virtual reality and training creation tools to design interactive virtual environments. Three main areas were explored in this research: (i) pedagogical design and specification requirements for integrating pedagogical principles into the creation of virtual environments, (ii) the human and didactic aspects of immersing participants in crisis management training, and (iii) the technical and functional aspects dedicated to the design of crisis exercises in virtual reality. The results highlight the importance of an immersive and personalised approach to crisis management training, opening up new avenues for the use of virtual reality as a crisis management training tool.Cette thèse explore le développement d'une approche outillée pour formation en réalité virtuelle dédiée à la gestion de crise. L'objectif principal de cette recherche est de faciliter l'expérience de l'enseignant. L'enseignant est considéré comme le scénariste durant l'entraînement en environnement virtuel interactif. Grâce à l'apport complémentaire de nouveau entraînement complémentaire, il est possible d'accroître la qualité de la réponse des parties prenantes à tout type de crise. Le contexte de cette étude consiste à intégrer les différents éléments constitutifs d'une crise dans une structure basée sur l'ingénierie des modèles dans l'objectif de concevoir un outil pratique pour les entraineurs. Cet outil s'utilise pour modéliser, concevoir et simuler des situations de crise de manière immersive, en environnements imaginés ou existants. La méthodologie adoptée repose sur l'exploration des effets cascades, de la réalité virtuelle et d'outils de création d'entraînement pour concevoir des environnements virtuels interactifs. Cette recherche inclut trois axes principaux explorés dans cette recherche : (i) la conception pédagogique et les exigences du cahier des charges pour intégrer les principes pédagogiques dans la création des environnements virtuels, (ii) les aspects humains et didactiques de l'immersion des participants dans la formation en gestion de crise, ainsi que (iii) les aspects techniques et fonctionnels dédiés à la conception d'exercices de crise en réalité virtuelle. Les résultats obtenus mettent en lumière l'importance d'une approche immersive et personnalisée dans la formation à la gestion de crise. Cette recherche ouvre de nouvelles voies pour l'utilisation de la réalité virtuelle comme outil de formation en gestion de crise
SAFE’ME: an Augmented Reality Application to Support Harm Awareness for the Safety Promotion of Children with Autism Disorder
No full textInternational audienceAutism Spectrum Disorder (ASD), as a developmental disorder,causes difficulties in social interaction, communication, sharing emotions, makingtheir education and daily life complicated. We aim to help these childrenand their parents to reduce cognitive impairment, maximize independence andimprove their quality of life and their well-being. Technology-based approaches,such as augmented reality (AR), are increasingly used in safety education ofthese children. Our study, which was conducted in Tunisia, developed an ARapplication by using virtual objects to teach hazard recognition and avoidanceto make autistic children more conscious of the dangers in their environment
Towards Tailoring Reinforcement Learning to Solve the Online Surgery-Planning and Scheduling Problem
No full textInternational audienceThis paper addresses the online surgery planning and scheduling problemfor operating rooms and recovery beds. We aim to minimize the makespanby dynamically assigning surgery dates, operating rooms, and recovery beds.Our integrated framework uses a Mixed-Integer Linear Program (MILP) solvedwith Python's PuLP package for initial scheduling and Reinforcement Learning(RL) for real-time adjustments. The MILP provides a static schedule, while RLhandles disruptions and updates schedules using experience replay and target networksfor stable training. Preliminary results demonstrate that this approach effectivelymanages scheduling complexities, improving operational efficiency andminimizing idle times. This highlights the potential of combining MILP and RLfor adaptive surgical scheduling
The impact of glycol water exposure on PA6/GF30 properties
No full textInternational audienceThe current paper aims to study the behavior of Polyamide 6 (PA6) and Polyamide 6 reinforced with 30% of short glass fibers (PA6GF30) under glycol water (GW) mixture used for cars as coolant fluid. Samples were fully immersed in the mixture of GW (20:80) at 50°C, 70°C and 90°C for up to 80 days and periodically weighted. Results gathered revealed the occurrence of substantial changes especially for PA6 samples aged at 90°C. The long term ageing leads to the formation of ester species due to thermo-oxidation as pointed out by Infrared spectroscopy analysis. Moreover, dynamic mechanical analysis measurements showed that glycol water acted as an effective plasticizer in lowering the Tg of the polyamide 6. This plasticizer effect was confirmed by the gain of ductility at early stage of ageing for both materials (PA6 and PA6GF30). Nevertheless, after 80 days of ageing the effect of glycol water ageing is no longer physical in nature and an overall loss of mechanical properties was noticed. Indeed, an embrittlement of PA6 was measured that can be attributed to a decrease of the level of entanglements within the polymer. Nevertheless, for PA6GF30 material, the loss of strength and stiffness was the result of the interfacial debonding between the fibers and the matrix
A Model-Driven Framework to Support Portfolio Management Under Uncertainties
No full textInternational audienceProject Portfolio management has become a standard in the business world as a means of structuring and achieving corporate objectives. It enables the progress of a project portfolio to be monitored andevaluated using a wide range of techniques, in particular the risks management approach. However, due to the complexity and uniqueness of these portfolios, it’s becoming increasingly difficult to predict the evolution of portfolios and assess the impact of every change on the performance regarding the predefined objectives. This leads to emerging uncertainties that portfolio managers must address at each stage of the portfolio management process. To address this problem, we focus on the execution phase of portfolio management process and present through the vision of uncertainty in a portfolio management context, a modeldriven framework using the Physics of Decision paradigm to support the execution phase of a projects portfolio. This framework presents a new approach to monitor projects portfolios which add flexibility and information interpretation to the risk management principles and could serve as a basis for the development of future information systems
R3XA: Toward a metadata standard for experimental (photo)mechanics datasets
No full textInternational audienceOpen-data are digital data whose access and use are free to users, and which may be produced under public funding. They are distributed in a structured manner following a metadata standard. The number of opendatasets increases in experimental mechanics, but, to our knowledge, there is no metadata standard specific to the field of experimental mechanics yet. In this talk, an initiative in this direction is presented
DIC data filtering proposal for complex environments
No full textInternational audienceA filtering method has been developed to provide reliable experimental results obtained using 3D DIC stereo correlation measurement method in cases noise cannot be controlled. This method based on numerical post processing, has shown to be efficient for experiments lead in complex environments, when noise sources cannot be controlled nor quantified (wind, vibrations, variable lighting, humidity…). Furthermore, this work describes a reliable filter parameter optimisation procedure
A physical model and a Monte Carlo estimate for the specific intensity spatial derivative, angular derivative and geometric sensitivity
No full textStarting from the radiative transfer equation and its usual boundary conditions, the objective of this work is to design Monte Carlo algorithms estimating the specific intensity spatial and angular derivatives as well as its geometric sensitivity. The present document is structured in three parts, each of them dedicated to a specific derivative of the intensity. Although they are all assembled here in one document each derivative is of interest independently whether it be for radiative transfers analysis or engineering conception. Therefore, they are thought to be written as three different papers and are presented here as such. Estimating derivatives of the specific intensity when solving radiative transfers using a Monte-Carlo algorithm is challenging. Finite differences are often not sufficiently accurate and directly estimating the derivative from a specific Monte-Carlo algorithm can lead to arduous formal or numerical developments. The proposition here is to work from the radiative transfer equation and its boundary conditions to design a physical model for each derivatives. Only then Monte-Carlo algorithms are built from the derivatives differential equations using the usual equivalent path integral. Since the same methodology is applied to the specific intensity spatial derivative, angular derivative and geometric sensitivity we chose to keep the same writing structure for all three parts so that all common ideas and developments appears exactly the same. We believe this choice to be coherent to facilitate the reader's understanding. Finally, these are preliminary versions of the final papers: for each parts the theory is fully described, but, although they have been implemented, the examples and algorithms sections are not always complete. This will be mentioned in the introductions of the concerned sections
Experimental and numerical analysis of structural inhomogeneity of grooved compacts
No full textInternational audienceStructural inhomogeneity developed within compacted powders using punches with grooved surfaces, is a critical issue due the cracking tendency. This study aims to provide a better understanding of the structural inhomogeneity according to the design of the punch surfaces using a finite element modelling. Two case studies are compared. In this modelling, the Drucker-Prager Cap model with density-dependent parameters is considered while the density distributions of manufactured grooved compacts are determined using X-ray tomography. The structural inhomogeneity is then analysed with the goal to determine which punch design is likely to improve the homogeneity.The contact-stress, stress and density distributions within the compact are computed. Predictions of the density distributions are validated from the measurements and show high densified regions localized essentially under the grooves and decreases drastically towards the compact center. The inhomogeneity of structure generated by each punch design is explained by the resulting contact-stress and the wall friction. A notable improvement of the homogeneity is qualitatively showed using the design of five instead of two grooves which was confirmed by X-ray tomography images and modelling thanks to the regular distribution of contact-pressure over the powder bed, even though this last is not uniform