Suchergebnisse

The term globalization, which is used to sum up the recent shift in international economic relations, refers to the increasing interdependence of world economies, combined with more intense exchanges of both goods and capital. Under the auspices of the GATT and then the WTO, this shift has also seen increasing State withdrawal, known as liberalization. Farming, which was long seen as an exceptional sector, has been at the heart of international talks since the Uruguay Round and the 1994 Marrakech agreement. The sector, which is involved in greenhouse gas emissions and sequestration, is also directly concerned by the international talks on global change. Liberalization is expected to result in a global increase in wellbeing, linked on the one hand to increased efficiency through specialization, and on the other hand to the fact that any shocks are spread over a broader market, thus reducing their adverse effects. This is one of the major results of economic theory, illustrated by many simulations backed up by figures, which have estimated the gains in terms of billions of dollars. However, although these results are guaranteed in a theoretical perfect market, they are in fact far from the reality of the current global market, whose very considerable fluctuations can induce huge efficiency losses.

Global Sourcing is becoming a common practice in industrial activities since it offers companies opportunities to improve its competitiveness in an increasingly competitive business environment. At the same time, it makes the flows more complex and the supply chain more fragile. Global Sourcing thus gives rise to a wide range of issues and impacts different levels of decision making. To address such a problem, we focus on tactical and operational decision making. We attempt to answer a variety of questions: What are possible actions for flow management in global sourcing? How to secure the procurement in the current industrial context? Are classical flow management policies also efficient in global sourcing? In collaboration with the industrial partners of the Chaire Supply Chain at Ecole Centrale Paris, we consider different problems. Firstly, we are interested in demand forecasting, an essential element for flow management in global sourcing and proposed a methodology to select an appropriate forecasting method and to update it dynamically. The fact that the lead times are long in global sourcing makes the forecast less reliable and less and less reliable when the forecast horizon increases, which requires an evaluation of the forecast accuracy. We propose a detailed model of the forecast accuracy and its evolution with time horizon involved. As the last step of the work, this forecast accuracy model is applied to a real life flow management problem in global sourcing. The case study carried out based on real life data from PSA demonstrates a clear superiority of the proposed method over existing ones in terms of both service level and inventory level. ; Le Global Sourcing est aujourd'hui en pleine expansion car il offre aux entreprises une source potentielle de compétitivité dans un environnement de plus en plus concurrentiel. Néanmoins, il génère aussi une complexification des flux et une fragilisation de la Supply Chain Globale. La problématique du Global Sourcing est vaste et touche les différents niveaux de décision de l'entreprise. Pour cela nous nous sommes focalisés dans ce travail sur les aspects tactiques et opérationnels de ce domaine. Nous avons abordé ainsi diverses questions : Quels leviers d'action pour un pilotage efficace des flux en approvisionnement lointain? Comment sécuriser les approvisionnements lointains dans le contexte industriel actuel ? Les politiques classiques de pilotage de flux sont-elles suffisantes pour les approvisionnements lointains ? En collaboration avec les partenaires industriels de la Chaire Supply Chain de l'Ecole Centrale Paris, nous avons abordé différentes facettes de cette problématique. Nous nous sommes intéressés tout d'abord à la prévision comme élément nécessaire au pilotage des flux lointains et nous avons proposé une méthodologie de sélection et de mise à jour de méthodes de prévision. Les délais longs en approvisionnement lointain font que les erreurs de prévision s'amplifient, ce qui nous a amenés à étudier l'erreur prévisionnelle. Nous avons proposé dans ce sens une modélisation fine de cette erreur et de son évolution en fonction de l'horizon temporelle de la prévision. Dans la dernière étape de ce travail, nous avons utilisé cette modélisation de l'incertitude pour piloter efficacement les flux lointains. Nous avons montré sur des cas réels issus de l'entreprise PSA l'efficacité de la méthode proposée en termes de respect du niveau de service avec un niveau de stock largement inférieur aux méthodes classiques.

The energy transition law passed by the French government has specific implications for renewable energies, in particular for their remuneration mechanism. Until 2015, a purchase obligation contract made it possible to sell electricity from wind power at a fixed rate. From 2015 onwards, some wind farms began to be exempted from the purchase obligation. This is because wind energy is starting to be sold directly on the market by the producers because of the breach of the purchase obligation contracts. Distribution system operators and transmission system operators require or even oblige producers to provide at least a production forecast one day in advance in order to rebalance the market. Over- or underestimation could be subject to penalties. There is, therefore, a huge need for accurate forecasts. It is in this context that this thesis was launched with the aim of proposing a model for predicting wind farms production by machine learning. We have production data and real wind measurements as well as data from meteorological models. We first compared the performances of the GFS and ECMWF models and studied the relationships between these two models through canonical correlation analysis. We then applied machine learning models to validate a first random forest prediction model. We then modeled the spatio-temporal wind dynamics and integrated it into the prediction model, which improved the prediction error by 3%. We also studied the selection of grid points by a variable group importance measure using random forests. Random forest prediction intervals associated with point forecasts of wind farm production are also studied. The forecasting model resulting from this work was developed to enable the ENGIE Group to have its own daily forecasts for all its wind farms. ; La loi de transition énergétique votée par l'Etat français a des implications précises sur les énergies renouvelables, en particulier sur leur mécanisme de rémunération. Jusqu'en 2015, un contrat d'obligation d'achat permettait de vendre ...

The energy transition law passed by the French government has specific implications for renewable energies, in particular for their remuneration mechanism. Until 2015, a purchase obligation contract made it possible to sell electricity from wind power at a fixed rate. From 2015 onwards, some wind farms began to be exempted from the purchase obligation. This is because wind energy is starting to be sold directly on the market by the producers because of the breach of the purchase obligation contracts. Distribution system operators and transmission system operators require or even oblige producers to provide at least a production forecast one day in advance in order to rebalance the market. Over- or underestimation could be subject to penalties. There is, therefore, a huge need for accurate forecasts. It is in this context that this thesis was launched with the aim of proposing a model for predicting wind farms production by machine learning. We have production data and real wind measurements as well as data from meteorological models. We first compared the performances of the GFS and ECMWF models and studied the relationships between these two models through canonical correlation analysis. We then applied machine learning models to validate a first random forest prediction model. We then modeled the spatio-temporal wind dynamics and integrated it into the prediction model, which improved the prediction error by 3%. We also studied the selection of grid points by a variable group importance measure using random forests. Random forest prediction intervals associated with point forecasts of wind farm production are also studied. The forecasting model resulting from this work was developed to enable the ENGIE Group to have its own daily forecasts for all its wind farms. ; La loi de transition énergétique votée par l'Etat français a des implications précises sur les énergies renouvelables, en particulier sur leur mécanisme de rémunération. Jusqu'en 2015, un contrat d'obligation d'achat permettait de vendre ...

The energy transition law passed by the French government has specific implications for renewable energies, in particular for their remuneration mechanism. Until 2015, a purchase obligation contract made it possible to sell electricity from wind power at a fixed rate. From 2015 onwards, some wind farms began to be exempted from the purchase obligation. This is because wind energy is starting to be sold directly on the market by the producers because of the breach of the purchase obligation contracts. Distribution system operators and transmission system operators require or even oblige producers to provide at least a production forecast one day in advance in order to rebalance the market. Over- or underestimation could be subject to penalties. There is, therefore, a huge need for accurate forecasts. It is in this context that this thesis was launched with the aim of proposing a model for predicting wind farms production by machine learning. We have production data and real wind measurements as well as data from meteorological models. We first compared the performances of the GFS and ECMWF models and studied the relationships between these two models through canonical correlation analysis. We then applied machine learning models to validate a first random forest prediction model. We then modeled the spatio-temporal wind dynamics and integrated it into the prediction model, which improved the prediction error by 3%. We also studied the selection of grid points by a variable group importance measure using random forests. Random forest prediction intervals associated with point forecasts of wind farm production are also studied. The forecasting model resulting from this work was developed to enable the ENGIE Group to have its own daily forecasts for all its wind farms. ; La loi de transition énergétique votée par l'Etat français a des implications précises sur les énergies renouvelables, en particulier sur leur mécanisme de rémunération. Jusqu'en 2015, un contrat d'obligation d'achat permettait de vendre ...

The energy transition law passed by the French government has specific implications for renewable energies, in particular for their remuneration mechanism. Until 2015, a purchase obligation contract made it possible to sell electricity from wind power at a fixed rate. From 2015 onwards, some wind farms began to be exempted from the purchase obligation. This is because wind energy is starting to be sold directly on the market by the producers because of the breach of the purchase obligation contracts. Distribution system operators and transmission system operators require or even oblige producers to provide at least a production forecast one day in advance in order to rebalance the market. Over- or underestimation could be subject to penalties. There is, therefore, a huge need for accurate forecasts. It is in this context that this thesis was launched with the aim of proposing a model for predicting wind farms production by machine learning. We have production data and real wind measurements as well as data from meteorological models. We first compared the performances of the GFS and ECMWF models and studied the relationships between these two models through canonical correlation analysis. We then applied machine learning models to validate a first random forest prediction model. We then modeled the spatio-temporal wind dynamics and integrated it into the prediction model, which improved the prediction error by 3%. We also studied the selection of grid points by a variable group importance measure using random forests. Random forest prediction intervals associated with point forecasts of wind farm production are also studied. The forecasting model resulting from this work was developed to enable the ENGIE Group to have its own daily forecasts for all its wind farms. ; La loi de transition énergétique votée par l'Etat français a des implications précises sur les énergies renouvelables, en particulier sur leur mécanisme de rémunération. Jusqu'en 2015, un contrat d'obligation d'achat permettait de vendre ...

The energy transition law passed by the French government has specific implications for renewable energies, in particular for their remuneration mechanism. Until 2015, a purchase obligation contract made it possible to sell electricity from wind power at a fixed rate. From 2015 onwards, some wind farms began to be exempted from the purchase obligation. This is because wind energy is starting to be sold directly on the market by the producers because of the breach of the purchase obligation contracts. Distribution system operators and transmission system operators require or even oblige producers to provide at least a production forecast one day in advance in order to rebalance the market. Over- or underestimation could be subject to penalties. There is, therefore, a huge need for accurate forecasts. It is in this context that this thesis was launched with the aim of proposing a model for predicting wind farms production by machine learning. We have production data and real wind measurements as well as data from meteorological models. We first compared the performances of the GFS and ECMWF models and studied the relationships between these two models through canonical correlation analysis. We then applied machine learning models to validate a first random forest prediction model. We then modeled the spatio-temporal wind dynamics and integrated it into the prediction model, which improved the prediction error by 3%. We also studied the selection of grid points by a variable group importance measure using random forests. Random forest prediction intervals associated with point forecasts of wind farm production are also studied. The forecasting model resulting from this work was developed to enable the ENGIE Group to have its own daily forecasts for all its wind farms. ; La loi de transition énergétique votée par l'Etat français a des implications précises sur les énergies renouvelables, en particulier sur leur mécanisme de rémunération. Jusqu'en 2015, un contrat d'obligation d'achat permettait de vendre ...

This PhD originates from a cooperation from CEREMA (Centre for Studies and Expertise on Risks, the Environment, Mobility and Urban Planning), EDF R&R, CERFACS (European Center for Avanced Research and Training in Computational Sciences) and SCHAPI (French national service for flood forecasting). In order to realize the twice-daily vigilance maps, the SCHAPI and the 19 SPC (Services de Prévision des Crues) distributed on the territory, use, among others, the results of numerical models generally launched with a deterministic approach (meteorological forecasts, hydrological and hydraulic modeling). The objective of the thesis is the implementation and evaluation of hydrological and hydraulic ensemble forecasts within the framework of the flood surveillance and forecast carried out by the governmental agencies in order to identify and reduce uncertainties for short to mid-term forecasts (24 hours). The originality of this work lies in the hybrid use of physics-based models and learning models on a large volume of data. For this purpose, meteorological forecasts are used to force a hydrological-hydraulic chained model to provide discharge and water level forecasts. In order to take into account the various sources of uncertainty related to the numerical models, model parameters and associated data, the limits of the deterministic are overcome by an ensemble approach; an ensemble of flow and water level forecasts is thus generated.The study basin is the Odet watershed located in Finistère (France, Brittany). The upstream part of the basin is represented by a hydrological model (GRP or MORDOR-TS). It provides a forecasted inflow for the 1D hydraulic model MASCARET that represents the dynamics of the river, and represents water heights at the downstream observing stations.First, a global sensitivity study (GSA) is carried out for the hydrological and hydraulic models. This is a prerequisite for the generation of ensemble forecasts. The GSA allows to identify the main sources of uncertainty and to perturb the ...

This PhD originates from a cooperation from CEREMA (Centre for Studies and Expertise on Risks, the Environment, Mobility and Urban Planning), EDF R&R, CERFACS (European Center for Avanced Research and Training in Computational Sciences) and SCHAPI (French national service for flood forecasting). In order to realize the twice-daily vigilance maps, the SCHAPI and the 19 SPC (Services de Prévision des Crues) distributed on the territory, use, among others, the results of numerical models generally launched with a deterministic approach (meteorological forecasts, hydrological and hydraulic modeling). The objective of the thesis is the implementation and evaluation of hydrological and hydraulic ensemble forecasts within the framework of the flood surveillance and forecast carried out by the governmental agencies in order to identify and reduce uncertainties for short to mid-term forecasts (24 hours). The originality of this work lies in the hybrid use of physics-based models and learning models on a large volume of data. For this purpose, meteorological forecasts are used to force a hydrological-hydraulic chained model to provide discharge and water level forecasts. In order to take into account the various sources of uncertainty related to the numerical models, model parameters and associated data, the limits of the deterministic are overcome by an ensemble approach; an ensemble of flow and water level forecasts is thus generated.The study basin is the Odet watershed located in Finistère (France, Brittany). The upstream part of the basin is represented by a hydrological model (GRP or MORDOR-TS). It provides a forecasted inflow for the 1D hydraulic model MASCARET that represents the dynamics of the river, and represents water heights at the downstream observing stations.First, a global sensitivity study (GSA) is carried out for the hydrological and hydraulic models. This is a prerequisite for the generation of ensemble forecasts. The GSA allows to identify the main sources of uncertainty and to perturb the ...

This PhD originates from a cooperation from CEREMA (Centre for Studies and Expertise on Risks, the Environment, Mobility and Urban Planning), EDF R&R, CERFACS (European Center for Avanced Research and Training in Computational Sciences) and SCHAPI (French national service for flood forecasting). In order to realize the twice-daily vigilance maps, the SCHAPI and the 19 SPC (Services de Prévision des Crues) distributed on the territory, use, among others, the results of numerical models generally launched with a deterministic approach (meteorological forecasts, hydrological and hydraulic modeling). The objective of the thesis is the implementation and evaluation of hydrological and hydraulic ensemble forecasts within the framework of the flood surveillance and forecast carried out by the governmental agencies in order to identify and reduce uncertainties for short to mid-term forecasts (24 hours). The originality of this work lies in the hybrid use of physics-based models and learning models on a large volume of data. For this purpose, meteorological forecasts are used to force a hydrological-hydraulic chained model to provide discharge and water level forecasts. In order to take into account the various sources of uncertainty related to the numerical models, model parameters and associated data, the limits of the deterministic are overcome by an ensemble approach; an ensemble of flow and water level forecasts is thus generated.The study basin is the Odet watershed located in Finistère (France, Brittany). The upstream part of the basin is represented by a hydrological model (GRP or MORDOR-TS). It provides a forecasted inflow for the 1D hydraulic model MASCARET that represents the dynamics of the river, and represents water heights at the downstream observing stations.First, a global sensitivity study (GSA) is carried out for the hydrological and hydraulic models. This is a prerequisite for the generation of ensemble forecasts. The GSA allows to identify the main sources of uncertainty and to perturb the significant uncertain parameters for the generation of an ensemble of forecasted discharges and water levels. The propagation of these uncertainties results in the generation of a raw ensemble for both hydrology and hydraulics, with the hydrological ensembles used to force the hydraulic ensembles. Two methods of ensemble correction are then investigated in the PhD: statistical calibration via the Quantile Regression Forest method and data assimilation calibration via an ensemble Kalman Filter (EnKF). It was shown that both approaches significantly improve the performance of the ensemble in terms of reliability and resolution. Finally, the performance of ensemble forecasts is compared for hydrology and hydraulics and recommendations are made for the operational generation of ensemble flow forecasts within SPCs ; Cette thèse s'inscrit dans le cadre d'un partenariat entre le CEREMA (Centre d'Études et d'expertise sur les Risques, l'Environnement, la Mobilité et l'Aménagement), EDF R&D, le CERFACS (Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique) et le SCHAPI (Service Central d'Hydrométéorologie et d'Appui à la Prévision des Inondations). Afin de réaliser les cartes de vigilance bi-quotidiennes, le SCHAPI et les 19 SPC (Services de Prévision des Crues) répartis sur le territoire utilisent entre autres des résultats de modèles numériques généralement lancés de manière déterministe (prévisions météorologiques, modélisations hydrologique et hydraulique). L'objectif de la thèse est la mise en place et l'évaluation de prévisions d'ensembles hydrologiques et hydrauliques dans le cadre de la vigilance crue-inondation réalisée par les services de l'État afin de mieux appréhender et réduire les incertitudes dans un contexte de prévision à courte et moyenne échéance (24 heures). L'originalité de ce travail réside dans l'utilisation hybride de modèles à base physique et de modèles d'apprentissage sur un important volume de données. Dans cet objectif, les prévisions météorologiques forcent un modèle chaîné hydrologie-hydraulique afin de fournir des prévisions de débit et de hauteurs d'eau. Afin de prendre en compte les diverses sources d'incertitude liées aux modèles numériques, aux paramètres des modèles et aux données associées, l'approche déterministe est remplacée par une approche ensembliste ; on fournit ainsi un ensemble de prévisions de débits et hauteurs d'eau.Le bassin d'étude est le bassin versant de l'Odet situé dans le Finistère. La partie amont du bassin est modélisée par un modèle hydrologique (GRP ou MORDOR-TS). Il fournit une prévision de débit qui sert de forçage au modèle hydraulique 1D MASCARET, qui lui prévoit des hauteurs d'eau aux stations de vigilance en aval.Dans un premier temps, une étude de sensibilité globale (GSA) est menée sur les modèles hydrologiques et hydrauliques. Ceci est un préalable à la génération des prévisions d'ensemble. La GSA permet d'identifier les sources principales d'incertitude et ainsi de perturber les paramètres incertains significatifs pour la représentation des débits et des hauteurs d'eau prévus. La propagation de ces incertitudes aboutit à la création d'un ensemble brut pour l'hydrologie et pour l'hydraulique, les ensembles hydrologiques étant utilisés pour forcer les ensembles hydrauliques. Deux méthodes de correction des ensembles sont alors investiguées dans la thèse : la calibration statistique via la méthode des forêts aléatoires « Quantile Regression Forest » et la calibration par assimilation de données via un filtre de Kalman d'ensemble (EnKF). On a montré que ces deux approches améliorent significativement les performances de l'ensemble en termes de fiabilité et résolution. Enfin la comparaison des performances des prévisions d'ensemble est finalement réalisée pour l'hydrologie et l'hydraulique et des préconisations sont émises pour la génération opérationnelle de prévisions d'ensemble au sein des SPC

This PhD in History of Science focuses on the Limits to Growth report, published in 1972, and more generally on the debate about the limits to growth, in the period from 1945 to 1990. ; Cette thèse d'Histoire des Sciences est centrée sur le rapport au Club de Rome de 1972, affirmant l'existence de limites globales à la croissance et la nécessité de ruptures sociétales radicales pour s'y ajuster. Elle pose la question des conditions matérielles, politiques et culturelles qui conduisent à la montée en force, autour de 1970, de discours énonçant le caractère nuisible de la croissance démographique et économique, à un niveau global. A cette fin, elle examine comment se développe un contexte favorable à l'appréhension d'un futur de l'humanité et à la modélisation mathématique d'un tel enjeu. Elle étudie l'émergence, l'évolution et la réappropriation d'une dénonciation de la croissance, en lien avec le développement du Tiers Monde, les premières mobilisations environnementales et la critique de la technologie. Elle suit le développement de l'entreprise du Club de Rome, afin d'élucider le paradoxe de l'appel par une élite industrielle et politique à une stabilisation de l'économie mondiale. Dans cette perspective, elle étudie les influences contrastées des discours environnementalistes et des études du futur sur le projet de modélisation de l'organisation, en particulier sur la représentation de la technologie qu'il porte. Elle étudie de manière précise comment le choix de la Dynamique des Systèmes comme méthodologie de modélisation concourt à une traduction particulière de la « Problématique » du Club en modèle mathématique, focalisée sur les limites à la croissance.Elle s'attache enfin à comprendre comment le vif débat du début des années 1970, pour ou contre la croissance, laisse rapidement la place à un consensus sur le bien-fondé de la croissance, tandis que les questions des inégalités entre Nord et Sud, et des meilleurs moyens technologiques et économiques pour dépasser les limites matérielles, deviennent alors ...