Dynamic Facility Location with Modular Capacities : Models, Algorithms and Applications in Forestry

Les décisions de localisation sont souvent soumises à des aspects dynamiques comme des changements dans la demande des clients. Pour y répondre, la solution consiste à considérer une flexibilité accrue concernant l’emplacement et la capacité des installations. Même lorsque la demande est prévisible, trouver le planning optimal pour le déploiement et l'ajustement dynamique des capacités reste un défi. Dans cette thèse, nous nous concentrons sur des problèmes de localisation avec périodes multiples, et permettant l'ajustement dynamique des capacités, en particulier ceux avec des structures de coûts complexes. Nous étudions ces problèmes sous différents points de vue de recherche opérationnelle, en présentant et en comparant plusieurs modèles de programmation linéaire en nombres entiers (PLNE), l'évaluation de leur utilisation dans la pratique et en développant des algorithmes de résolution efficaces. Cette thèse est divisée en quatre parties. Tout d’abord, nous présentons le contexte industriel à l’origine de nos travaux: une compagnie forestière qui a besoin de localiser des campements pour accueillir les travailleurs forestiers. Nous présentons un modèle PLNE permettant la construction de nouveaux campements, l’extension, le déplacement et la fermeture temporaire partielle des campements existants. Ce modèle utilise des contraintes de capacité particulières, ainsi qu’une structure de coût à économie d’échelle sur plusieurs niveaux. L'utilité du modèle est évaluée par deux études de cas. La deuxième partie introduit le problème dynamique de localisation avec des capacités modulaires généralisées. Le modèle généralise plusieurs problèmes dynamiques de localisation et fournit de meilleures bornes de la relaxation linéaire que leurs formulations spécialisées. Le modèle peut résoudre des problèmes de localisation où les coûts pour les changements de capacité sont définis pour toutes les paires de niveaux de capacité, comme c'est le cas dans le problème industriel mentionnée ci-dessus. Il est appliqué à trois cas particuliers: l'expansion et la réduction des capacités, la fermeture temporaire des installations, et la combinaison des deux. Nous démontrons des relations de dominance entre notre formulation et les modèles existants pour les cas particuliers. Des expériences de calcul sur un grand nombre d’instances générées aléatoirement jusqu’à 100 installations et 1000 clients, montrent que notre modèle peut obtenir des solutions optimales plus rapidement que les formulations spécialisées existantes. Compte tenu de la complexité des modèles précédents pour les grandes instances, la troisième partie de la thèse propose des heuristiques lagrangiennes. Basées sur les méthodes du sous-gradient et des faisceaux, elles trouvent des solutions de bonne qualité même pour les instances de grande taille comportant jusqu’à 250 installations et 1000 clients. Nous améliorons ensuite la qualité de la solution obtenue en résolvent un modèle PLNE restreint qui tire parti des informations recueillies lors de la résolution du dual lagrangien. Les résultats des calculs montrent que les heuristiques donnent rapidement des solutions de bonne qualité, même pour les instances où les solveurs génériques ne trouvent pas de solutions réalisables. Finalement, nous adaptons les heuristiques précédentes pour résoudre le problème industriel. Deux relaxations différentes sont proposées et comparées. Des extensions des concepts précédents sont présentées afin d'assurer une résolution fiable en un temps raisonnable.

On the design of customized risk measures in insurance, the problem of capital allocation and the theory of fluctuations for Lévy processes

Dans cette thèse, nous étudions quelques problèmes fondamentaux en mathématiques financières et actuarielles, ainsi que leurs applications. Cette thèse est constituée de trois contributions portant principalement sur la théorie de la mesure de risques, le problème de l’allocation du capital et la théorie des fluctuations. Dans le chapitre 2, nous construisons de nouvelles mesures de risque cohérentes et étudions l’allocation de capital dans le cadre de la théorie des risques collectifs. Pour ce faire, nous introduisons la famille des "mesures de risque entropique cumulatifs" (Cumulative Entropic Risk Measures). Le chapitre 3 étudie le problème du portefeuille optimal pour le Entropic Value at Risk dans le cas où les rendements sont modélisés par un processus de diffusion à sauts (Jump-Diffusion). Dans le chapitre 4, nous généralisons la notion de "statistiques naturelles de risque" (natural risk statistics) au cadre multivarié. Cette extension non-triviale produit des mesures de risque multivariées construites à partir des données financiéres et de données d’assurance. Le chapitre 5 introduit les concepts de "drawdown" et de la "vitesse d’épuisement" (speed of depletion) dans la théorie de la ruine. Nous étudions ces concepts pour des modeles de risque décrits par une famille de processus de Lévy spectrallement négatifs.

Reproduire pour déconstruire : le paradoxe de l'engagement dans la série Body techniques de Carey Young

La série Body Techniques (2007) a été réalisée par l’artiste britannique Carey Young dans le cadre d’une résidence offerte par la biennale de Sharjah, aux Émirats Arabes Unis. Les huit photographies de format tableau constituant la série montrent l’artiste qui, portant l’uniforme d’une femme d’affaires, réinterprète huit œuvres célèbres associées à la mouvance de l’art conceptuel. Des paysages singuliers, situés aux abords des villes de Sharjah et Dubaï, servent de toile de fond à ces actions et leur confèrent une aura futuriste. La présente analyse tâche de démontrer que la série est habitée par un paradoxe remettant en question le statut d’art engagé que l’artiste revendique pour son œuvre. Ce paradoxe se manifeste à travers trois axes, autour desquels s’articule notre réflexion : les médiations se glissant entre Body Techniques et les œuvres que la série réinterprète, la déconstruction du « médium » du paysage, et le rôle actif occupé par le dispositif photographique. Cet examen attentif de chacune des occurrences du paradoxe permet de révéler Body Techniques comme une incarnation exemplaire de la double contrainte traversant toute œuvre d’art contemporain engagé : celle permettant aux artistes de critiquer le système auquel ils participent, mais les forçant en retour à participer au système qu’ils critiquent.

Advances in scaling deep learning algorithms

Les algorithmes d'apprentissage profond forment un nouvel ensemble de méthodes puissantes pour l'apprentissage automatique. L'idée est de combiner des couches de facteurs latents en hierarchies. Cela requiert souvent un coût computationel plus elevé et augmente aussi le nombre de paramètres du modèle. Ainsi, l'utilisation de ces méthodes sur des problèmes à plus grande échelle demande de réduire leur coût et aussi d'améliorer leur régularisation et leur optimization. Cette thèse adresse cette question sur ces trois perspectives. Nous étudions tout d'abord le problème de réduire le coût de certains algorithmes profonds. Nous proposons deux méthodes pour entrainer des machines de Boltzmann restreintes et des auto-encodeurs débruitants sur des distributions sparses à haute dimension. Ceci est important pour l'application de ces algorithmes pour le traitement de langues naturelles. Ces deux méthodes (Dauphin et al., 2011; Dauphin and Bengio, 2013) utilisent l'échantillonage par importance pour échantilloner l'objectif de ces modèles. Nous observons que cela réduit significativement le temps d'entrainement. L'accéleration atteint 2 ordres de magnitude sur plusieurs bancs d'essai. Deuxièmement, nous introduisont un puissant régularisateur pour les méthodes profondes. Les résultats expérimentaux démontrent qu'un bon régularisateur est crucial pour obtenir de bonnes performances avec des gros réseaux (Hinton et al., 2012). Dans Rifai et al. (2011), nous proposons un nouveau régularisateur qui combine l'apprentissage non-supervisé et la propagation de tangente (Simard et al., 1992). Cette méthode exploite des principes géometriques et permit au moment de la publication d'atteindre des résultats à l'état de l'art. Finalement, nous considérons le problème d'optimiser des surfaces non-convexes à haute dimensionalité comme celle des réseaux de neurones. Tradionellement, l'abondance de minimum locaux était considéré comme la principale difficulté dans ces problèmes. Dans Dauphin et al. (2014a) nous argumentons à partir de résultats en statistique physique, de la théorie des matrices aléatoires, de la théorie des réseaux de neurones et à partir de résultats expérimentaux qu'une difficulté plus profonde provient de la prolifération de points-selle. Dans ce papier nous proposons aussi une nouvelle méthode pour l'optimisation non-convexe.

Acoustic radiation and scattering from cylindrical bodies and analysis of transducer arrays

New mathematical methods to analytically investigate linear acoustic radiation and scattering from cylindrical bodies and transducer arrays are presented. Three problems of interest involving cylinders in an infinite fluid are studied. In all the three problems, the Helmholtz equation is used to model propagation through the fluid and the beam patterns of arrays of transducers are studied. In the first problem, a method is presented to determine the omni-directional and directional far-field pressures radiated by a cylindrical transducer array in an infinite rigid cylindrical baffle. The solution to the Helmholtz equation and the displacement continuity condition at the interface between the array and the surrounding water are used to determine the pressure. The displacement of the surface of each transducer is in the direction of the normal to the array and is assumed to be uniform. Expressions are derived for the pressure radiated by a sector of the array vibrating in-phase, the entire array vibrating in-phase, and a sector of the array phase-shaded to simulate radiation from a rectangular piston. It is shown that the uniform displacement required for generating a source level of 220 dB ref. μPa @ 1m that is omni directional in the azimuthal plane is in the order of 1 micron for typical arrays. Numerical results are presented to show that there is only a small difference between the on-axis pressures radiated by phased cylindrical arrays and planar arrays. The problem is of interest because cylindrical arrays of projectors are often used to search for underwater objects. In the second problem, the errors, when using data-independent, classical, energy and split beam correlation methods, in finding the direction of arrival (DOA) of a plane acoustic wave, caused by the presence of a solid circular elastic cylindrical stiffener near a linear array of hydrophones, are investigated. Scattering from the effectively infinite cylinder is modeled using the exact axisymmetric equations of motion and the total pressures at the hydrophone locations are computed. The effect of the radius of the cylinder, a, the distance between the cylinder and the array, b, the number of hydrophones in the array, 2H, and the angle of incidence of the wave, α, on the error in finding the DOA are illustrated using numerical results. For an array that is about 30 times the wavelength and for small angles of incidence (α<10), the error in finding the DOA using the energy method is less than that using the split beam correlation method with beam steered to α; and in some cases, the error increases when b increases; and the errors in finding the DOA using the energy method and the split beam correlation method with beam steered to α vary approximately as a7 / 4 . The problem is of interest because elastic stiffeners – in nearly acoustically transparent sonar domes that are used to protect arrays of transducers – scatter waves that are incident on it and cause an error in the estimated direction of arrival of the wave. In the third problem, a high-frequency ray-acoustics method is presented and used to determine the interior pressure field when a plane wave is normally incident on a fluid cylinder embedded in another infinite fluid. The pressure field is determined by using geometrical and physical acoustics. The interior pressure is expressed as the sum of the pressures due to all rays that pass through a point. Numerical results are presented for ka = 20 to 100 where k is the acoustic wavenumber of the exterior fluid and a is the radius of the cylinder. The results are in good agreement with those obtained using field theory. The directional responses, to the plane wave, of sectors of a circular array of uniformly distributed hydrophones in the embedded cylinder are then computed. The sectors are used to simulate linear arrays with uniformly distributed normals by using delays. The directional responses are compared with the output from an array in an infinite homogenous fluid. These outputs are of interest as they are used to determine the direction of arrival of the plane wave. Numerical results are presented for a circular array with 32 hydrophones and 12 hydrophones in each sector. The problem is of interest because arrays of hydrophones are housed inside sonar domes and acoustic plane waves from distant sources are scattered by the dome filled with fresh water and cause deterioration in the performance of the array.

Development of Free Space Method for Material Characterisation and GPR Study & A Case Study on Biological Samples

The thesis mainly focuses on material characterization in different environments: freely available samples taken in planar fonn, biological samples available in small quantities and buried objects.Free space method, finds many applications in the fields of industry, medicine and communication. As it is a non-contact method, it can be employed for monitoring the electrical properties of materials moving through a conveyor belt in real time. Also, measurement on such systems at high temperature is possible. NID theory can be applied to the characterization of thin films. Dielectric properties of thin films deposited on any dielectric substrate can be determined. ln chemical industry, the stages of a chemical reaction can be monitored online. Online monitoring will be more efficient as it saves time and avoids risk of sample collection.Dielectric contrast is one of the main factors, which decides the detectability of a system. lt could be noted that the two dielectric objects of same dielectric constant 3.2 (s, of plastic mine) placed in a medium of dielectric constant 2.56 (er of sand) could even be detected employing the time domain analysis of the reflected signal. This type of detection finds strategic importance as it provides solution to the problem of clearance of non-metallic mines. The demining of these mines using the conventional techniques had been proved futile. The studies on the detection of voids and leakage in pipes find many applications.The determined electrical properties of tissues can be used for numerical modeling of cells, microwave imaging, SAR test etc. All these techniques need the accurate determination of dielectric constant. ln the modem world, the use of cellular and other wireless communication systems is booming up. At the same time people are concemed about the hazardous effects of microwaves on living cells. The effect is usually studied on human phantom models. The construction of the models requires the knowledge of the dielectric parameters of the various body tissues. lt is in this context that the present study gains significance. The case study on biological samples shows that the properties of normal and infected body tissues are different. Even though the change in the dielectric properties of infected samples from that of normal one may not be a clear evidence of an ailment, it is an indication of some disorder.ln medical field, the free space method may be adapted for imaging the biological samples. This method can also be used in wireless technology. Evaluation of electrical properties and attenuation of obstacles in the path of RF waves can be done using free waves. An intelligent system for controlling the power output or frequency depending on the feed back values of the attenuation may be developed.The simulation employed in GPR can be extended for the exploration of the effects due to the factors such as the different proportion of water content in the soil, the level and roughness of the soil etc on the reflected signal. This may find applications in geological explorations. ln the detection of mines, a state-of-the art technique for scanning and imaging an active mine field can be developed using GPR. The probing antenna can be attached to a robotic arm capable of three degrees of rotation and the whole detecting system can be housed in a military vehicle. In industry, a system based on the GPR principle can be developed for monitoring liquid or gas through a pipe, as pipe with and without the sample gives different reflection responses. lt may also be implemented for the online monitoring of different stages of extraction and purification of crude petroleum in a plant.Since biological samples show fluctuation in the dielectric nature with time and other physiological conditions, more investigation in this direction should be done. The infected cells at various stages of advancement and the normal cells should be analysed. The results from these comparative studies can be utilized for the detection of the onset of such diseases. Studying the properties of infected tissues at different stages, the threshold of detectability of infected cells can be determined.

Child labour in India

This thesis entitled “Child labour in india”Children are "supremely important assets" of the nation, India proudly asserts in the National Policy for Children (1974) gracefully acknowledging that they are future citizens on whose shoulders the destiny of our nation rests.Childhood is a time of discovery as the world and all it contains are new to children. It is a time of excitement and anticipation. lt is a time of dreams and fantasies. And it is a time of receiving love and appreciation. Paradoxically, a picture of contrast is a common experience in India as a vast majority of children who are starved of basic needs of nutrition, health and education are made to work at an early age in exploitative conditions. The specter of child labour is a glaring anomaly in a country graciously adorning human right. In an exposition of the problem involving human rights abuse, Chapters from Two to Five of this study have shot into focus the human rights jurisprudence with special reference to the rights of children. Children have a particular identity as children and they also have a universal identity as human beings.The concern for mankind expressed unequivocally and transcending the globe will be real and moving and not mere rhetoric and ritual if and only when it begins with children, as, to quote the words of Nehru, the human being counts much more as a child than as a grown up.The first three of these rights namely right to health, right to nutritive food and right to education are dealt with in Chapter Four. Finally, the positive effects of education have been sketched in that chapter to impress upon its significance in the development of human capitals.legitimization. The theme of legitimacy was rationalised on the ground of poverty as a strategy for achieving eradication of child labour ultimately by enforcing minimum wages, shorter working hours, leave compensation, non-formal education etc., as the employer would soon discover that child labour is not cheap and would be obliged to substitute adult labour. However, humanising the work life is only a promise to the detriment of children as the Act of 1986 enacted as a part of the new strategy is nearingcompletion of a decade of existence but nowhere near the fulfilment of the mission.As similar urge is more necessary and overdue, it has been suggested that a special body be established with all powers for cognisance of human rights abuse of children.It is proposed to conclude this study with a brief summary of the inferences drawn from the foregoing chapters along with a few suggestions emerging out of those inferences

Socio-Economic Dimensions of Health Care Development Among Tribals in Kerala: A Study of Tribal Communities in Wynad

The contemporary explanations and discussions of the relationship between medicine and health, and society centre around assumptions that can be broadly classified into three setsl. The first set considers health and illness as predominantly ‘biological’ and therefore, having nothing to do with the social and economic environment in which it occurs. The struggle to combat illness therefore, lies entirely within the purview of modern medicine which is neutral to economic or social change. The second considers practice of medicine as a natural science. It allows the doctor to separate himself from his subject matter, the patient, in the samelway as the natural scientist is assumed to separate himself from his subject matter, the natural world. As a 'science' and with the scientific method, it can produce unchallengable and autonomous body of knowledge which is free from the wider social and economic context. The third, different from the above, recognises the relationship between health, medicine and society. Social and environmental aspects as determinants of illness or of health comes to sharp focus here and it assigns to medicine the status of a mediator, the only viable mediator, between people and diseases. In this scheme of things the usefulness of medicine is unquestionable but the problem lies in not having enough of it to go arounds.

Reinforcement Learning approaches to Economic Dispatch problem

This paper presents Reinforcement Learning (RL) approaches to Economic Dispatch problem. In this paper, formulation of Economic Dispatch as a multi stage decision making problem is carried out, then two variants of RL algorithms are presented. A third algorithm which takes into consideration the transmission losses is also explained. Efficiency and flexibility of the proposed algorithms are demonstrated through different representative systems: a three generator system with given generation cost table, IEEE 30 bus system with quadratic cost functions, 10 generator system having piecewise quadratic cost functions and a 20 generator system considering transmission losses. A comparison of the computation times of different algorithms is also carried out.

Three-electron radiative transitions

A new type of many-electron radiative transitions involving three electrons is predicted. The results of their investigation by many-body perturbation theory are presented. New spectral lines observed in the wavelength range of 37.5 to 54.0 nm by means of photon-induced fluorescence spectroscopy (PIFS) following the excitation of the Kr I 3d{^-1}np resonances are reported and compared with the predictions.

Approximate solutions and error estimates for a Stokes boundary value problem

The aim of this paper is the numerical treatment of a boundary value problem for the system of Stokes' equations. For this we extend the method of approximate approximations to boundary value problems. This method was introduced by V. Maz'ya in 1991 and has been used until now for the approximation of smooth functions defined on the whole space and for the approximation of volume potentials. In the present paper we develop an approximation procedure for the solution of the interior Dirichlet problem for the system of Stokes' equations in two dimensions. The procedure is based on potential theoretical considerations in connection with a boundary integral equations method and consists of three approximation steps as follows. In a first step the unknown source density in the potential representation of the solution is replaced by approximate approximations. In a second step the decay behavior of the generating functions is used to gain a suitable approximation for the potential kernel, and in a third step Nyström's method leads to a linear algebraic system for the approximate source density. For every step a convergence analysis is established and corresponding error estimates are given.

Mathematical problem solving, modelling, applications, and links to other subjects

The paper will consist of three parts. In part I we shall present some background considerations which are necessary as a basis for what follows. We shall try to clarify some basic concepts and notions, and we shall collect the most important arguments (and related goals) in favour of problem solving, modelling and applications to other subjects in mathematics instruction. In the main part II we shall review the present state, recent trends, and prospective lines of development, both in empirical or theoretical research and in the practice of mathematics instruction and mathematics education, concerning problem solving, modelling, applications and relations to other subjects. In particular, we shall identify and discuss four major trends: a widened spectrum of arguments, an increased globality, an increased unification, and an extended use of computers. In the final part III we shall comment upon some important issues and problems related to our topic.

Applied mathematical problem solving, modelling, applications, and links to other subjects

The paper will consist of three parts. In part I we shall present some background considerations which are necessary as a basis for what follows. We shall try to clarify some basic concepts and notions, and we shall collect the most important arguments (and related goals) in favour of problem solving, modelling and applications to other subjects in mathematics instruction. In the main part II we shall review the present state, recent trends, and prospective lines of development, both in empirical or theoretical research and in the practice of mathematics instruction and mathematics education, concerning (applied) problem solving, modelling, applications and relations to other subjects. In particular, we shall identify and discuss four major trends: a widened spectrum of arguments, an increased globality, an increased unification, and an extended use of computers. In the final part III we shall comment upon some important issues and problems related to our topic.

The Koundoum sheep breed in Niger: morpho-biometric study and description of the production system

The diffusion of highly productive breeds across developing countries goes along with a neglect of indigenous breeds, which are well suited to their environment but often show low yields. Thus, in Niger, the flock of Koundoum sheep are rapidly decreasing. The Koundoum is one of the few wool sheep breeds of Africa and shows important adaptive feature to its native environment, i.e. the humid pastures on the banks of the Niger River. To characterise the breed and to understand its production context, a survey has been conducted in 104 herds in four communes along the Niger River (Kollo, Tillabery, Say and Tera). Nine body measurements, including live weight, were taken on 180 adult sheep (101 females and 79 males). The herds varied from 2 to 60 heads, with a median size of eight animals and two thirds of the herds having less than 10 animals. Mainly fed on natural pastures, 85.6% of the herds received crop residues. Only natural mating was practiced. Veterinary care was restricted to anti-helminthic and some indigenous treatments. The frequent affiliation of breeders to professional unions appeared as favourable to the implementation of a collective conservation program. The Koundoum sheep were white or black coated, with the black colour being most frequent (75.6%). Wattles were present in both sexes at similar frequencies of around 14%. All biometric variables were significantly and positively correlated between them. The thoracic perimeter showed the best correlation with live weight in both males and females. Three variables were selected for live weight prediction: thoracic perimeter, height at withers and rump length. From the present study, it is expected that the in situ conservation of the Koundoum sheep will be highly problematic, due to lack of market opportunities for wool and the willingness of smallholders to get involved in pure Koundoum rearing.

Enhanced Reality Visualization in a Surgical Environment

Enhanced reality visualization is the process of enhancing an image by adding to it information which is not present in the original image. A wide variety of information can be added to an image ranging from hidden lines or surfaces to textual or iconic data about a particular part of the image. Enhanced reality visualization is particularly well suited to neurosurgery. By rendering brain structures which are not visible, at the correct location in an image of a patient's head, the surgeon is essentially provided with X-ray vision. He can visualize the spatial relationship between brain structures before he performs a craniotomy and during the surgery he can see what's under the next layer before he cuts through. Given a video image of the patient and a three dimensional model of the patient's brain the problem enhanced reality visualization faces is to render the model from the correct viewpoint and overlay it on the original image. The relationship between the coordinate frames of the patient, the patient's internal anatomy scans and the image plane of the camera observing the patient must be established. This problem is closely related to the camera calibration problem. This report presents a new approach to finding this relationship and develops a system for performing enhanced reality visualization in a surgical environment. Immediately prior to surgery a few circular fiducials are placed near the surgical site. An initial registration of video and internal data is performed using a laser scanner. Following this, our method is fully automatic, runs in nearly real-time, is accurate to within a pixel, allows both patient and camera motion, automatically corrects for changes to the internal camera parameters (focal length, focus, aperture, etc.) and requires only a single image.