Le thème principal de cette thèse est l'étude des processus de Fragmentation et de Coalescence dit homogènes. Le premier chapitre se concentre sur les propriétés des fragmentations de masses, en particulier on établit une correspondance utile entre les fragmentations de masses et les fragmentations de partitions. La deuxième partie se concentre sur les fragmentations d'intervalles. On s'intéresse particulièrement aux ensembles de points ayant une vitesse de fragmentation anormale et on calcule explicitement la dimension de Hausdorff de tels ensembles. Enfin, dans le dernier tiers on définit et on étudie les processus de fragmentation-coalescence homogènes qui décrivent l'évolution de systèmes de particules dans lesquels le phénomènes de coalescences et de fragmentations sont présent simultanément. On montre que la condition d'homogénéité est suffisante pour assurer l'existence d'une mesure d'équilibre que l'on étudie.

Industrial metal forming operations generally require several deformation steps in order to create the final product. The mechanical behavior of materials undergoing strain path changes can be very different from those deformed in a given direction to fracture. The work presented here employed laser drilled model materials to better understand the effect of pre-strains and strain path changes on void growth and linkage leading to fracture is studied. The experimental results show that increasing pre-strain results in faster void growth which was justified in terms work hardening rate in the sample. Scanning electron microscope images revealed that the ductility of the sample decreased with increasing pre-strain but only slightly compared to the large decrease in far field strain at failure. This suggests that pre-strain affects strain localization significantly and to a lesser extent the ductility. Finally a finite element model has been built to predict the linkage between voids.

Dans le domaine de la recherche de gènes causaux, il est maintenant connu que plusieurs caractères complexes peuvent en fait être influencés par une multitude de gènes. Dans ce mémoire, nous présentons l'adaptation d'une méthode de cartographie génétique fine à la cartographie de caractère polygénique. Nous présentons tout d'abord un aperçu de certains outils statistiques utilisés en génétique. En particulier, certaines mesures d'association généralement employées en cartographie génétique. Puis, nous présentons la méthode de cartographie que nous souhaitons adapter: méthode qui suppose que le caractère est causé par l'effet d'un seul gène. Nous supposons plutôt que le caractère est causé par la combinaison de deux gènes. Après avoir présenté notre modélisation et les aspects théoriques de l'adaptation proposée, nous utilisons des données simulées pour tester nos développements. Nous comparons aussi nos résultats avec ceux obtenus avec une mesure d'association, ainsi qu'avec la méthode de cartographie dont nous proposons une adaptation. Les résultats démontrent la nécessité de développer des méthodes de cartographie génétique adaptées aux caractères polygéniques ; avec quelques améliorations concernant l'inférence des génotypes aux gènes causaux, notre adaptation devrait offrir de meilleurs résultats que les autres méthodes présentées. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : statistique génétique, cartographie génétique, caractère polygénique, processus de coalescence, arbre de recombinaison ancestral

Located in this technology island, not only because of large global IT products but also due to product relationship between industrial characteristics, business models need to change in response to raise the necessary decisions to respond more quickly. Therefore need to review internal auxiliary production resources, such as:collaboration with the organization of the mobilization, equipment and information systems support; but unfortunately in the past often used in information systems into the methodology of "process oriented", this method has become standardized work considerable help, but the allocation of resources to response or re-engineering the organization become very rigid and difficult but flexible rapid reaction with the corporate policy, so we need a rapid reaction response to organizational decision-making methodology of doing things. This research adopts enterprise architecture method to construct the service-oriented manufacturing resource planning management model (SOMRPMM). By this methodology, "service-oriented" is used to replace the traditional "process-oriented" planning. From the very beginning of manufacturing resource planning, the services, operations, and task behaviors are all clearly defined as to meet the realm of structure-behavior coalescence (SBC). Policy makers can create cost-effective coordination of resource allocation, adjusting the organizational changes to enhance the implementation of reengineering and organizational learning.

Down the ages, business management has been intensely discussed, but to implement is not always correct. The organization is like a building. A great building needs a great plan which is usually called a blueprint. Blueprint describes the relationship among components of the building. And the workers must follow the plan to finish the building. A decision-maker is an architect of the enterprise; he must have an enterprise-blueprint inside his mind first, then to govern the enterprise effectively later. This study focuses on constructing an Architecture-Oriented Petroleum Business Management Model (AOPBMM) which uses structure-behavior coalescence (SBC) architecture to model the organization of petroleum business. By integrating the organizational structure and organizational, AOPBMM is able to achieve a desirable blueprint of an enterprise. Overall, since SBC architecture is good at describing complicated business and makes the petroleum business easy to understand AOPBMM helps an enterprise communicate better and execute faster.

In these days, container yards always operate in the face of large-scale, high volume, and shipping companies requesting a large number of containers to be handled in a short time. It then becomes an important issue that container yards are able to maximize the operational capability under the limited land to meet the ship company¡¦s "accurate, fast, stable and cheaper" demand.In addition to having the general functions of storage yards, container yards, due to close to wharf berth line, also need to support the loading and unloading of containers. In such a situation, container yards shall play an important indicator of ports¡¦ operational efficiency. In this study, we construct an Architecture-Oriented Container Yard Management Model (AOCYMM) which is based on the six fundamental diagrams of Structure-Behavior Coalescence (SBC) architecture. AOCYMM, not only provides an integrated description of structure and behavior on the container yard operations, but also helps the employees within the enterprise easily manage the container yard business. AOCYMM covers all structure and behavior of the whole container yard operations. Therefore, AOCYMM describes the complete picture of container yards so that every employee shall understand and communicate well to achieve the high quality and efficiency of overall management. Keywords: Architecture-Oriented Container Yard Management Model, Structure-Behavior Coalescence (SBC) Architecture

Cloud computing has updated the appearance of the Information Technology (IT) infrastructure, and in addition to lower operation costs provides real-time services and reduces the information service barrier. In order to adapt to the rapidly changing market demand, enterprises are beginning to consider the feasibility of the deployment of cloud computing. The business environment changes so fast that an integrated dynamic framework and intelligent service system to achieve enterprises¡¦ visions, objectives and strategies, and to quick response is needed. Regarding to the three main service types of cloud computing: Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure as a Service (IaaS), this study proposes an integration model for enterprise administration. Cloud computing packs the functionality of dynamic resource adjustment. From the deployment of organizations to the customer interactions, cloud computing is divided into a public cloud, private cloud and mixed cloud based on its deployment model. As for the private cloud, its information security and efficiency allow enterprises execute their operations smoothly according to the business rules. Thus more and more enterprises are inclined to deploy private clouds. This study uses structure-behavior coalescence architecture description language (SBC-ADL) to accomplish the systems architecture, and provides thorough suggestions of dynamic resources allocation as a reference model for any enterprise which plans to deploy the cloud computing service. For those enterprises that have already implemented cloud computing services, the systems architecture can be referred to better their business management. Describing the relationship between the various systems architecture is helpful in quickly understanding the system operation. Consider reducing misunderstanding and increasing work efficiency and information correctness, SBC-ADL works very well as an effective tool for training and communication within the IT department.

This thesis is about the phase transformation, shape, size distribution and coalescence of TiO2 (part I) and ZrO2 (part II) nanopartilces produced by Nd-YAG laser ablation on metal targets under oxygen background gas, and characterized by analytical electron microscopy. The optimum laser ablation condition that satisfactory and routinely yield high-pressure phases of TiO2 (i.e. £\-PbO2-type and fluorite-related structures) and ZrO2 with high residual stress were reported. Part I-1 focuses on physical coagulation, by Van der Waals force, of the TiO2 condensates at temperatures up to about 1000 K as a result of post-condensation radiant heating. In part I-2, imperfect oriented attachment of nanoparticles over specific surfaces is rationalized to cause accretion and defects for the rutile condensates. Brownian motion may proceed above a critical temperature for anchorage release at the interface of imperfect attached nanoparticles until an epitaxial relationship is reached. Part I-3 deals with further the Brownian-type rotation of the imperfectly impinged £\-PbO2-type TiO2 and rutile nanocondensates until interfacial-energy cusp was reached. In part I-4 laser ablation condensation synthesis of dense TiO2 polymorphs and their phase transformations were documented. Part II-1 is about dense tetragonal (t)-ZrO2 and cubic (c-) nanocondensates which were synthesized under very rapid heating and cooling by pulsed Nd-YAG laser ablation with oxygen background gas. The t-ZrO2 nanoparticles were found to form deformation twins/faults and followed unique transformation path upon local electron dosage. Electron diffraction indicated that the dense c- and t- phase with specific size and residual stress were allowed to relax and/or kinetically phase change into lower-energy state as constrained by the intersections of the internal energy vs. cell volume plots calculated for the two polymorphs (Part II-2).

Nanocrystalline £\-Zr condensates deposited by ion beam sputtering on the NaCl (100) surfaces and then annealed at 100 ¢J to 750 ¢J in air. The phases present were identified by transmission electron microscopy to be nanometer-size £\-Zr+ZrO¡B£\-Zr+ZrO+c-ZrO2¡Bc-ZrO2¡Bc-+t-ZrO2¡Bt-ZrO2¡Band t-+m-ZrO2 phase assemblages with increasing annealing temperature. The zirconia showed strong {100} preferred orientation due to parallel epitaxy with NaCl (100) when annealed between 150 ¢J and 500 ¢J in air. The c- and t-zirconia condensates also showed (111)-specific coalescence among themselves. The c- and/or t-ZrO2 formation can be accounted for by the small grain size, the presence of low-valence Zr cation and the lateral constraint of the neighboring grains. (Part 1) Nanocrystalline £\-Zr condensates were deposited on the NaCl (100) plane at 25 to 450 ¢J by radio frequency ion beam sputtering from a pure 99.9¢H Zr disk. The nano condensates were identified by transmission electron microscopy to be quasiamorphous, £\-Zr, £\-Zr+ZrO and £\-Zr+ZrO+c-ZrO2 phase assemblages with increasing substrate temperature. At 400 ¢J and under 1-20 sccm oxygen, c- and t-ZrO2 nanocondensates were assembled on NaCl (100) as monolayer nanocrystalline material and showed strong preferred orientation. The c- and/or t-ZrO2 were retained by small grain size, low-valence Zr cation and 2-D matrix constraint of the film. (Part 2) Nanosized c- and t-ZrO2 were formed as monolayer nanocrystalline film on NaCl (100) plane by radio frequency ion beam sputtering. The microstructure and the epitaxy relationship with the NaCl (100) plane were studied by a high resolution transmission electron microscope. The epitaxy orientation was found to be [001]Z//[001]N, [100]Z//[1 0]N (group A), and [011]Z//[001]N, [100]Z//[100]N (group B) between zirconia (Z) and NaCl (N). Group B has two variants and is the dominant type. The possible causes for the epitaxy relationship are discussed. Crystallites within the same group can merge by rotation and coalesce into a single crystal, whereas crystallites in different groups can form high-angle grain boundaries. (Part 3) Special interfaces were formed for the c- and/or t-ZrO2 (Z) nano-crystals when deposited on the NaCl (N) (100) cleavage plane by ion beam sputtering to follow the epitaxy relationships of [001]Z//[001]N, (100)Z//(1 0)N (group A); and [011]Z//[001]N, (100)Z//(100)N (group B1) or (100)Z//(010)N (group B2). The nanoparticles in group A and B were impinged and coalesced to form {220}A/{200}B and {200}A/{111}B interfaces; with anchored dislocation whereas those in group B1 and B2 form {220}B1/{200}B2 interface. The {220}A/{200}B interface is found to be of especially low energy due to good match O2¡V lattice sites, and smoothly joints {200} and {220} planes across the interfaces without mismatch strain and dislocations. The special interfaces may shed light on the epitaxial mechanism of nanocrystalline materials in general. (Part 4)

In recent years, owing to the network, the information science and technology, and the telecommunications industry¡¦s popularization, make the interaction between the enterprise and customer more frequent. The more rigid marketing to customer, the higher quality of product and service can be applied to the customer request. How to dig the high profit rate customer in the general customer group, and maintain good the interaction relationship among the customers, and detain the customers who already has, then becomes a key factor to adopt the Customer Relationship Management (CRM). In this research, we find that most CRM literature focused on the case study, or adopted the key factor discussion and the theory of CRM elaboration. The main purpose of our research is to define and develop an Architecture-Oriented Customer Relationship Management Model (AOCRMM). This model provides the information and the way of construction of CRM, so the entire construction of CRM can be easily understood. In this model, the sufficient defines structure element, service of structure element, and behavior. By unifying structure viewpoints, behavior viewpoints, and resources viewpoints, we then construct the structure behavior coalescence for a CRM system. We conclude that AOCRMM, being an integration model, enables enterprise organization, IT system, internal business processes, etc. to be a single interface to customers. Also AOCRMM makes the organization behavior be presented easily the resources be used effortlessly, and have a good communication effect of organization and external environment.