Strategic network design has significant impacts on the operational performance of transportation and telecommunications industries. The corresponding networks are typically characterized by a multicommodity ow structure where a commodity is defined by a unique origin-destination pair and an associated amount of ow. In turn, multicommodity network design and hub location models are commonly employed when designing strategic networks in transportation and telecommunications applications. In this dissertation, these two modeling approaches are integrated and generalized to address important requirements in network design for truckload transportation and long-distance telecommunications networks. To this end, we first introduce a cost effective relay network design model and then extend this base model to address the specific characteristics of these applications. The base model determines relay point (RP) locations where the commodities are relayed from their origins to destinations. In doing this, we explicitly consider distance constraints for the RP-RP and nonRPRP linkages. In truckload transportation, a relay network (RP-network) can be utilized to decrease drivers' driving distances and keep them within their domiciles. This can potentially help alleviate the high driver turnover problem. In this case, the percentage circuitry, load-imbalance, and link-imbalance constraints are incorporated into the base model to control related performance metrics that are affected by the distance constraints. When compared to the networks from other modeling approaches, the RP-network is more effective in controlling drivers' tour lengths and capable of controlling the empty mileage to low levels without adding a large amount of additional travel distance. In telecommunications, an RP-network can be beneficial in long-distance data transfers where the signals' delity must be improved/regenerated at RPs along their travel paths. For this setting, we extend the base model to include fixed link setup costs and capacities. From our computational results, our models provide better network configuration that is cost effective and facilitates a better service quality (shorter delays and better connectivity). Concerning methodology, we develop effcient exact solution algorithms based on Benders decomposition, Lagrangean decomposition, and Lagrangean relaxation. The performance of the typical solution frameworks are enhanced via numerous accelerating techniques to allow the solution of large-sized instances in reduced solution times. The accelerating techniques and solution approaches are transferable to other network design problem settings with similar characteristics.

Network-on-Chip (NoC) is emerging as a promising communication structure, which is scalable with respect to chip complexity. Meanwhile, latest chip designs are increasingly leveraging multiple voltage-frequency domains for energy-efficiency improvement. In this work, we propose a simultaneous task and voltage scheduling algorithm for energy minimization in NoC based designs. The energy-latency tradeoff is handled by Lagrangian relaxation. The core algorithm is a clustering based approach which not only assigns voltage levels and starting time to each task (or Processing Element) but also naturally finds voltage-frequency clusters. Compared to a recent previous work, which performs task scheduling and voltage assignment sequentially, our method leads to an average of 20 percent energy reduction.

Studying dynamics of magnetization relaxation in excited magnetic materials is important both for understanding the rates and pathways of magnetization relaxation and for the potential use in spin-based electronics and data storage devices in the future. Previous studies have demonstrated that the size of nanocrystals is an important factor for energy relaxation in quantum dots and metal nanoparticles. Since magnetization relaxation is one of energy relaxation pathways, the size of nanocrystals may be also an important factor for magnetization relaxation in nanoscale magnetic materials. The goal of this study is to have a better understanding of magnetization relaxation in nanoscale magnetic materials. In particular, we focused on the correlation between the nanocrystal size and the rates of spin-lattice relaxation (SLR), a magnetization relaxation pathway, in magnetic nanocrystals. The size-dependent magnetization relaxation rate after optically induced demagnetization in colloidal Fe3O4 nanocrystals was measured by using time-resolved Faraday rotation (FR). Fe3O4 nanocrystals were chosen as the model system to study the correlation between the size of nanocrystals and the rates of SLR due to the wellestablished synthetic procedure of making nanocrystals with various sizes and narrow size dispersion. Faster SLR rates were observed in smaller Fe3O4 nanocrystals. The results suggested the surface of nanocrystals have higher efficiency of SLR than the interior region by using a simple model to analyze the SLR rates of Fe3O4 nanocrystals with various sizes. Higher efficiency of SLR at the surface may be due to the stronger spin-orbit coupling at the surface relative to the interior region. In addition to magnetization dynamics studies, the effect of oxidation on static FR in iron oxide nanocrystals (between Fe3O4 and y-Fe2O3) was studied. The results indicated FR signal is linearly correlated to the strength of optical transition between Fe2 and Fe3 in Fe3O4 for a given size of nanocrystals.

none

Recently, lifetime imaging has become a subject of intensive research. Lifetime is an important parameter to understand the dynamics of targeted objects and its applications ranges from fluorescence decay of biological objects to relaxation of semiconductor materials and devices. Many methods, such as time-correlated single photon counting (TCSPC) and phase detection in frequency domain, were developed to measure the characteristic lifetime. These methods are now rather matured and widely applied in various studies. However, these methods are only effective for lifetime longer than 100 picoseconds due to the bandwidth limitation of high-speed electronics. For even faster temporal resolution, novel techniques that do not rely on high-speed electronics will be required. In this study, we are integrating an autocorrelator with a galvo-based laser scanning microscope to enable imaging with very high temporal resolution. The principle and technique of pump-probe is implemented through the autocorrelator. In this way, imaging based pump-probe measurements can be realized. Specifically, we have applied the experimental setup so developed in measuring fluorescent dyes and semiconductor devices.

Microscopic magnetic resonance imaging (£gMRI) is a powerful tool for studying the structure and dynamics of various biological systems in vivo. A particular advantage of£gMRI related to this type of applications is that many different images can be acquired on the same object, each of them being important and meaningful. In this work, we report the results of applying£gMRI to the study of the biological process of a real plant (sesame seed). With weighting of longitudinal and transverse relaxation, diffusion as well as multi-quantum coherences, a series of images of sesame seed in the process of germination, with spatial resolution of several micrometers or tens of micrometers, has been obtained (examples given in Fig.1). The images are analyzed based on the physiological characteristics of the sesame seed. Some insights are drawn from these images obtained with different weightings.

In this study we consider an inventory routing problem in which a supplier distributes a single product to multiple retailers in a finite planning horizon. Retailers should satisfy the deterministic and dynamic demands of end customers in the planning horizon, but the retailers can backorder the demands of end customers considering the supply chain costs. In each period the supplier decides the retailers to be visited, and the amount of products to be supplied to each retailer by a fleet of vehicles. The decision problems of the supplier are about when, to whom and how much to deliver products, and in which order to visit retailers while minimizing system-wide costs. We propose a mixed integer programming model and a Lagrangian relaxation based solution approach in which both upper and lower bounds are computed. We test our solution approach with test instances taken from the literature and provide our computational results.

A ferromagnetic particle with single domain, at some kinds of applied field (at some angle or strangth), the particle's free energy would be two state model. The rate of barrier crossing could be solve by Fokker-Planck equation .And use master equation to find out the Total rate between two potential well. In this thysis, we use the upper method to simulate particle's magnetic moment under time varying magnetic field at fixed angle or fixed magnetic applied rotate the particle. In numerical method, we use the back Euler method to prevent the divergence of the calculation.

The purpose of this research is to develop a heuristic algorithm to minimize total costs in multi-commodity, multi-mode generalized networks with time windows problems. The proposed mathematical model incorporates features of the congestion of vehicle flows and time restriction of delivering commodities. The heuristic algorithm, HA, has two phases. Phase 1 provides lower and upper bounds based on Lagrangian relaxations with subgradient methods. Phase 2 applies two methods, early due date with overdue-date costs and total transportation costs, to search for an improved upper bound. Two application networks are used to test HA for small and medium-scale problems. A different number of commodities and various lengths of planning time periods are generated. Results show that HA can provide good feasible solutions within the reasonable range of optimal solutions. If optimal solutions are unknown, the average gap between lower and upper bounds is 0.0239. Minimal and maximal gaps are 0.0007 and 0.3330. If optimal solutions are known, the maximal gap between upper bounds and optimal solutions is less than 10% ranges of optimal solutions.

Thèse de doctorat : Physique : Tours : 2008. / Titre provenant de l'écran-titre.