Inom logistiken finns det ett antal frågeställningar, där struktur är ett viktigt område. Strukturfrågorna berör bland annat distributionen, exempelvis om en färdig produkt ska lagras eller skickas direkt till kund samt hur själva transporten ska gå till. Distributionsstrukturen syftar till att överbrygga de gap som uppstår mellan producerande företag och konsumenterna av deras produkter. Distributionen är i vissa fall mer komplex än företaget i sig, där beslutsfattandet kan handla om vilka delar som ska outsourcas och vilka externa aktörer som ska samarbetas med. Val gällande distributionen kan vara avgörande för ett företags lönsamhet och konkurrensförmåga. Många författare hävdar dock att det inte finns någon generellt svar på hur en distributionsstruktur ska utformas utan det skiljer sig från situation till situation, exempelvis beroende på produkttyp. En produktgrupp där distributionen är av betydelse är FMCG-varor, då de bland annat kännetecknas av en stor geografisk spridning, hög lageromsättningshastighet och frekventa inköp.

In this thesis, I analyze the weight functions used in moving least squares (MLS) methods to construct implicit surfaces that interpolate or approximate polygon soup. I found that one previous method that presented an analytic solution to the integrated moving least squares method has issues with degeneracies because they changed the weight functions to decrease too slowly. Inspired by their method, I derived a bound for the choice of weight function for implicit moving least squares (IMLS) methods to avoid these degeneracies in two-dimensions and in three-dimensions. Based on this bound, I give a theoretical proof of the correctness of the moving least squares interpolation and approximation scheme with weight function used in Shen et al. when used on closed polyhedrons. Further, previous IMLS implicit surface reconstruction algorithms that ll holes and gaps create surfaces with obvious bulges due to an intrinsic property of MLS. I propose a generalized IMLS method using a Gaussian distribution function to re-weight each polygon, making nearer polygons dominate and reducing the bulges on holes and gaps.

When a microphone array is mounted on a mobile aerial platform, such as an unmanned aerial vehicle (UAV), most existing beamforming methods cannot be used to adequately identify continuous and impulsive ground. Here, numerical simulation results and laboratory experiments are presented that validate a proposed time-frequency beamforming method based on the Multiple Signal Classification (MUSIC) algorithm to detect these acoustic sources from a mobile aerial platform. In the numerical simulations three parameters were varied to test the proposed algorithm’s location estimation performance: 1) the acoustic excitation types; 2) the moving receiver’s simulated flight conditions; and 3) the number of acoustic sources. Also, a distance and angle error analysis was done to quantify the proposed algorithm’s source location estimation accuracy when considering microphone positioning uncertainty. For experimental validation, three laboratory experiments were conducted. Source location estimations were done for: a 600 Hz sine source, a banded white noise source between 700-800 Hz, and a composite source combined simultaneously with both the sine and banded white noise sources. The proposed algorithm accurately estimates the simulated monopole’s location coordinates no matter the excitation type or simulated trajectory. When considering simultaneously-excited, multiple monopoles at high altitudes, e.g. 50 m, the proposed algorithm had no error when estimating the source’s locations. Finally, a distance and angle error analysis exposed how relatively small microphone location error, e.g. 1 cm maximum error, can propagate into large averaged distance error of about 10 m in the far-field for all monopole excitation types. For all simulations, however, the averaged absolute angle error remained small, e.g. less than 4 degrees, even when considering a 5 cm maximum microphone location error. For the laboratory experiments, the sine source had averaged distance and absolute angle errors of 0.9 m and 14.07 degrees from the source’s true location, respectively. Similarly, the banded white noise source’s averaged distance and absolute angle errors were 1.9 m and 47.14 degrees; and lastly, the averaged distance and absolute angle errors of 0.78 m and 8.14 degrees resulted when both the sources were simultaneously excited.

The application of membrane bioreactors (MBRs) to wastewater treatment is increasing due to their ability to operate at high biomass concentrations and to deliver effluents of high quality. The major challenges associated with the application of MBRs is fouling which can shorten the useful life of the membrane, increase in the amount of energy consumed, and the cost for membrane cleaning. The main reasons for fouling are the deposition of solids as a cake layer, pore plugging by colloidal particles, adsorption of soluble compounds and biofouling. Fouling is a particular problem for activated sludge membrane bioreactors (AS-MBRs) since this process deals with liquors having a high concentration of total solids as well as dissolved compounds such as extracellular polymeric substances (EPS). The combination of a moving bed biofilm reactor and a membrane reactor (MBBR-MR) has significant potential. It may be considered as a compact wastewater treatment process which can compensate for the drawbacks of AS-MBRs. Readily biodegradable COD is removed in the MBBR while particulate matter is separated by the membrane. To further reduce the membrane fouling the effects of adding an intermediate coagulation stage was investigated critically on membrane fouling. The present study includes an overall assessment of the performance of a combined MBBR-MR system, based on the chemical oxygen demand (COD) removal efficiency and membrane fouling mechanism. The required test runs were conducted using pilot-scale MBBR and ultra filtration membrane. The pilot MBBR had a working volume of 1.8 m3 with a 60% carrier fill fraction. The MBBR was operated with loading rate of 78 ± 21 g/m2/d (HRT of 4 h). The ultra-filtration was spiral wound and composed of polyethersulfone (PES) with a pore size of 0.03 microns. The MBBR feed was obtained from a final treated wastewater effluent in a food processing plant located in SW Ontario. In this research, ferric chloride was also employed as a coagulant and influences of different coagulant doses and permeate fluxes on membrane fouling were studied. Based on the experimental results, it was found that the combination of MBBR with membrane filtration can produce a constant high quality permeate that is appropriate for water reuse purposes. The composition analysis of permeate showed that the stream is free of suspended solids and the average COD turns to 75 ± 25 mg/l. In addition, the MBBR had a SCOD removal of 76% ± 7% which is considered as a reasonable efficiency for a single reactor. Operating the membrane without adding coagulant caused rapid fouling in a short time period and the Trans Membrane Pressure (TMP) reached the maximum allowable pressure of 10 psi. However, addition of coagulant was found to decrease the fouling of the membrane as well as increasing the filtration time. The extent of the pre-coagulation effect on membrane fouling was found to strongly depend on the dosage of the coagulant and the MBBR effluent characteristics. A coagulant dose of 400 mg/l with a permeate flux of 7.6 LMH performed the best at reducing membrane fouling. Colloidal fouling was found to be a significant fouling mechanism at low coagulant dose (e.g. 200 mg/l), while cake formation appeared to be mainly responsible for fouling at higher coagulant doses. Permeate flux was found to have a significant effect on the fouling of the membrane. The presence of colloidal matters at low fluxes and TSS at higher fluxes were responsible for fouling of the membrane by blocking the pores and formation of the cake layer on the membrane surface, respectively. Then later addition of Dissolved Air Flotation (DAF) inside the factory had a noticeable effect on wastewater characteristics and consequently on fouling of the membrane. A 22% and 31% improvement in TCOD and TSS in the wastewater was observed leading to reduction in the fouling.

This dissertation challenges an unstudied area in moving objects database domains; predicting (long-term) future locations of moving objects. Moving object prediction enables us to provide a wide range of applications, such as traffic prediction, pre-detection of an aircraft collision, and reporting attractive gas prices for drivers along their routes ahead. Nevertheless, existing location prediction techniques are limited to support such applications since they are generally capable only of short-term predictions. In the real world, many objects exhibit typical movement patterns. This pattern information is able to serve as an important background to tackle the limitations of the existing prediction methods. We aims at offering foundations of pattern-aware prediction for moving objects, rendering more precise prediction results. Specifically, this thesis focuses on three parts. The first part of the thesis studies the problem of predicting future locations of moving objects in Euclidean space. We introduce a novel prediction approach, termed the hybrid prediction model, which utilizes not only the current motion of an object, but also the object's trajectory patterns for prediction. We define, mine, and index the trajectory patterns with a novel access method for efficient query processing. We then propose two different query processing techniques along given query time, i.e., for near future and for distant future. The second part covers the prediction problem for moving objects in network space. We formulate a network mobility model that offers a concise representation of mobility statistics extracted from massive collections of historical objects trajectories. This model captures turning patterns of the objects at junctions, at the granularity of individual objects as well as globally. Based on the model, we develop three different algorithms for predicting the future path of a mobile user moving in a road network, named the PathPredictors. The third part of the thesis extends the prediction problem for a single object to that for multiple objects. We introduce a convoy query that retrieves all groups of objects, i.e., convoys, from the objects' historical trajectories, each convoy consists of objects that have traveled together for some time; thus they may also move together in the future. We then propose three efficient algorithms for the convoy discovery, called the CuTS family, that adopt line simplification methods for reducing the size of the trajectories, permitting efficient query processing. For each part, we demonstrate comprehensive experimental results of our proposals, which show significantly improved accuracies for moving object prediction compared with state-of-the-art methods, while also facilitating efficient query processing.

Deploying networks onboard mass transit vehicles, e.g., trains and buses, and connecting these moving networks to the Internet using cellular or satellite services is seen as a promising technology to meet the broadband demand in the transport sector. This thesis examines the problem of service disruptions in such moving networks and proposes original solutions to manage these disruptions. Service disruption is caused by two main reasons. First, unavailability of wireless resources in the back-haul (e.g., cellular) network may prevent successful handoff of all passenger calls when the moving vehicle leaves the coverage of one radio tower and enters another. Second, the moving network may face frequent network disconnections when the vehicle travels through environments with severe radio coverage, e.g., tunnels. To address service disruption due to back-haul resource unavailability, an advance resource reservation frame work is proposed whereby the deterministic mobility of public transport is exploited to reserve bandwidth at target radio base-stations along the route of the vehicle. The viability of the proposed advance reservation framework is validated by (i) a predictability analysis of resource demand of moving networks using real passenger data, which show that it is possible to predict the demand at an upcoming location based on past data and the passenger count at the current location, and (ii) design of cellular extensions which show that advance reservation not only can be accommodated in existing cellular architectures with minimal modifications, it can also guarantee fast and scalable access to massive reservation data. To quantify the effect of temporary network disconnections on the probability of service disruption, new mathematical models have been developed and validated by computer simulation. Finally, a novel proactive admission control (PAC) approach is proposed to reduce service disruption by rejecting new call requests when the moving network is predicted to experience an imminent network disconnection. Using mathematical modelling, it has been demonstrated that PAC reduces service disruption probability exponentially as a function of the proactive decision time with only a linear increase in the new call blocking probability. Practicality of the PAC approach has been confirmed using empirical data from actual vehicular trips.

Double Bind: Splitting identity and the body as an object is a research project consisting of studio-based practice presented mainly in video installation format. This work looks at hysterical symptoms as a performance of a body’s split identity. The project draws on the Lacanian theory of Mirror Stage which proposes that the self experienced by the subject, and the image of that self (represented in a mirror-like reflection, or an image) are different to each other, and the development of self-awareness as misrecognition of one’s self. As a conspicuous example of split body, Chapter One describes how the hysterical body, in clinical and artistic representation, is dissociated into multiple selves. In Chapter Two, I discuss some examples of contemporary performance artists who use themselves as subjects, but whose bodies become objects that do not portray the self. In the final chapter I explain how, in my video work, I objectify my own body and how I assess whether this is a mode of self-portraiture. / During the course of this research, I studied a wide range of medical resources and psychoanalytical literature, much of which employed visual illustration and documentation. For example, I have drawn inspiration from Jean-Martin Charcot’s photographic documents of female hysterics whom he treated as patients at the French hospital of La Salpêtrière in the late 19th century; in particular the figure of his most famous patient, known as Augustine. My research also involved studio-based investigation, such as experimentations with the performance of my own body in video format, and the contextual study of artistic and critical texts relating to contemporary media art. / The aim of this research is to demonstrate the ways in which my video performances split the body, creating an Other within one body that can be compared with the hysterical body of a patient, like Augustine, performing for her doctor. In this condition, I perform as the subject and the object of the gaze at the same time. My self-portrait is split in this way: it creates a body double, which I misrecognise as myself. But in doing so, I am both the director and the performer of the image. This is the double bind that my video work puts me into.

This dissertation challenges an unstudied area in moving objects database domains; predicting (long-term) future locations of moving objects. Moving object prediction enables us to provide a wide range of applications, such as traffic prediction, pre-detection of an aircraft collision, and reporting attractive gas prices for drivers along their routes ahead. Nevertheless, existing location prediction techniques are limited to support such applications since they are generally capable only of short-term predictions. In the real world, many objects exhibit typical movement patterns. This pattern information is able to serve as an important background to tackle the limitations of the existing prediction methods. We aims at offering foundations of pattern-aware prediction for moving objects, rendering more precise prediction results. Specifically, this thesis focuses on three parts. The first part of the thesis studies the problem of predicting future locations of moving objects in Euclidean space. We introduce a novel prediction approach, termed the hybrid prediction model, which utilizes not only the current motion of an object, but also the object's trajectory patterns for prediction. We define, mine, and index the trajectory patterns with a novel access method for efficient query processing. We then propose two different query processing techniques along given query time, i.e., for near future and for distant future. The second part covers the prediction problem for moving objects in network space. We formulate a network mobility model that offers a concise representation of mobility statistics extracted from massive collections of historical objects trajectories. This model captures turning patterns of the objects at junctions, at the granularity of individual objects as well as globally. Based on the model, we develop three different algorithms for predicting the future path of a mobile user moving in a road network, named the PathPredictors. The third part of the thesis extends the prediction problem for a single object to that for multiple objects. We introduce a convoy query that retrieves all groups of objects, i.e., convoys, from the objects' historical trajectories, each convoy consists of objects that have traveled together for some time; thus they may also move together in the future. We then propose three efficient algorithms for the convoy discovery, called the CuTS family, that adopt line simplification methods for reducing the size of the trajectories, permitting efficient query processing. For each part, we demonstrate comprehensive experimental results of our proposals, which show significantly improved accuracies for moving object prediction compared with state-of-the-art methods, while also facilitating efficient query processing.

This thesis is a practical project which involves moving images and paintings together as a medium that explores phenomena of contemporary China relating to personal identity, independence and its relationship with the traditional importance on collective groups, group centredness and interdependence. The project’s approach draws on sociological research on Western thought, values and beliefs naturally occurring in China since the “Open Door” policy as raw data to focus on the transition and transformation of contemporary Chinese individuality, and translates these data to form the concepts underpinning the metaphoric method of my artwork. Bubbles are the main visual symbols that metaphorically imply the incessantly transformable Chinese individuality and social cultural identity. My aim has been to portray this phenomenon through artistic practices on screens. By reflecting and engaging with moving images and paintings, underpinned by theoretical research and methods including data collecting, self-reflecting on data, practical manifestation and self-inquiry, I have attempted to unfold the phenomenon of contemporary Chinese individuality through my art practice. The thesis is composed as a creative work of moving images accompanied by an exegesis component. The moving image represents a nominal 80%, and the exegesis 20% of the final submission.

Thesis (Ph.D.) -- University of Adelaide, Dept. of Social Inquiry, 2000? / Bibliography: leaves 261-276. Also available in an electronic version via the Internet (ADT).