In the US, breast cancer is one of the most common forms of cancer and the most lethal. There are many decisions that must be made by the doctor and/or the patient when dealing with a potential breast cancer. Many of these decisions are made under uncertainty, whether it is the uncertainty related to the progression of the patient's health, or that related to the accuracy of the doctor's tests. Each possible action under consideration can have positive effects, such as a surgery successfully removing a tumour, and negative effects: a post-surgery infection for example. The human mind simply cannot take into account all the variables involved and possible outcomes when making these decisions. In this report, a detailed Partially Observable Markov Decision Process (POMDP) for breast cancer screening decisions is presented. It includes 151 states, covering 144 different cancer states, and 2 competing screening methods. The necessary parameters were first set up using relevant medical literature and a patient history simulator. Then the POMDP was solved optimally for an infinite horizon, using the Perseus algorithm. The resulting policy provided several recommendations for breast cancer screening. The results indicated that clinical breast examinations are important for screening younger women. Regarding the decision to operate on a woman with breast cancer, the policy showed that invasive cancers with either a tumour size above 1.5 cm or which are in metastasis, should be surgically removed as soon as possible. However, the policy also recommended that patients who are certain to be healthy should have a breast biopsy. The cause of this error was explored further and the conclusion was reached that a finite horizon may be more appropriate for this application.

A prática da terceirização por parte das organizações é uma prática que vem sendo utilizada há muito tempo e tornou-se cada vez mais comum. O mesmo também ocorre em relação às atividades da área de Tecnologia da Informação, onde a terceirização vem assumindo um papel muito importante devido ao crescente do número de organizações que estão adotando essa prática, pelos altos valores envolvidos, ou ainda pelas formas e arranjos cada vez mais inovadores que estão se formando para o uso da terceirização em Tecnologia da Informação. Diversas são as razões que levam os executivos a terceirizar, parcial ou totalmente, suas diferentes atividades e serviços na área de Tecnologia da Informação, sendo muitos deles associados à redução de custos, melhoria da qualidade e foco nas atividades essenciais da organização. Os motivos pelos quais as organizações terceirizam têm sido foco de muitos estudos acadêmicos importantes, onde ainda são identificadas lacunas. Nesse contexto, esse trabalho tem como objetivo identificar os fatores que impactam no processo decisório de terceirização em Tecnologia da Informação. Diante disso, o principal objetivo desse trabalho é responder a seguinte questão: “que fatores devem ser considerados no processo decisório com relação à terceirização da Tecnologia da Informação?” O método utilizado é a pesquisa Survey realizada a partir de modelo validado apresentado pela literatura. O modelo precisou ser adaptado para a área de Tecnologia da Informação que resultou em um modelo contendo as dimensões Estratégia, Ambiente, Características do Processo de TI, Custos e Características dos Fornecedores de TI. O processo de validação e refinamento utilizou a Análise de Confiabilidade (Coeficiente Alfa de Cronbach) e a Análise Fatorial Fatorial Exploratória Discriminante. A partir do resultado dessas validações, foi identificada a possibilidade de diferentes arranjos de variáveis nas dimensões, porém manteve-se o padrão validado anteriormente sem contestação da literatura. A principal contribuição dessa pesquisa para a prática gerencial é o fornecimento de um instrumento de pesquisa que prevê a identificação dos fatores de determinantes para a decisão de terceirização em Tecnologia da Informação sob o ponto de vista do contratante. Esse modelo pode auxiliar executivos na decisão de terceirizar ou não suas atividades em Tecnologia da Informação. / The practice of Outsourcing by organizations has been used for years and it has become more and more common. The same has been occurring as to the activities of the area of Information Technology , where outsourcing has been occupying a relevant role due to the growing number of companies adopting this practice, the high amounts involved, or, still, the more and more innovating arrangements being formed for the utilization of outsourcing in Information Technology. There are many reasons for the executives to outsource, partial or totally, their different activities and services in the area of Information Technology, many of them associated to the reduction of costs, quality improvement and focus on the essential activities of the organization. The reasons why the organizations outsource have been the focus of many important academic studies, where many gaps are still identified. In this context, the aim of this work is to identify the factors that impact on the decision process of outsourcing in Information Technology. So, the main goal of this work is the answer for the following question: “which are the factors that shall be considered in the decision process relating to outsourcing of Information Technology".This work is a survey developed from a validated model presented by literature. The model had to be adapted to the area of Information Technology and resulted in a model containing the following dimensions: Strategy, Environment, Features of IT Process, Costs and Characteristics of IT Suppliers. The validation process and refining have used the Analysis of Reliability (Cronbach's Alpha Coefficient) and Discriminatory and Exploratory Factor Analysis. From the result of such validations, it has been identified the possibility of different arrangements of variables in the dimensions; however, it has been maintained the standard previously validated with no objection of literature. The main contribution of this survey for the managerial practice is the supply of a research tool which foresees the identification of determinant factors for outsourcing in Information Technology related to the contractor. The model may help the executives decide on the outsourcing of their activities related to Information Technology or not.

Atualmente as dificuldades enfrentadas no deslocamento urbano são consideradas um dos maiores problemas, especialmente nas grandes cidades. O planejamento adequado do sistema de transporte urbano é fundamental para minimizar o tempo e os custos de deslocamento, melhorando a qualidade de vida da população e o próprio funcionamento do ambiente urbano. Esta dissertação parte da premissa de que os sistemas de simulação podem ser usados para estudar diferentes alternativas para melhorar o sistema de transportes, servindo de base para a tomada de decisão de forma a otimizar o deslocamento urbano. Assim, este trabalho apresenta o desenvolvimento de um modelo computacional para simulação do sistema de transporte urbano. O modelo proposto combina características dos modelos mesoscópicos e microscópicos, incluindo o comportamento dos usuários no planejamento da rota. Um framework para o desenvolvimento de aplicações de simulação é descrito e testado com uma implementação usando como cenário o Metropolitano de São Paulo (Metrô), considerando-se dados da pesquisa Origem-Destino para teste e validação do modelo aqui proposto. / Nowadays, the difficulties faced in urban displacement are considered a major problem, especially in big cities. Proper urban transport system planning is essential to minimize travel time and costs, improving quality of life and improving the urban environment. This master thesis starts from the premise that simulation systems can be used to study different alternatives to improve the transport system, so that decision making can be better justified and can optimize the urban displacement. Therefore, this work presents the development of a computational model for urban transportation system simulation. The model proposed targets mesoscopic and microscopic models, including user behaviors of route planning. A framework for development of simulation applications is described, with an implementation using as scenario the Metropolitano of São Paulo (Metrô) for model testing, considering data from Origem-Destino survey for testing and validation of the model here proposed.

Medical decision problems are extremely complex owing to their dynamic nature, large number of variable factors, and the associated uncertainty. Decision support technology entered the medical field long after other areas such as the airline industry and the manufacturing industry. Yet, it is rapidly becoming an indispensable tool in medical decision making problems including the class of sequential decision problems. In these problems, physicians decide on a treatment plan that optimizes a benefit measure such as the treatment cost, and the quality of life of the patient. The last decade saw the emergence of many decision support applications in medicine. However, the existing models have limited applications to decision problems with very few states and actions. An urgent need is being felt by the medical research community to expand the applications to more complex dynamic problems with large state and action spaces. This thesis proposes a methodology which models the class of sequential medical decision problems as a Markov decision process, and solves the model using a simulation based reinforcement learning (RL) algorithm. Such a methodology is capable of obtaining near optimal treatment strategies for problems with large state and action spaces. This methodology overcomes, to a large extent, the computational complexity of the value-iteration and policy-iteration algorithms of dynamic programming. An average reward reinforcement-learning algorithm is developed. The algorithm is applied on a sample problem of treating hereditary spherocytosis. The application demonstrates the ability of the proposed methodology to obtain effective treatment strategies for sequential medical decision problems.

<p>Undersökningen visar att Folkuniversitetet i Stockholms befintliga kunder beter sig på ett likartat sätt, i alla olika steg i köpbeslutsprocessen, när de söker information om Folkuniversitetet samt deras kurser och utbildningar. Hemsidan är den informationskanal som de flesta kunderna kommer i kontakt med, följt av kurskatalogen, personalen och rekommendationer från vänner och bekanta.</p> / <p>The survey shows that the current customers of Folkuniversitetet in Stockholm act in a similar way, in all steps of the buyer behaviour process, when they search for information about Folkuniversitetet, their courses and educations. The homepage is the main information channel that most of the customers get in contact with, followed by the course catalogue, the personnel and recommendations from friends and acquaintances.</p>

Internet har gett konsumenten en ökad marknadsöverblick, detta har gjort det möjligt för konsumenten att tidseffektivt jämföra varor och priser. Detta innebär att dagens konsument är betydligt mer välinformerad. Medvetenheten om en stor och komplex marknad innebär ett högt engagemang och upplevda risker med köpet som följd. Det höga engagemanget innebär att man som konsument bland annat söker information för att reducera sina upplevda risker. Syftet med denna uppsats är att minska luckan i teorin om konsumenters riskreducerande informationsbeteende då de genomgår köpprocessen i ett högengagemangsköp. Ett annat syfte blir att se vilka faktorer kring informationssökning som är viktiga i ett högengagemangsköp på en högteknologisk marknad 2006. Frågeställningarna rör hur informationskällorna används idag, vilka källor som har störst utvecklingspotential och hur denna utveckling borde se ut. Datainsamlingen har skett genom åtta stycken djupintervjuer med studenter från olika fackhögskolor inom Högskolan i Jönköping. Studenternas köp av bärbar dator de senaste sex månaderna har fungerat som underlag för samtalet. Resultatet av studien visar att de ekonomiska och funktionella riskerna upplevts vara mycket omfattande för våra respondenter. Priset och krav på en viss funktionalitet har varit klart tidigt i köpprocessen, funktionalitetskravet har under köpets gång översatts från behovsbeskrivningar till teknisk specifikation. Översättningen har mestadels inneburit omfattande informationssökning på Internet, vilket i många fall inneburit att respondenten engagerat personliga källor för värdering av informationen. De ekonomiska och funktionella riskerna har varit aktuella växelvis under köpprocessens olika faser. Många upplevelser kring informationssökningen präglas av ineffektivitet. Det icke kommersiella intryck information från jämförelsesajter på nätet och personliga källor ger ser vi ha en stor potential att kunna utvecklas. Detta genom att både i butik och på Internet aktivt arbeta med konsumenternas referenspunkter. Lösningsorienterad information med fokus på att bygga relationer och förtroende har enligt studiens resultat utvecklingsmöjligheter både på Internet och bland säljare. / Today Internet has given consumers a greater possibility to have a more complete view over the range of products on a market. This has made it possible to in a time effective way compare products and prices. A greater amount of perceived risks when purchasing a high involvement product is one result of the consumers awareness of a large and complex market. To reduce the risks, that high involvement purchase in hazes, the consumer is in a greater need of information The purpose with this essay is to reduce the gap in the theory about consumers risk reducing information behaviour when following the buyer’s decision process in a high involvement purchase. Another purpose is to identify which factors that is important when seeking information during a high involvement purchase. The ambition with this study is to explain how different information channels are used today, which channels that has the best potentials and how the development of these should be done. Interviews with eight different students from Jönköping University underlie the result of this study. During the six latest months these students have purchased a laptop, this purchase has been the starting-point for the interviews. The result shows that the economical and functional risks are those that have been most frequent perceived among our respondents. Requirements concerning price and functionality where also those things that the respondents early in the buyer’s decision process hade a clear opinion about. Functional requirements was in the beginning spoken out as to solve certain needs. These needs have been translated to technical specifications during the process. This translation has been signified with a lot of time spent on seeking information on the Internet. It has also engaged personal contacts with friends etc for valuation of information. Many experiences of information seeking shows on inefficiency. The non-commercial impression that personal contacts and Internet pages, with product comparison, has given is something that we mean should be taken to account in a larger scale. This could be done by actively working with the consumer’s frame of references to the product. Information that is solution orientated, on Internet and from sellers, and which has the ambition to build long term relations has from this reports result shown to have great possibilities for development.

Undersökningen visar att Folkuniversitetet i Stockholms befintliga kunder beter sig på ett likartat sätt, i alla olika steg i köpbeslutsprocessen, när de söker information om Folkuniversitetet samt deras kurser och utbildningar. Hemsidan är den informationskanal som de flesta kunderna kommer i kontakt med, följt av kurskatalogen, personalen och rekommendationer från vänner och bekanta. / The survey shows that the current customers of Folkuniversitetet in Stockholm act in a similar way, in all steps of the buyer behaviour process, when they search for information about Folkuniversitetet, their courses and educations. The homepage is the main information channel that most of the customers get in contact with, followed by the course catalogue, the personnel and recommendations from friends and acquaintances.

The purpose of this dissertation is to understand how algorithms can efficiently learn to solve new tasks based on previous experience, instead of being explicitly programmed with a solution for each task that we want it to solve. Here a task is a series of decisions, such as a robot vacuum deciding which room to clean next or an intelligent car deciding to stop at a traffic light. In such a case, state-of-the-art learning algorithms are difficult to employ in practice because they often make thou- sands of mistakes before reliably solving a task. However, humans learn solutions to novel tasks, often making fewer mistakes, which suggests that efficient learning algorithms may exist. One advantage that humans have over state- of-the-art learning algorithms is that, while learning a new task, humans can apply knowledge gained from previously solved tasks. The central hypothesis investigated by this dissertation is that learning algorithms can solve new tasks more efficiently when they take into consideration knowledge learned from solving previous tasks. Al- though this hypothesis may appear to be obviously true, what knowledge to use and how to apply that knowledge to new tasks is a challenging, open research problem. I investigate this hypothesis in three ways. First, I developed a new learning algorithm that is able to use prior knowledge to constrain the exploration space. Second, I extended a powerful theoretical framework in machine learning, called Probably Approximately Correct, so that I can formally compare the efficiency of algorithms that solve only a single task to algorithms that consider knowledge from previously solved tasks. With this framework, I found sufficient conditions for using knowledge from previous tasks to improve efficiency of learning to solve new tasks and also identified conditions where transferring knowledge may impede learning. I present situations where transfer learning can be used to intelligently constrain the exploration space so that optimality loss can be minimized. Finally, I tested the efficiency of my algorithms in various experimental domains. These theoretical and empirical results provide support for my central hypothesis. The theory and experiments of this dissertation provide a deeper understanding of what makes a learning algorithm efficient so that it can be widely used in practice. Finally, these results also contribute the general goal of creating autonomous machines that can be reliably employed to solve complex tasks.

Visuell kommunikation är något som vi dagligen stöter på i vår vardag, i form av reklamannonser, reklamfilmer, skyltningar osv. Skyltfönster är en form av visuell kommunikation, och fungerar som företagens ansikte utåt och det skapar ofta en första bild hos mottagaren om företaget.  Enligt köpbeslutsprocessen är reklam den mest förekommande informationskälla som påverkar konsumenten i köpbeslutet. Denna studie kommer att utgå från Kotler och Kellers modell av köpbeslutsprocessen, som följer kunden före, under samt efter köpet. Modellen består av fem steg, problem erkännande, informationssökning, utvärdering av valmöjligheter, köpbeslut och efter köpbeteende. Forskningen behandlar endast de tre första stegen som sker före själva köpet, och den klargör fönsterskyltnings påverkan på den kvinnliga köpbeslutsprocessen före köpet. Kopplingar dras mellan köpbeslutsprocessen och den visuella kommunikationens tre arbetsområden, text, form och färg. Avslutad studie tyder på en påverkande effekt av skyltfönster i köpbeslutsprocessen. Då studien visade på att antal spenderade pengar ökade beroende på antal shoppingtillfällen. Ett högre antal shoppingtillfällen innebär alltså en större påverkan i köpbeslutsprocessen då kunden oftare stöter på den visuella kommunikationen, skyltfönster. / Visual Communication is something we daily run into in our everyday lives, in form of advertising, commercials, billboards, etc. Storefronts are a form of visual communication, andserves as a corporate public face for the company. Often it creates a first image of the companies’ image among the receivers. According to the buying decision process, advertising is the most frequent source of information affecting the consumers buying decision. This study will be based on Kotler and Keller's model of the buying decision process, which follows the customer before, during and after the purchase. The model consists of five stages, problem recognition, information search, evaluation of options, purchase decision and post purchase behavior. This research only covers the first three steps that occur before the actual purchase, and it clarifies storefronts impacts on the female buying decision process before the purchase. Connections are drawn between the buying decision process and the visual communications three areas of work, text, shapes and colors. The completed study indicates an influence of storefronts in the purchase decision process. Since the study showed that the number of spent money increased depending on the number of shopping opportunities. A higher number of shopping opportunities is therefore a major influence in the buying decision process, when customer more frequently are getting exposed to storefronts when they shop.

Wireless technology has become an increasingly popular way to gain network access. Wireless networks are expected to provide efficient and reliable service and support a broad range of emerging applications, such as multimedia streaming and video conferencing. However, limited wireless spectrum together with interference and fading pose signi cant challenges for network designers. The novel technique of network coding has a significant potential for improving the throughput and reliability of wireless networks by taking advantage of the broadcast nature of wireless medium. Reverse carpooling is one of the main techniques used to realize the benefits of network coding in wireless networks. With reverse carpooling, two flows are traveling in opposite directions, sharing a common path. The network coding is performed in the intermediate (relay) nodes, which saves up to 50% of transmissions. In this thesis, we focus on the scheduling at the relay nodes in wireless networks with reverse carpooling. When two packets traveling in opposite directions are available at the relay node, the relay node combines them and broadcasts the resulting packet. This event is referred to as a coding opportunity. When only one packet is available, the relay node needs to decide whether to wait for future coding opportunities, or to transmit them without coding. Though the choice of holding packets exploits the positive aspects of network coding, without a proper policy in place that controls how long the packets should wait, it will have an adverse impact on delays and thus the overall network performance. Accordingly, our goal is to find an optimal control strategy that delicately balances the tradeoff between the number of transmissions and delays incurred by the packets. We also address the fundamental question of what local information we should keep track of and use in making the decision of of whether to transmit uncoded packet or wait for the next coding opportunity. The available information consists of queue length and time stamps indicating the arrival time of packets in the queue. We could also store history of all previous states and actions. However, using all this information makes the control very complex and so we try to find if the overhead in collecting waiting times and historical information is worth it. A major contribution of this thesis is a stochastic control framework that uses state information based on what can be observed and prescribes an optimal action. For that, we formulate and solve a stochastic dynamic program with the objective of minimizing the long run average cost per unit time incurred due to transmissions and delays. Subsequently, we show that a stationary policy based on queue lengths is optimal, and the optimal policy is of threshold-type. Then, we describe a non-linear optimization procedure to obtain the optimal thresholds. Further, we substantiate our analytical ndings by performing numerical experiments under varied settings. We compare systems that use only queue length with those where more information is available, and we show that optimal control that uses only the queue length is as good as any optimal control that relies on knowing the entire history.