Safety risk evaluation is critical in autonomous vehicle applications. This research aims to develop, implement, and validate new safety monitoring methods for navigation in Global Navigation Satellite System (GNSS)-denied environments. The methods quantify uncertainty in sensors and algorithms that exploit the complementary properties of light detection and ranging (LiDAR) and inertial measuring units (IMU). This dissertation describes the following four contributions. First, we focus on sensor augmentation for landmark-based localization. We develop new IMU/LiDAR integration methods that guarantee a bound on the integrity risk, which is the probability that the navigation error exceeds predefined acceptability limits. IMU data improves LiDAR position and orientation (pose) prediction and LiDAR limits the IMU error drift over time. In addition, LiDAR return-light intensity measurements improve landmarks recognition. As compared to using the sensors individually, tightly-coupled IMU/LiDAR not only increases pose estimation accuracy but also reduces the risk of incorrectly associating perceived features with mapped landmarks. Second, we consider algorithm improvements. We derive and analyze a new data association method that provides a tight bound on the risk of incorrect association for LiDAR feature-based localization. The new data association criterion uses projections of the extended Kalman filter's (EKF) innovation vector rather than more conventional innovation vector norms. This method decreases the integrity risk by improving our ability to predict the risk of incorrect association. Third, we depart from landmark-based approaches. We develop a spherical grid-based localization method that leverages quantization theory to bound navigation uncertainty. This method is integrated with an iterative EKF to establish an analytical bound on the vehicle's pose estimation error. Unlike landmark-based localization which requires feature extraction and data association, this method uses the entire LiDAR point cloud and is robust to extraction and association failures. Fourth, to validate these methods, we designed and built two testbeds for indoor and outdoor experiments. The indoor testbed includes a sensor platform installed on a rover moving on a figure-eight track in a controlled lab environment. The repeated figure-eight trajectory provides empirical pose estimation error distributions that can directly be compared with analytical error bounds. The outdoor testbed required another set of navigation sensors for reference truth trajectory generation. Sensors were mounted on a car to validate our algorithms in a realistic automotive driving environment. / Doctor of Philosophy / Advances in computing and sensing technologies have enabled large scale demonstrations of autonomous vehicle operations including pilot programs for self-driving cars on public roads. However, a key question that has yet to be answered is about how safe these vehicles really are. "Autonomously" driving millions of miles (with a trained safety driver taking over control to prevent potential collisions) is insufficient to prove fatality rates matching human performance, i.e., lower than 1 per 100,000,000 miles driven. The safety of an autonomous vehicle depends on the safety of its individual subsystems, components, connected infrastructure, etc. In this research, we evaluate the safety of the navigation subsystem which uses sensor information to determine the vehicle's location and orientation. We focus on light detection and ranging (LiDAR)and inertial measuring units (IMU). A LiDAR provides a point cloud representation of the environment by measuring distances to surrounding objects using beams of infrared light (laser beams) sent at regular angular intervals. An IMU measures the acceleration and angular velocity of the vehicle. We assume that a map of the environment is available. In the first part of this research, we extract recognizable objects from the LiDAR point cloud and match them with those in the map: this process helps estimate the vehicle's position and orientation. We identify the process' limitations that include incorrectly matching sensed and mapped landmarks. We develop new methods to quantify their impacts on localization errors, which we then reduce by incorporating additional IMU data. In the second part of this dissertation, we design and evaluate a new approach specifically aimed at provably increasing confidence in landmark matching, thereby improving vehicle navigation safety. Third, instead of isolating individual landmarks, we use the LiDAR point cloud as a whole and match it directly with the map. The challenge with this approach was in efficiently and accurately quantifying the confidence that can be placed in the vehicle's navigation solution. We tested these navigation methods using experimental data collected in a controlled lab environment and in a real-world scenario.

Kiekvieno banko viena iš pagrindinių veiklos sričių – paskolų išdavimas. Su kiekviena išduodama paskola bankas prisiima vieną svarbiausių savo veikloje rizikų – kredito riziką, kurios nuostoliai bankui gali būti labai dideli. To geriausias įrodymas – 2007 metais JAV ištikusi kredito rizikos krizė, nusidriekusi per visą pasaulį ir atnešusi milžiniškus nuostolius. Todėl kredito rizikos tikslus įvertinimas ir tinkamas valdymas yra ypatingai svarbus tiek komerciniams bankams, tiek bankus prižiūrinčioms institucijoms, kurios privalo užtikrinti stabilų finansinio sektoriaus vystymąsi. Tinkamų kredito rizikos vertinimo modelių naudojimas bankuose leidžia sumažinti kredito riziką, padidinti banko veiklos stabilumą ir patikimumą. Todėl yra aktualu išanalizuoti kredito rizikos vertinimo ir reitingavimo metodus, ir atlikus atitinkamas korekcijas pritaikyti juos Lietuvos įmonių kredito rizikai vertinti. Darbo objektas – kredito rizikos vertinimo ir kredito reitingų nustatymo modeliai. Mokslinė problema: nors kredito rizikos vertinimo ir kredito reitingų nustatymo modeliai ir metodai yra plačiai išanalizuoti ir taikomi praktikoje, tačiau visi jie yra labiau pritaikyti stambioms užsienio rinkoms, kurios reikšmingai skiriasi nuo Lietuvos rinkos, todėl egzistuoja modelio, pritaikyto konkrečiai Lietuvos rinkai, problema. Darbo tikslas – įmonių kredito rizikos vertinimo modelio, pritaikyto Lietuvos rinkai, suformulavimas. Darbą sudaro trys skyriai: teorinė, analitinė ir rezultatų. Teorinėje... [toliau žr. visą tekstą] / Issuing credits is one of the main bank’s activities. Each bank takes a credit risk by giving the credits. Credit risk is the most important risk of all and requires exceptional consideration, because potential losses caused by credit risk can be huge. If correct and accurate credit risk evaluation models are used to evaluate the credit risk, it helps to reduce the credit risk and increase the stability and reliability of the bank. That is why it is so important and topical to perform the analysis of the credit risk evaluation and credit ratings models and to make corrections for adoption these models for credit risk evaluation of Lithuanian companies. The object of this working paper – credit risk evaluation and assigning credit ratings models. The scientific problem: although credit risk evaluation and credit ratings methods and models are broadly analyzed and applied in practice, but these models are more designed for large foreign markets, which significantly differs from Lithuania’s market and because of that the problem of an adopted credit risk evaluation model for Lithuanian companies exists. The purpose of this working paper – to formulate the credit risk evaluation model adopted for Lithuanian companies. This working paper consists of three chapters: theoretical, analytical and results. In theoretical chapter risk, risk kinds, risk measurement models allowing to measure creditworthiness and assigning credit ratings models are analyzed. In analytical chapter the... [to full text]

It is the purpose of this study to examine the underlying nature of the relative risk associated with an audit engagement.

Cybersecurity threats have surged in the past decades. Experts agree that conventional security measures will soon not be enough to stop the propagation of more sophisticated and harmful cyberattacks. Recently, there has been a growing interest in mastering the complexity of cybersecurity by adopting methods borrowed from Artificial Intelligence (AI) in order to support automation. Moreover, entire security frameworks, such as DETECT (Decision Triggering Event Composer and Tracker), are designed aimed to the automatic and early detection of threats against systems, by using model analysis and recognising sequences of events and other tropes, inherent to attack patterns. In this project, I concentrate on cybersecurity threat assessment by the translation of Attack Trees (AT) into probabilistic detection models based on Bayesian Networks (BN). I also show how these models can be integrated and dynamically updated as a detection engine in the existing DETECT framework for automated threat detection, hence enabling both offline and online threat assessment. Integration in DETECT is important to allow real-time model execution and evaluation for quantitative threat assessment. Finally, I apply my methodology to some real-world case studies, evaluate models with sample data, perform data sensitivity analyses, then present and discuss the results.

A qualidade do ar nas áreas urbanas e industriais tende a apresentar concentrações indesejáveis de contaminantes, sem que haja um sistema abrangente de monitoramento, dada a sofisticação dos métodos físico-químicos convencionais, que requerem custos elevados de implantação, operação e manutenção, custos estes, que podem ser minimizados pela adoção de metodologia complementar de biomonitoramento. O biomonitoramento é um método experimental que permite avaliar a resposta de organismos vivos à poluição, oferecendo vantagens como: custos reduzidos, eficiência para o monitoramento de áreas amplas e por longos períodos de tempo e, também, avaliação de elementos químicos em baixas concentrações ambientais. As medidas e registros efetuados por redes convencionais de monitoramento da qualidade do ar permitem verificar se normas e limites estabelecidos ou recomendados pela legislação, agências ambientais e órgãos de promoção da saúde humana estão sendo respeitados. Entretanto, tais medições não permitem conclusões imediatas sobre as conseqüências de poluentes nos seres vivos. Assim, o biomonitoramento deve ser considerado como um método complementar na análise de poluentes, podendo constituir-se em um terceiro sistema de informações, além dos inventários de emissões e de concentrações ambientais. O presente trabalho teve por objetivo identificar, por meio de revisão sistemática de literatura desenvolvida por dois revisores independentes, espécies vegetais (vasculares, musgos e líquens) utilizadas como bioindicadores, referente ao período de janeiro de 1997 a junho de 2003, em estudos experimentais e observacionais, associando-as a poluentes atmosféricos. De um total de 4547 trabalhos científicos sobre bioindicadores, foram pré-selecionados 279 estudos referentes ao uso de vegetais bioindicadores de poluição atmosférica, publicados nos idiomas inglês, espanhol e português. Estes trabalhos foram analisados pela aplicação de dois testes de relevância, sendo selecionadas 240 referências e obtidos 154 estudos na íntegra. Deste total, foram incluídos, após aplicação dos dois testes de relevância, 126 trabalhos científicos, sobre o tema considerado, realizados em 34 diferentes países. Constatou-se que o uso da metodologia de revisão sistemática permitiu levantar o conhecimento das experiências acadêmicas nesta área de estudo, ampliando o conhecimento sobre esse tema. Os resultados ainda revelaram a utilização de 112 espécies vegetais, sendo 64 espécies pertencentes à divisão Angiospermae; 11 espécies da divisão Coniferophyta; 22 espécies de líquens e 15 espécies de musgos, relacionadas ao monitoramento de um ou mais dos seguintes poluentes atmosféricos: metais pesados, ozônio, material particulado, dióxido de enxofre, óxidos de nitrogênio, monóxido de carbono, fluoretos, compostos orgânicos voláteis e hidrocarbonetos. Constatou-se, assim, a existência de uma quantidade significativa de estudos dessa natureza, principalmente nos países europeus, onde está implantado o projeto EUROBIONET de biomonitoramento de poluição atmosférica, baseado na padronização de ensaios e biomonitores, desde o ano 2000. Tendo em vista que determinados bioindicadores já estão consagrados ou mesmo validados para o monitoramento de poluentes atmosféricos específicos, considera-se ser possível a instalação de uma rede de biomonitoramento ambiental no Estado de São Paulo, a partir de um trabalho conjunto e coordenado entre universidades, municípios e agência de proteção ambiental, associada à rede existente de monitoramento convencional da qualidade do ar. Tal iniciativa permitirá que mais um passo seja dado na universalização dos cuidados com os ambientes natural e social, promovendo e garantindo melhorias no padrão de qualidade de vida das sociedades atuais e futuras. / Air quality in urban and industrial areas tends to present undesirable concentrations of contaminants, without the availability of a broad monitoring system, given the sophistication of conventional physical-chemical methods, which require high expenditure for implantation, operation and maintenance. These costs can be reduced by the adoption of the complementary biomonitoring methodology. Biomonitoring is an experimental methodology that allows us to evaluate the response of living organisms to pollution, including advantages such as: reduced costs; efficiency to monitor large areas over long periods of time; evaluation of chemical elements in low environmental concentrations. Measurements and registers by conventional air quality monitoring networks allow us to verify whether the standards and limits set by legislation, environmental agencies and human health promotion bodies are being respected. However, these measurements do not allow for immediate conclusions about the consequences of polluting agents for human beings. Thus, biomonitoring must be considered a complementary method for analyzing pollutant agents and can constitute a third information system, besides emission and environmental concentration inventories. This study aimed to identify, by means of a systematic literature review carried out by two peer independent reviewers, vegetal species (vascular, moss and lichen) that are used as bioindicators, in the period from January 1997 to June 2003, in experimental and observational studies, linking them up with atmospheric pollutant agents. Out of a total of 4,547 scientific studies on bioindicators, 279 studies about the use of vegetal atmospheric pollution indicators were preselected, published in English, Spanish and Portuguese. These studies were analyzed by means of two relevancy tests, resulting in the selection of 240 references and 154 full studies. Out of this total, after applying the two relevancy tests, we included 126 scientific studies from 34 different countries. The systematic review methodology allowed us to survey the knowledge resulting from academic experiments in this field of study, thus broadening the knowledge related to that theme. Results disclosed the use of 111 vegetal species, 63 of which belonged to the Angiospermae group; 11 to the Coniferophyta group; 15 moss species and 22 lichen species, related to the monitoring of one or more of the following atmospheric polluting agents: heavy metals, ozone, particles, sulfur dioxide, nitrogen oxides, carbon monoxide, fluorides, volatile organic compounds. Thus, we observed that there is a significant amount of this kind of studies, mainly in European countries, where the EUROBIONET project for biomonitoring atmospheric pollution has been implanted since 2000, based on test and biomonitor standardization. In view of the fact that certain bioindicators are already acclaimed or even validated for monitoring specific atmospheric polluting agents, we believe it is possible to install an environmental biomonitoring network in São Paulo State, starting from a joint and coordinated effort among universities, municipalities and the environmental protection agency, in cooperation with the existing conventional air quality monitoring network. This initiative will allow for another step in the universalization of natural and social environment care, promoting and guaranteeing improvement in the quality of life of current and future societies.

Made available in DSpace on 2015-12-03T18:33:14Z (GMT). No. of bitstreams: 1 Guilherme_Arevalo_Leal.pdf: 867348 bytes, checksum: 54a59ca7b3f8a507c316893086a7d458 (MD5) Previous issue date: 2014-12-15 / The work aims to verifythe determinants oftrade creditof listed companies in Brazil'sStock Exchange overthe period from 2003 to2013, considereda credit expansion period for our economy. International studieshave testeddataincross sectionandshort-periodpanel, with significant resultsin explainingthe volume of non-financialcompanies trade credit (measured by accounts receivable and accounts payable). International andlocalliterature on this matteranalyzed theimpactsofvariablessuchthatsize, leverage, liquidity accounts andcyclicalprofitability.This study analyzed84non-financiallisted companieswith available data for the whole period. For the Accounts Receivable Model,thevolume of credit grantedisnegatively related tofirm size, the volume of highliquidityresourcesin other accounts(cash andstock), indicating that the greater amount invested in other high liquidityaccounts is inversely related tothe credit supply to customers.The variables ofaccounts payable,debt andoperating profitshowed positive andsignificant coefficient with the volume oftrade credittakenby companies.The companytends togivemorecreditto their customersas theytakemorecreditfrom their suppliersand obtainbank financing.In addition, ahigher generation ofoperating incomeis relatedto increasedlending.ForAccounts Payablemodel, thefirm sizevariablewas found to benegativeandsignificant coefficient.This resultis supported by thepositive and significantcoefficient of thevariableaccountsreceivablepresent in the model. Inventories,however,had a positiveand significantsignal on thepanelspecification, indicating that companies withhigher inventoriestend to gethigher volume ofcredit fromsuppliers. / O objetivo deste trabalho ?? analisar os fatores determinantes do trade credit das empresas listadas na bolsa de valores do Brasil ao longo do per??odo de 2003 a 2013, considerado macroeconomicamente como intervalo de expans??o do cr??dito brasileiro. Estudos internacionais j?? testaram dados em cross section e per??odos curtos em painel, apresentando resultados significantes na explica????o da concess??o e da tomada de cr??dito de empresas n??o financeiras. A literatura internacional e nacional sobre o tema j?? analisaram os impactos de vari??veis tais quais tamanho, endividamento, contas c??clicas e rentabilidade. Este estudo analisou 84 empresas n??o financeiras listadas que apresentaram seus dados anualmente entre 2003 e 2013, evidenciando a capacidade explicativa das vari??veis j?? utilizadas tanto para modelos cuja vari??vel dependente ?? tanto o volume de contas a receber quanto o volume de contas a pagar.Para o modelo de Contas a Receber, o volume de cr??dito concedido ?? negativamente relacionado com o tamanho da empresa, o volume de recursos de alta liquidez em outras contas (caixa e estoques), indicando que maior a presen??a de liquidez imediata est?? inversamente relacionada com a oferta de financiamento aos seus clientes, por meio de concess??o de prazo de pagamento. As vari??veis de contas a pagar, endividamento e rentabilidade operacional apresentaram rela????o positiva e significante com o volume de trade credit tomado pelas empresas. A empresa tende a conceder mais cr??dito aos seus clientes ?? medida em que toma mais cr??dito de seus fornecedores e obt??m financiamento banc??rio. Al??m disso, uma maior gera????o de resultado operacional est?? relacionada a uma maior concess??o de cr??dito. Para o modelo de Contas a Pagar, a vari??vel de tamanho da empresa mostrou-se com coeficiente negativo e significante, indicando que empresas de maior tamanho tendem a apresentar menores volumes tanto de cr??dito concedido aos clientes como tamb??m de cr??dito tomado de fornecedores. Este resultado ?? corroborado pelo coeficiente positivo e significante da vari??vel de contas a receber presente no modelo. A vari??vel de estoques, no entanto, apresentou sinal positivo e significante na especifica????o de painel, indicando que empresas com estoques tendem a obter maior volume de cr??dito concedido por fornecedores.

Das Ziel dieser Studie war eine Darstellung der aktuellen Situation der präoperativen Risikoevaluierung in der Anästhesie-Ambulanz der Universitätsmedizin Göttingen unter besonderer Beachtung des kardialen Risikos. Es erfolgte ein Vergleich der Vorstellungen der Anästhesisten mit dem realen Vorgehen bei der Prämedikationsvisite. In den meisten Fällen kommt es zu einer Übereinstimmung. Es gibt jedoch auch Fälle, in denen es Unterschiede mit möglicherweise weitreichenden Konsequenzen gibt. Beispielsweise zeigt sich eine hundertprozentige Einigkeit unter den Anästhesisten, Betablocker präoperativ nicht abzusetzen. In der Praxis wird dies bei jedem 20. Patienten dennoch getan. Zudem ergeben sich auffällige Diskrepanzen zwischen internen Vorgaben oder Leitlinien und dem Vorgehen der prämedizierenden Anästhesisten hinsichtlich der Nachforderung weiterführender präoperativer Untersuchungen. Insbesondere spiegelt sich dies bei den Laborparametern wider, welche die Anästhesisten bereits routinemäßig wesentlich häufiger bestimmen lassen wollen und dies in der Praxis dann auch bei immerhin 9 % aller Patienten ohne ersichtlichen Grund anhand der Leitlinien oder klinischen Befunde durchführen. Hinsichtlich der kardialen Risikoevaluation zeigt sich, dass in fast 90 % der Fälle eine solche anhand der Dokumentation nach den aktuellen Leitlinien nachvollziehbar ermöglicht wird. Zur Optimierung wäre sicher eine Erstellung von Pflichtfeldern auf dem Prämedikationsprotokoll sinnvoll. Trotz hoher Korrelation zwischen ASA-Einstufung und kardialer Risikobeurteilung gibt es dennoch immer wieder Patienten, welche trotz höherer ASA-Einstufung nur ein geringes kardiales Risiko aufweisen und auch wenige, bei denen trotz niedriger ASA-Stufe dennoch ein kardiales Risiko zu finden ist.

A alta ocorrência de atropelamentos nas áreas urbanas acarreta altos custos econômicos e sociais. Por isso, é importante a adoção de medidas de segurança viária que visem previnir e mitigar os riscos de acidentes envolvendo pedestres. Para que essas medidas sejam eficazes, sua escolha deve estar embasada em informações sobre onde, como e porque os atropelamentos acontecem. Uma forma de obtenção dessas informações envolve as avaliações de risco. Este trabalho propõe um método de avaliação de risco potencial de atropelamentos em travessias urbanas em meio de quadra. O método proposto incorpora análises quantitativas e qualitativas através de técnicas de modelagem que relacionam a incidência de atropelamentos e a percepção de risco dos pedestres com os fatores de risco. A aplicação do método envolveu a análise dos dados de atropelamentos registrados na cidade de Porto Alegre durante o período de 1998 e 2006 e a identificação das travessias em meio de quadra onde ocorreu o maior número de atropelamentos nesse período. Foram identificadas e, então, analisadas 21 travessias em meio de quadra. A análise das travessias incluiu a coleta de dados referentes aos fatores de risco e uma simulação de vídeo, onde pedestres e especialistas avaliaram as travessias no que concerne a sua segurança. A partir dos dados coletados, foram construídos dois modelos de regressão, um embasado em dados de ocorrência de atropelamentos e outro em dados de percepção de risco. A análise desses modelos mostrou que o risco de atropelamento é influenciado por uma combinação de fatores de risco, entre os quais as características do transporte coletivo, a largura da via, o número de faixas de tráfego de veículos, o volume de pedestres e de veículos. Ressalta-se que os modelos construídos podem ser utilizados para relativizar o risco observado e percebido de travessias em meio de quadra de Porto Alegre e fornecer indicativos para o gerenciamento da segurança dos pedestres. O estudo de caso mostrou que a aplicação do método proposto é simples, sendo possível a sua transferência a realidades onde há escassez de dados e recursos para sua obtenção. / The high occurrence of pedestrian crashes in urban areas results in high social and economic costs. To address this problem, it is important to adopt countermeasures that aim to prevent and reduce the risk of pedestrian crash. In order to obtain efficient countermeasures, their selection must be based on information of where, when and why the pedestrian crashes occur. This information can be obtained through risk evaluation. This dissertation proposes a method to evaluate the potential risk of pedestrian crash at midblock crosswalks. The proposed method incorporates quantitative and qualitative analyses, using modeling techniques to understand the relationship of risk factors with pedestrian crash occurrence and the relationship of risk factors with pedestrian risk perception. The method application comprised the analysis of reported pedestrian crashes in the city of Porto Alegre, in southern Brazil, over the period of 1998 to 2006 and the identification of midblock crosswalks with the highest incidence of pedestrian crash. Twenty one midblock crosswalks were indentified and, then analyzed. The crosswalks’ analysis included data collection about risk factors and video simulations. During the video simulations, pedestrians and experts evaluated the crosswalks’ safety. Two regression models were developed using the data collected, one based on pedestrian crash occurrence and one based on risk perception data. The analysis of models showed that pedestrian crash risk is influenced by a combination of risk factors, such as public transportation characteristics, road width, number of through lanes, pedestrian and vehicle volumes. The developed models can be used to relativize the observed and perceived risk of midblock crosswalks in Porto Alegre and to provide information to the pedestrian safety management. The case study results showed that the application of the proposed method is simple and it is possible to use it on real world cases where data and resources to obtain such data are scarce.

Made available in DSpace on 2010-04-20T21:00:25Z (GMT). No. of bitstreams: 4 sandrayoshida.pdf.jpg: 18559 bytes, checksum: 89330efa6ff64db03630f55c21d0cbec (MD5) sandrayoshida.pdf: 690177 bytes, checksum: 63543bceee748a4231a59b679cc35088 (MD5) 1_167660.pdf: 704396 bytes, checksum: 0801d7a4c995db7f173722b0ff530263 (MD5) sandrayoshida.pdf.txt: 102582 bytes, checksum: eeb87b7e6c9f90db6038bf6fbbd74155 (MD5) Previous issue date: 2008-04-02T00:00:00Z / When companies decide whether they should invest or not in a long run (5 to 10 years ahead) investment project, some alternative methodologies to the Discount Cash Flow (DCF) may become useful in order to confirm the business feasibility or to define the best moment to start the enterprise. These analysis take into consideration both the cash flow uncertainty and the project starting date flexibility. It can also be built using a stochastic approach, such as the differential equation solutions for geometric Brownian motion. Under certain conditions, project opportunities can be seen as call options with no maturity date, as the model proposed by McDonald-Siegel (1986) for the decision makers process and the optimum moment to start the project. This work analyses the feasibility of investments in the telecommunication market using non deterministic models, where the most relevant variable is the return dispersion, i.e., the variance represents the risk associated to the enterprise. / Quando as empresas decidem se devem ou não investir em determinado projeto de investimentos a longo prazo (horizonte de 5 a 10 anos), algumas metodologias alternativas ao Fluxo de Caixa Descontado (FCD) podem se tornar úteis tanto para confirmar a viabilidade do negócio como para indicar o melhor momento para iniciar o Empreendimento. As análises que levam em conta a incerteza dos fluxos de caixa futuros e flexibilidade na data de início do projeto podem ser construídos com a abordagem estocástica, usando metodologias como a solução de equações diferenciais que descrevem o movimento browniano. Sob determinadas condições, as oportunidades de investimentos em projetos podem ser tratados como se fossem opções reais de compra, sem data de vencimento, como no modelo proposto por McDonald-Siegel (1986), para a tomada de decisões e momento ótimo para o investimento. Este trabalho analisa a viabilidade de investimentos no mercado de telecomunicações usando modelos não determinísticos, onde a variável mais relevante é a dispersão dos retornos, ou seja, que a variância representa o risco associado a determinado empreendimento.

A alta ocorrência de atropelamentos nas áreas urbanas acarreta altos custos econômicos e sociais. Por isso, é importante a adoção de medidas de segurança viária que visem previnir e mitigar os riscos de acidentes envolvendo pedestres. Para que essas medidas sejam eficazes, sua escolha deve estar embasada em informações sobre onde, como e porque os atropelamentos acontecem. Uma forma de obtenção dessas informações envolve as avaliações de risco. Este trabalho propõe um método de avaliação de risco potencial de atropelamentos em travessias urbanas em meio de quadra. O método proposto incorpora análises quantitativas e qualitativas através de técnicas de modelagem que relacionam a incidência de atropelamentos e a percepção de risco dos pedestres com os fatores de risco. A aplicação do método envolveu a análise dos dados de atropelamentos registrados na cidade de Porto Alegre durante o período de 1998 e 2006 e a identificação das travessias em meio de quadra onde ocorreu o maior número de atropelamentos nesse período. Foram identificadas e, então, analisadas 21 travessias em meio de quadra. A análise das travessias incluiu a coleta de dados referentes aos fatores de risco e uma simulação de vídeo, onde pedestres e especialistas avaliaram as travessias no que concerne a sua segurança. A partir dos dados coletados, foram construídos dois modelos de regressão, um embasado em dados de ocorrência de atropelamentos e outro em dados de percepção de risco. A análise desses modelos mostrou que o risco de atropelamento é influenciado por uma combinação de fatores de risco, entre os quais as características do transporte coletivo, a largura da via, o número de faixas de tráfego de veículos, o volume de pedestres e de veículos. Ressalta-se que os modelos construídos podem ser utilizados para relativizar o risco observado e percebido de travessias em meio de quadra de Porto Alegre e fornecer indicativos para o gerenciamento da segurança dos pedestres. O estudo de caso mostrou que a aplicação do método proposto é simples, sendo possível a sua transferência a realidades onde há escassez de dados e recursos para sua obtenção. / The high occurrence of pedestrian crashes in urban areas results in high social and economic costs. To address this problem, it is important to adopt countermeasures that aim to prevent and reduce the risk of pedestrian crash. In order to obtain efficient countermeasures, their selection must be based on information of where, when and why the pedestrian crashes occur. This information can be obtained through risk evaluation. This dissertation proposes a method to evaluate the potential risk of pedestrian crash at midblock crosswalks. The proposed method incorporates quantitative and qualitative analyses, using modeling techniques to understand the relationship of risk factors with pedestrian crash occurrence and the relationship of risk factors with pedestrian risk perception. The method application comprised the analysis of reported pedestrian crashes in the city of Porto Alegre, in southern Brazil, over the period of 1998 to 2006 and the identification of midblock crosswalks with the highest incidence of pedestrian crash. Twenty one midblock crosswalks were indentified and, then analyzed. The crosswalks’ analysis included data collection about risk factors and video simulations. During the video simulations, pedestrians and experts evaluated the crosswalks’ safety. Two regression models were developed using the data collected, one based on pedestrian crash occurrence and one based on risk perception data. The analysis of models showed that pedestrian crash risk is influenced by a combination of risk factors, such as public transportation characteristics, road width, number of through lanes, pedestrian and vehicle volumes. The developed models can be used to relativize the observed and perceived risk of midblock crosswalks in Porto Alegre and to provide information to the pedestrian safety management. The case study results showed that the application of the proposed method is simple and it is possible to use it on real world cases where data and resources to obtain such data are scarce.