Following the introduction of the discounted penalty function by Gerber and Shiu (1998), significant progress has been made on the analysis of various ruin-related quantities in risk theory. As we know, the discounted penalty function not only provides a systematic platform to jointly analyze various quantities of interest, but also offers the convenience to extract key pieces of information from a risk management perspective. For example, by eliminating the penalty function, the Gerber-Shiu function becomes the Laplace-Stieltjes transform of the time to ruin, inversion of which results in a series expansion for the associated density of the time to ruin (see, e.g., Dickson and Willmot (2005)). In this thesis, we propose to analyze the long-standing finite-time ruin problem by incorporating the number of claims until ruin into the Gerber-Shiu analysis. As will be seen in Chapter 2, many nice analytic properties of the original Gerber-Shiu function are preserved by this generalized analytic tool. For instance, the Gerber-Shiu function still satisfies a defective renewal equation and can be generally expressed in terms of some roots of Lundberg's generalized equation in the Sparre Andersen risk model. In this thesis, we propose not only to unify previous methodologies on the study of the density of the time to ruin through the use of Lagrange's expansion theorem, but also to provide insight into the nature of the series expansion by identifying the probabilistic contribution of each term in the expansion through analysis involving the distribution of the number of claims until ruin. In Chapter 3, we study the joint generalized density of the time to ruin and the number of claims until ruin in the classical compound Poisson risk model. We also utilize an alternative approach to obtain the density of the time to ruin based on the Lagrange inversion technique introduced by Dickson and Willmot (2005). In Chapter 4, relying on the Lagrange expansion theorem for analytic inversion, the joint density of the time to ruin, the surplus immediately before ruin and the number of claims until ruin is examined in the Sparre Andersen risk model with exponential claim sizes and arbitrary interclaim times. To our knowledge, existing results on the finite-time ruin problem in the Sparre Andersen risk model typically involve an exponential assumption on either the interclaim times or the claim sizes (see, e.g., Borovkov and Dickson (2008)). Among the few exceptions, we mention Dickson and Li (2010, 2012) who analyzed the density of the time to ruin for Erlang-n interclaim times. In Chapter 5, we propose a significant breakthrough by utilizing the multivariate version of Lagrange's expansion theorem to obtain a series expansion for the density of the time to ruin under a more general distribution assumption, namely when interclaim times are distributed as a combination of n exponentials. It is worth emphasizing that this technique can also be applied to other areas of applied probability. For instance, the proposed methodology can be used to obtain the distribution of some first passage times for particular stochastic processes. As an illustration, the duration of a busy period in a queueing risk model will be examined. Interestingly, the proposed technique can also be used to analyze some first passage times for the compound Poisson processes with diffusion. In Chapter 6, we propose an extension to Kendall's identity (see, e.g., Kendall (1957)) by further examining the distribution of the number of jumps before the first passage time. We show that the main result is particularly relevant to enhance our understanding of some problems of interest, such as the finite-time ruin probability of a dual compound Poisson risk model with diffusion and pricing barrier options issued on an insurer's stock price. Another closely related quantity of interest is the so-called occupation times of the surplus process below zero (also referred to as the duration of negative surplus, see, e.g., Egidio dos Reis (1993)) or in a certain interval (see, e.g., Kolkovska et al. (2005)). Occupation times have been widely used as a contingent characteristic to develop advanced derivatives in financial mathematics. In risk theory, it can be used as an important risk management tool to examine the overall health of an insurer's business. The main subject matter of Chapter 7 is to extend the analysis of occupation times to a class of renewal risk processes. We provide explicit expressions for the duration of negative surplus and the double-barrier occupation time in terms of their Laplace-Stieltjes transform. In the process, we revisit occupation times in the content of the classical compound Poisson risk model and examine some results proposed by Kolkovska et al. (2005). Finally, some concluding remarks and discussion of future research are made in Chapter 8.

A neutropenia induzida pela quimioterapia é o mais comum efeito adverso da quimioterapia sistêmica para o câncer e é frequentemente complicada por neutropenia febril (NF). O uso profilático de fatores de crescimento hematopoiéticos pode reduzir o risco, a severidade e a duração da NF. Na prática clínica atual, a decisão de administrar ao paciente profilaxia com fatores de crescimento é baseada principalmente no potencial mielotóxico dos esquemas de QT, mas riscos específicos dos regimes não são definidos. Em muitos estudos, a toxicidade da quimioterapia é analizada em termos de alta dosagem versus baixa dosagem, sem uma regra geral que considere os diferentes esquemas de QT em uma única escala. O objetivo desse estudo é validar uma classificação de toxicidade de um esquema de QT e avaliar sua utilidade em um modelo de predição de risco de neutropenia febril em pacientes com câncer hematológico no começo de um ciclo de quimioterapia. Foram avaliados prospectivamente duzentos e sessenta e oito pacientes e acompanhados durante 1053 ciclos de quimioterapia na Bélgica, entre 2001 e 2005. Informações relevantes foram coletadas no começo do primeiro ciclo e o número de dias de neutropenia febril foi contabilizado no acompanhamento dos pacientes [dicotomizada (sem neutropenia versus >= dia de NF)]. A relação entre o desfecho e as co-variáveis foi analisada usando a Equação de Estimativa Generalizada (GEE). Um regime de quimioterapia agressiva é o maior preditor de NF [razão de chances (OR) 5.2 (3.2-8.4)]. Os outros preditores independentes são: doença subjacente, o comprometimento de medula óssea, superfície corporal <= 2m², uma contagem pré-tratamento de monócitos <150µl e a interação entre o primeiro ciclo na mesma linha de tratamento e uma dosagem de hemoglobina pré-tratamento. Usando as estimativas dos coeficientes de regressão, uma regra de predição clínica de NF foi desenvolvida com essas características: sensibilidade 78.6%, especificidade 62.3%, valor preditivo positivo de 42.7%% e um valor preditivo negativo de 89.1%. Estudos posteriores são necessários para validar esse escore bem como investigar novos potenciais fatores com o intuito de melhor prever a NF. / Chemotherapy-induced neutropenia is the most common adverse effect of chemotherapy and is often complicated by febrile neutropenia (FN). As prophylactic use of colonystimulating factors (CSF) can reduce the risk, severity, and duration of FN, it is of great importance to identify as soon as possible after or even before the start of chemotherapy, the patients who will develop FN. In the current clinical practice, the decision to give to the patient a colony-stimulating factor (CSF) prophylaxis is mainly based on the myelosuppressive potential of the chemotherapy regimen. The objective of this study is to validate a classification of aggressiveness of a chemotherapy regimen and to evaluate its usefulness in a risk prediction model of FN in patients with hematological cancer at the beginning of a chemotherapy cycle. Two hundred and sixty-six patients were prospectively enrolled and followed during 1053 cycles. Relevant patient informations were collected at the beginning of the first cycle and the number of days of FN were counted in the follow-up [dichotomized (no FN versus>= 1 day of FN)]. Aggressive chemotherapy regimen is the major predictor of FN [odds ratio 5.2 (3.2 - 8.4)]. The other independent predictors are the underlying disease, an involvement of bone marrow, body surface<= 2m², a baseline monocyte count <150/µl and the interaction between the first cycle in the same treatment line and a baseline hemoglobin dosage. A rule of prediction of FN was computed with these characteristics: sensitivity 78.6%, specificity 62.3%, positive predictive value 42.7% and negative predictive value 89.1%. Further studies are needed to validate this score.

A neutropenia induzida pela quimioterapia é o mais comum efeito adverso da quimioterapia sistêmica para o câncer e é frequentemente complicada por neutropenia febril (NF). O uso profilático de fatores de crescimento hematopoiéticos pode reduzir o risco, a severidade e a duração da NF. Na prática clínica atual, a decisão de administrar ao paciente profilaxia com fatores de crescimento é baseada principalmente no potencial mielotóxico dos esquemas de QT, mas riscos específicos dos regimes não são definidos. Em muitos estudos, a toxicidade da quimioterapia é analizada em termos de alta dosagem versus baixa dosagem, sem uma regra geral que considere os diferentes esquemas de QT em uma única escala. O objetivo desse estudo é validar uma classificação de toxicidade de um esquema de QT e avaliar sua utilidade em um modelo de predição de risco de neutropenia febril em pacientes com câncer hematológico no começo de um ciclo de quimioterapia. Foram avaliados prospectivamente duzentos e sessenta e oito pacientes e acompanhados durante 1053 ciclos de quimioterapia na Bélgica, entre 2001 e 2005. Informações relevantes foram coletadas no começo do primeiro ciclo e o número de dias de neutropenia febril foi contabilizado no acompanhamento dos pacientes [dicotomizada (sem neutropenia versus >= dia de NF)]. A relação entre o desfecho e as co-variáveis foi analisada usando a Equação de Estimativa Generalizada (GEE). Um regime de quimioterapia agressiva é o maior preditor de NF [razão de chances (OR) 5.2 (3.2-8.4)]. Os outros preditores independentes são: doença subjacente, o comprometimento de medula óssea, superfície corporal <= 2m², uma contagem pré-tratamento de monócitos <150µl e a interação entre o primeiro ciclo na mesma linha de tratamento e uma dosagem de hemoglobina pré-tratamento. Usando as estimativas dos coeficientes de regressão, uma regra de predição clínica de NF foi desenvolvida com essas características: sensibilidade 78.6%, especificidade 62.3%, valor preditivo positivo de 42.7%% e um valor preditivo negativo de 89.1%. Estudos posteriores são necessários para validar esse escore bem como investigar novos potenciais fatores com o intuito de melhor prever a NF. / Chemotherapy-induced neutropenia is the most common adverse effect of chemotherapy and is often complicated by febrile neutropenia (FN). As prophylactic use of colonystimulating factors (CSF) can reduce the risk, severity, and duration of FN, it is of great importance to identify as soon as possible after or even before the start of chemotherapy, the patients who will develop FN. In the current clinical practice, the decision to give to the patient a colony-stimulating factor (CSF) prophylaxis is mainly based on the myelosuppressive potential of the chemotherapy regimen. The objective of this study is to validate a classification of aggressiveness of a chemotherapy regimen and to evaluate its usefulness in a risk prediction model of FN in patients with hematological cancer at the beginning of a chemotherapy cycle. Two hundred and sixty-six patients were prospectively enrolled and followed during 1053 cycles. Relevant patient informations were collected at the beginning of the first cycle and the number of days of FN were counted in the follow-up [dichotomized (no FN versus>= 1 day of FN)]. Aggressive chemotherapy regimen is the major predictor of FN [odds ratio 5.2 (3.2 - 8.4)]. The other independent predictors are the underlying disease, an involvement of bone marrow, body surface<= 2m², a baseline monocyte count <150/µl and the interaction between the first cycle in the same treatment line and a baseline hemoglobin dosage. A rule of prediction of FN was computed with these characteristics: sensitivity 78.6%, specificity 62.3%, positive predictive value 42.7% and negative predictive value 89.1%. Further studies are needed to validate this score.

A neutropenia induzida pela quimioterapia é o mais comum efeito adverso da quimioterapia sistêmica para o câncer e é frequentemente complicada por neutropenia febril (NF). O uso profilático de fatores de crescimento hematopoiéticos pode reduzir o risco, a severidade e a duração da NF. Na prática clínica atual, a decisão de administrar ao paciente profilaxia com fatores de crescimento é baseada principalmente no potencial mielotóxico dos esquemas de QT, mas riscos específicos dos regimes não são definidos. Em muitos estudos, a toxicidade da quimioterapia é analizada em termos de alta dosagem versus baixa dosagem, sem uma regra geral que considere os diferentes esquemas de QT em uma única escala. O objetivo desse estudo é validar uma classificação de toxicidade de um esquema de QT e avaliar sua utilidade em um modelo de predição de risco de neutropenia febril em pacientes com câncer hematológico no começo de um ciclo de quimioterapia. Foram avaliados prospectivamente duzentos e sessenta e oito pacientes e acompanhados durante 1053 ciclos de quimioterapia na Bélgica, entre 2001 e 2005. Informações relevantes foram coletadas no começo do primeiro ciclo e o número de dias de neutropenia febril foi contabilizado no acompanhamento dos pacientes [dicotomizada (sem neutropenia versus >= dia de NF)]. A relação entre o desfecho e as co-variáveis foi analisada usando a Equação de Estimativa Generalizada (GEE). Um regime de quimioterapia agressiva é o maior preditor de NF [razão de chances (OR) 5.2 (3.2-8.4)]. Os outros preditores independentes são: doença subjacente, o comprometimento de medula óssea, superfície corporal <= 2m², uma contagem pré-tratamento de monócitos <150µl e a interação entre o primeiro ciclo na mesma linha de tratamento e uma dosagem de hemoglobina pré-tratamento. Usando as estimativas dos coeficientes de regressão, uma regra de predição clínica de NF foi desenvolvida com essas características: sensibilidade 78.6%, especificidade 62.3%, valor preditivo positivo de 42.7%% e um valor preditivo negativo de 89.1%. Estudos posteriores são necessários para validar esse escore bem como investigar novos potenciais fatores com o intuito de melhor prever a NF. / Chemotherapy-induced neutropenia is the most common adverse effect of chemotherapy and is often complicated by febrile neutropenia (FN). As prophylactic use of colonystimulating factors (CSF) can reduce the risk, severity, and duration of FN, it is of great importance to identify as soon as possible after or even before the start of chemotherapy, the patients who will develop FN. In the current clinical practice, the decision to give to the patient a colony-stimulating factor (CSF) prophylaxis is mainly based on the myelosuppressive potential of the chemotherapy regimen. The objective of this study is to validate a classification of aggressiveness of a chemotherapy regimen and to evaluate its usefulness in a risk prediction model of FN in patients with hematological cancer at the beginning of a chemotherapy cycle. Two hundred and sixty-six patients were prospectively enrolled and followed during 1053 cycles. Relevant patient informations were collected at the beginning of the first cycle and the number of days of FN were counted in the follow-up [dichotomized (no FN versus>= 1 day of FN)]. Aggressive chemotherapy regimen is the major predictor of FN [odds ratio 5.2 (3.2 - 8.4)]. The other independent predictors are the underlying disease, an involvement of bone marrow, body surface<= 2m², a baseline monocyte count <150/µl and the interaction between the first cycle in the same treatment line and a baseline hemoglobin dosage. A rule of prediction of FN was computed with these characteristics: sensitivity 78.6%, specificity 62.3%, positive predictive value 42.7% and negative predictive value 89.1%. Further studies are needed to validate this score.

This thesis calculates the value of statistical life (VSL) in the Czech Republic based on estimation of fatal risk computed across occupation within industries. Using the regression analysis, I estimate the impact of the fatal risk on wage. From this I calculate the VSL using the hedonic wage model. I use randomized wage data sample with 150 000 observations from the year 2013 combined with injury data from the years 2013-2015. Based on this data I estimated the VSL to be in a range between 6 965 277CZK (260 092EURO) and 37 355 674CZK (1. 394mil.EURO) based on statistically significant estimates of fatal risk. Furthermore, I estimate my model of fatal risk calculated across industries to compare the impact of the level of risk in the computation. This regression indicates that estimates based on risk in industries are overestimated. My results show that the method of risk computation is a key factor affecting the VSL.

Statistical inference is a tool for creating models that can accurately display real-world events. Special importance is given to the financial methods that model risk and large price movements. A parameter that describes tail heaviness, and risk overall, is α. This research finds a representative distribution that models α. The absolute value of standardized stock returns from the Center for Research on Security Prices are used in this research. The inference is performed using R. Approximations for α are found using the ptsuite package. The GAMLSS package employs maximum likelihood estimation to estimate distribution parameters using the CRSP data. The distributions are selected by using AIC and worm plots. The Skew t family is found to be representative for the parameter α based on subsets of the CRSP data. The Skew t type 2 distribution is robust for multiple subsets of values calculated from the CRSP stock return data.

Introduction: Lower extremity (LE) work-related musculoskeletal disorders (WMSDs) are known to occur with cumulative exposure to occupational and personal risks. The objective of this dissertation study was to find if creating a quantifiable risk detection model for the LE was feasible. The primary product of the literature review conducted for this study resulted in focusing the attention of the model development process onto creating the initial model of the LE for assessing knee disorder risk factors. Literature Review: LE occupational disorders affect numerous industries and thousands of people each year by affecting any one of the musculoskeletal systems deemed susceptible by the occupational and personal risk factors involved. Industries known to be affected tend to have labor intensive job descriptions. Some of the numerous industry examples include mining, manufacturing, firefighting, and carpet laying. Types of WMSDs noticed by the literature include bursitis, osteoarthritis, stress fractures, tissue inflammation, and nerve entrapment. In addition to the occupationally related disorders that may develop, occupationally related discomforts were also taken into consideration by this study. Generally, both the disorders and the discomforts can be traced to either a personal or occupational risk factor or both. Personal risk factors noted by the literature include a person's physical fitness and health history (such as past injuries). Meanwhile, occupational risks can be generalized to physical postures, activities, and even joint angles. Prevalence data over a three year interval (2003-2005) has found that LE WMSDs make up on average approximately 7.5% of all the WMSD cases reported to the US Occupational Safety and Health Administration (OSHA). When the literature is refined to the information pertaining to occupational knee disorders, the mean prevalence percentage of the same three year range is about 5%. Mean cost for knee injuries were found to be $18,495 (for the year between 2003 and 2004). Methodology: Developing a risk model for the knee meant using groups of subject matter experts for model development and task hazard analysis. Sample occupational risk data also needed to be gathered for each of a series of tasks so that the model could be validated. These sample data were collected from a sample aircraft assembly plant of a US aerospace manufacturer. Results: Based on the disorder and risk data found in the literature, a knee risk assessment model was developed to utilize observational, questionnaire, and direct measure data collection methods. The final version of this study's knee model has an inventory of 11 risk factors (8 occupational and 3 personal) each with varying degrees of risk exposure thresholds (e.g., high risk, moderate risk, or minimal risk). For the occupational risk assessment portion of the model, the results of task evaluations include both an occupational risk resultant score (risk score) and a task risk level (safe or hazardous). This set of results is also available for a cumulative (whole day) assessment. The personal risk assessment portion only produces a risk resultant score. Validation of the knee risk model reveals statistically (t (34) = 1.512, p = 0.156), that it is functioning as it should and can decide between hazardous and safe tasks. Additionally, the model is also capable of analyzing tasks as a series of cumulative daily events and providing an occupational and personal risk overview for individuals. Conclusion: While the model proved to be functional to the given sample site and hypothetical situations, further studies are needed outside of the aerospace manufacturing environment to continue testing both the model's validity and applicability to other industrial environments. The iterative adjustments generated for the occupational risk portion of the model (to reduce false positives and negatives) will need additional studies that will further evaluate professional human judgment of knee risk against this model's results. Future investigations must also make subject matter experts aware of the minimal risk levels of this knee risk assessment model so that task observational results are equally comparable. Additional studies are moreover needed to assess the intimate nature between variable interactions; especially multiple model defined minimal risks within a single task.

The role of homologous recombination deficiency (HRD) in lower grade glioma (LGG) has not been elucidated, and accurate prognostic prediction is also important for the treatment and management of LGG. The aim of this study was to construct an HRD-based risk model and to explore the immunological and molecular characteristics of this risk model. The HRD score threshold = 10 was determined from 506 LGG samples in The Cancer Genome Atlas cohort using the best cut-off value, and patients with highHRDscores had worse overall survival. A total of 251 HRD-related genes were identified by analyzing differentially expressed genes, 182 of which were associated with survival. A risk score model based on HRD-related genes was constructed using univariate Cox regression, least absolute shrinkage and selection operator regression, and stepwise regression, and patients were divided into high- and low-risk groups using the median risk score. High-risk patients had significantly worse overall survival than lowrisk patients. The risk model had excellent predictive performance for overall survival in LGG and was found to be an independent risk factor. The prognostic value of the riskmodel was validated using an independent cohort. In addition, the risk score was associated with tumor mutation burden and immune cell infiltration in LGG. High-risk patients had higher HRD scores and “hot” tumor immune microenvironment, which could benefit from poly-ADP-ribose polymerase inhibitors and immune checkpoint inhibitors. Overall, this big data study determined the threshold of HRD score in LGG, identified HRD-related genes, developed a risk model based on HRD-related genes, and determined the molecular and immunological characteristics of the risk model. This provides potential new targets for future targeted therapies and facilitates the development of individualized immunotherapy to improve prognosis.

Este trabalho apresenta as principais causas para a mortalidade das empresas paulistas criadas entre 2003 e 2007 a partir de base de dados cedida pelo SEBRAE-SP para o desenvolvimento dessa pesquisa. A amostra final, construída a partir de dados disponibilizados pela primeira vez para estudos desta natureza, contou com 662 empresas e 33 variáveis coletadas por meio de questionário aplicado diretamente às próprias empresas. A análise consistiu no teste de modelos econométricos, baseados na literatura dos modelos de duração, de forma a traduzir quais fatores são mais críticos para a sobrevivência das empresas a ponto de distingui-las em dois grupos: o das empresas vencedoras, cuja longevidade está pautada em ações que promovem ganhos de produtividade e eficiência, e aquelas desprovidas dessas ações e que muito provavelmente deixarão o mercado. Os três tipos de modelos abordados neste trabalho - não paramétrico, semi-paramétrico (riscos proporcionais) e paramétrico - apresentaram resultados similares, sendo que na abordagem de riscos proporcionais os resultados foram segmentados por tamanho e setor de atuação das empresas. Para as micro empresas, a idade do empreendedor e a iniciativa em investir na qualificação da mão de obra dos funcionários mostraram-se importantes mitigadores do risco de falha desse grupo de empresa, enquanto que para as pequenas empresas, a inovação em processos e a elaboração de um plano de negócios se destacaram dentre o conjunto de variáveis. Entre empresas dos setores de comércio e serviços, as empresas do primeiro grupo que faziam o acompanhamento das finanças (fluxo de caixa) apresentaram menor risco de falhar. Para aquelas do setor de serviços, a idade do empreendedor, o investimento em qualificação dos funcionários e o tamanho da empresa ao nascer foram importantes para reduzir o risco de falha no tempo. Outro resultado encontrado, por meio do modelo paramétrico utilizando distribuição Weibull, foi que o risco de a empresa deixar o mercado mostrou-se crescente, pelo menos nos cinco primeiros anos de existência da empresa. Entretanto, esse resultado não deve ser generalizado para períodos de tempo maiores que cinco anos. / This thesis presents the main results that determined the bankruptcy of enterprises located in the São Paulo State from 2003 to 2007. The models used in this work were possible due to the partnership with SEBRAE, Small Business Service Supporting, located in the State of São Paulo. This institution provided the data basis for this research and its final version was compound by 662 enterprises and 33 variables, which were collected from a survey done by SEBRAE and the related enterprise. For first time available for research like this The research was supported by econometrics models, more precisely duration models, which identified the most important factors regarding enterprises survival. Two enterprise groups were distinguished: that one that will survive and grow and another will fail. In this work, three models were used: parametric, non-parametric and proportional risk with all of them presenting similar results. The proportional risk approach was applied for economic sectors and enterprises size. For the micro size business, the entrepreneurship\'s age and the resources applied on the employee\'s qualification were important to reduce the risk to fail in the time, whereas for small enterprises, variables like innovation and business plan building were the most important variables. For the commerce and service sectors, the enterprises related to the first one, the enterprises which kept attention on financial results (cash flow) presented lower risk to fail. For service sector, variables such as: entrepreneur\'s age, investment on the employee\'s qualification and enterprise\'s size were the most important variables to explain the difference the risk to fail between the enterprises. Another result presented was the risk to fail, which indicates the likelihood of an enterprise to leave its business activity. In this case, the parametric model using Weibull distribution concluded that the risk grows in the first five years. However, this result must be carefully evaluated since it would be necessary a longer term data to ensure this result.

É elaborada uma proposta de procedimentos de projetos de comboios fluviais adaptados às condições existentes das vias navegáveis, a partir de uma visão ampliada da necessidade de obtenção de menores custos de transporte, com níveis adequados de segurança e de respeito ao meio ambiente. Uma análise das inter-relações técnicas entre hidrovias e embarcações, assim como dos condicionantes e implicações do conceito de embarcações adaptadas às vias navegáveis, orientou a elaboração do modelo. Este foi estruturado em módulos, com o objetivo de reproduzir, um a um, os fatores mais importantes que influenciam a eficiência, a segurança e o nível de interferência ambiental do transporte de cargas por comboios. Um programa computacional foi desenvolvido como instrumento de aplicação do modelo, consolidando os procedimentos propostos para a escolha das melhores alternativas de projeto e de operação de comboios. Resultados experimentais com comboios em escala real e com modelos em tanques de provas, foram utilizados para validação dos procedimentos adotados. Dados de acidentes com comboios em várias hidrovias do mundo foram utilizados como bases para avaliações de risco. O modelo foi aplicado aos casos de transportes de soja pela hidrovia Tietê-Paraná e pelo rio Araguaia, exemplificando as formas de análise e de escolha das alternativas de soluções de projeto. Ao final, os resultados obtidos comprovaram a utilidade da adoção de um enfoque mais abrangente do processo de projeto de comboios fluviais. / It is proposed a procedure model for design of barges push-tow adapted to waterway actual conditions, with the purpose of minimize transportation costs but always making verifications of navigation safety and ambient interferences levels. An analysis of inter-relations on inland waterways and cargo ships and, also, detailed conditionings and implications of waterway adapted ships concept, was used for model elaboration. It was structured in blocks to reproduce, one to one, the most important factors that modify efficiency, safety and environmental interference levels of barges push-tow cargo transportation. A computational program was developed to consolidate the proposed model and to apply procedures to choose best design and operational alternatives. Results of full scale and towing tank tests with push-tows were used to verify the mathematical and semi-empirical models. Barges push-tows accidents data from waterways of the world was used as risk model basis. To evaluate its effectiveness, the model was applied to bulk grain transportation cases by Tietê-Paraná waterway and by Araguaia river. The results shows that the special emphasis on three factors (efficiency, safety and ambient) improves the quality of barges push-tow design process.