Fundamentos para incorporar la cláusula de progreso de seguridad vial en las concesiones viales encargados de la infraestructura nacional”, es el contenido que se ha desarrollado en este presente tema, habiéndose realizado un tipo de investigación cualitativa, junto a un método analítico, con un análisis documental, tomándose como objeto de estudio el proponer la implementación de la cláusula de progreso de Seguridad Vial en las carreteras concesionadas de la Red Vial Nacional, explicando la relevancia de implementar esta condición en los contratos de concesión, debido a que establece la obligación continuada por parte del concesionario de incorporar normas que permitan la ejecución, desarrollo y cumplimiento de una obra pública. En ese mismo orden de ideas, se realizó un estudio de distintos autores con aportes importantes, que hablen sobre los contratos de concesión y cláusula de progreso, complementado con al análisis de las Asociaciones Público Privadas (en adelante APP) y seguridad vial. De tal manera, se pudo concluir que es muy común ver accidentes de tránsito y que éstos, en lugar de disminuir, siguen aumentando, debido a hechos imprevisibles, como casos fortuitos o fuerza mayor, que no se pudieron prever o predecir, sin embargo, no son las únicas causas, pues, también está el factor humano, como la imprudencia de los peatones o los conductores, y el factor mecánico, que serían aquellas fallas que puedan tener los vehículos automotores, por tanto, implementar una cláusula de progreso para la seguridad vial, resultaría una medida que busca reducir accidentes de tránsito mediante el equilibrio económico en los contratos de concesión ante normativas que puedan surgir después de celebrado dicho acto. / Fundamentals for incorporating the road safety progress clause in the concession contracts of the national road network, is the content that has been developed in this topic, having carried out a type of qualitative research, together with an analytical method, with an analysis documentary, taking as an object of study the proposal to implement the Road Safety progress clause on the concessioned highways of the National Road Network, explaining the relevance of implementing this condition in the concession contracts, since it establishes the continued obligation for part of the concessionaire to incorporate regulations that allow the execution, development and fulfillment of a public work. In the same order of ideas, a study was carried out by different authors with important contributions, who talk about concession contracts and progress clauses, complemented by the analysis of Public Private Associations (hereinafter APP) and road safety. In this way, it was possible to conclude that it is very common to see traffic accidents and that these, instead of decreasing, continue to increase, due to unforeseeable events, such as acts of God or force majeure, which could not be foreseen or predicted, however, These are not the only causes, since there is also the human factor, such as the recklessness of pedestrians or drivers, and the mechanical factor, which would be those failures that motor vehicles may have, therefore, implement a progress clause for the road safety, would be a measure that seeks to reduce traffic accidents through economic balance in concession contracts before regulations that may arise after said act is celebrated.

This research assessed the criminological significance of motor vehicle crash data within the criminal justice context. The study was conducted at the southern region Accident Bureau of the Ekurhuleni Metropolitan Police Department (EMPD). The objectives of this research are to: a. Explore and describe the extent and criminological impact of road traffic crashes on a macro level. b. Explore and describe the impact of road traffic crashes on a micro level. c. Explain the criminological significance of road traffic crash data in relation to effective road safety management. d. Evaluate and describe the attendance and recording procedure of road crashes. e. Evaluate and describe the supervisor’s responsibility in relation to the quality of crash data recorded. f. Evaluate and describe the effectiveness of crash statistics accessed by management and/or officials. g. Evaluate and explain the current prosecution procedure as part of the CJS. h. Evaluate existing policy about the crash-recording process applicable to operational law enforcement practitioners of the EMPD. i. Explore and describe the current state of motor vehicle insurance and its necessity from a criminological perspective. The researcher followed a mixed-methods approach, collecting quantitative data from recorded Accident Report Forms and qualitative data from interviews with participants active in their fields of expertise. Road crashes are characterised by injury and damage; victims suffer trauma and economic consequences, and the society and the economy as a whole are affected. The research suggests that accurate road crash data form a crucial component in the investigation, prosecution and development of road safety strategies in the quest towards improved road safety efforts. It also indicates that existing crash data are of such a sub-standard that traffic professionals will not be able to develop effective and efficient road safety strategies in pursuit of road safety / Criminology and Security Science / M.A. (Criminology)

In der vorliegenden Arbeit werden Erfahrungen von Dreiradfahrer_innen im deutschsprachigen Raum (vorwiegend Deutschland) untersucht, die mit Hilfe eines Online-Fragebogens erhoben wurden. Die betrachteten Themenfelder sind: Nutzungsgründe und -häufigkeit, Infrastruktur für den fließenden und ruhenden Radverkehr, Benutzungspflicht von Radwegen, Interaktion mit anderen Verkehrsteilnehmer_innen und mit Passant_innen sowie Verkehrssicherheit. Abschließend werden Empfehlungen gegeben, wie Infrastruktur für Dreiradfahrer attraktiver gestaltet werden kann, um die Verbreitung der Fahrzeuge zu fördern.

INTRODUCTION : En milieu urbain, l’amélioration de la sécurité des piétons pose un défi de santé publique. Pour chaque décès attribuable aux collisions routières, il y a des centaines de personnes blessées et, dans les pays riches, la diminution du nombre annuel de piétons décédés s’expliquerait en partie par la diminution de la marche. Les stratégies préventives prédominantes n’interviennent pas sur le volume de circulation automobile, un facteur pourtant fondamental. De plus, les interventions environnementales pour améliorer la sécurité des infrastructures routières se limitent habituellement aux sites comptant le plus grand nombre de décès ou de blessés. Cette thèse vise à décrire la contribution des volumes de circulation automobile, des pratiques locales de marche et de la géométrie des routes au nombre et à la répartition des piétons blessés en milieu urbain, et d’ainsi établir le potentiel d’une approche populationnelle orientée vers la reconfiguration des environnements urbains pour améliorer la sécurité des piétons. MÉTHODE : Le devis est de type descriptif et transversal. Les principales sources de données sont les registres des services ambulanciers d’Urgences-santé (blessés de la route), l’enquête Origine-Destination (volumes de circulation automobile), la Géobase du réseau routier montréalais (géométrie des routes) et le recensement canadien (pratiques locales de marche, position socioéconomique). Les analyses descriptives comprennent la localisation cartographique (coordonnées x,y) de l’ensemble des sites de collision. Des modèles de régression multi-niveaux nichent les intersections dans les secteurs de recensement et dans les arrondissements. RÉSULTATS : Les analyses descriptives démontrent une grande dispersion des sites de collision au sein des quartiers. Les analyses multivariées démontrent les effets significatifs, indépendants du volume de circulation automobile, de la présence d’artère(s) et d’une quatrième branche aux intersections, ainsi que du volume de marche dans le secteur, sur le nombre de piétons blessés aux intersections. L’analyse multi-niveaux démontre une grande variation spatiale de l’effet du volume de circulation automobile. Les facteurs environnementaux expliquent une part substantielle de la variation spatiale du nombre de blessés et du gradient socioéconomique observé. DISCUSSION : La grande dispersion des sites de collision confirme la pertinence d’une approche ne se limitant pas aux sites comptant le plus grand nombre de blessés. Les résultats suggèrent que des stratégies préventives basées sur des approches environnementales et populationnelle pourraient considérablement réduire le nombre de piétons blessés ainsi que les inégalités observées entre les quartiers. / INTRODUCTION: In an urban environment, improving pedestrian safety is a public health challenge. For every death attributable to road injuries, there are hundreds of injured people and, in richer countries, decreases in the number of pedestrian deaths may reflect a reduction of walking as a mode of transportation. The predominant preventive strategies have limited or no effect on traffic volume, which is the most important predictor of the number of pedestrians injured by motor vehicles. Moreover, environmental interventions are usually limited to sites that have had the greatest number of deaths or injuries. The purpose of this thesis is to describe the contribution of environmental factors such as traffic volume, local walking practices and street geometry, on the number and the distribution of pedestrian injuries in an urban setting. METHOD: The design of the study is descriptive and cross-sectional. The main data sources were Urgences-santé ambulance service records (road injuries), the Origin-Destination survey (traffic volume), the Geobase for the Montreal road network (street geometry) and the census of Canada (local walking practices, socioeconomic status). The descriptive analyses included map locations (x, y coordinates) for all collision sites. Multi-level regression models nested intersections in census sectors and in boroughs. RESULTS: The descriptive analyses showed a great deal of dispersion of collision sites within neighbourhoods. The multivariate analyses showed the significant effects, independent of traffic volume, of the presence of an artery and of a fourth leg at the intersection, and of pedestrian volume in the neighbourhood, on the number of pedestrian injuries at intersections. The multi-level analysis showed extensive spatial variation in the effect of traffic volume. This spatial variation in the number of casualties, and the socioeconomic gradient observed, were in large part explained by traffic volume, the presence of an artery and of a fourth leg at the intersection. DISCUSSION: The great dispersion of collision sites supports an approach which is not limited to sites with the highest number of injured pedestrians. This thesis suggests that preventive strategies based on environmental and population approaches could considerably reduce the number of injured pedestrians , as well as road injury inequalities across urban neighbourhoods.

Godstransporter ökar i takt med att utrikeshandeln växer, vilket bidrar till utsläpp av växthusgaser. Hållbar utveckling och miljö är viktiga faktorer i dagens samhälle vilket gör att det ställs krav på nya och innovativa transportlösningar. Ett alternativ att bemöta dessa krav är genom högkapacitetstranporter, som i dagsläget testas, utförs och forskas kring frekvent i Sverige. Högkapacitetstransport avser lastbilsfordon som är längre och/eller tyngre än dagens konventionella lastbil som har fått dispens att trafikera vissa zoner på det svenska vägnätet. Denna studie baseras på att utvärdera två intermodala transportlösningar för transport av gods på sträckan mellan Göteborgs hamn och Viared, Borås. Den ena transportlösningen baseras på högkapacitetstransport och den andra baseras på en järnvägsbaserad transportlösning. De alternativa transportlösningarna utvärderas med avseende på effektivitet omfattande transportkostnad, flexibilitet och tid samt miljöpåverkan omfattande koldioxidutsläpp genom en fallstudie.Slutsatserna har resulterat i att högkapacitetstransport genererar en högre transporteffektivitet jämfört med dagens dragbil som trafikerar den utpekade vägsträckan i fallstudien. I brist på material har inga konkreta kostnader kunnat beräknats för de olika transportlösningarna. Däremot kunde en slutsats dras kring att de alternativa transportlösningarna ej bör uppgå till en högre kostnad än vad dagens transportlösning genererar. Ur miljösynpunkt har koldioxidutsläpp för de olika transportlösningarna beräknats. Implementeras hela vägsträckan med högkapacitetstransport har resultatet gällande koldioxidutsläpp genererat i en minskning med 30,6 procent jämfört med dagens dragbil. Däremot visade studien att järnvägstransporten vinner miljömässiga fördelar då den jämfört med dragbilen som trafikerar vägsträckan idag reducerar utsläppen med 80,4 procent. Dock visar koldioxidutsläppet i den intermodala järnvägslösningen att vägtransporterna bidrar med 98 procent av utsläppen.Denna studie väger för- och nackdelar gentemot varandra för en implementering av högkapacitetstransport. Högkapacitetstransporten jämförs även gentemot järnvägens för- och nackdelar vid transport av containergods. Studien bidrar dessutom till att stärka och reflektera över tidigare publicerade studier som finns inom ämnet. / Freight transport increases as foreign trade grows, which contributes to greenhouse gas emission. Sustainable development and the environment are important factors in today's society, which imposes demands on new and innovative transport solutions. One way of meeting these demands is through high capacity transports, which are currently being tested, performed and researched frequently in Sweden. High capacity transport refers to truck vehicles that are longer and/or heavier than today's conventional truck that has been given the dispensation to traffic certain areas on the Swedish road network. This study is based on evaluating two intermodal transport solutions for the carriage of goods on the route between Port of Gothenburg and Viared, Borås. One transport solution is based on high capacity transport and the other is based on a rail-based transport solution. Through a case study, the alternative transport solutions are evaluated based on efficiency, including transportation costs, flexibility and time, and the environmental impact including CO2 emissions.The conclusions have resulted in high capacity transport generating a higher transport efficiency compared with today's truck that operates the designated road distance in the case study. In lack of material, no concrete costs have been calculated for the various transport solutions. On the other hand, a conclusion could be drawn that alternative transport solutions should not be at a higher cost than today's transport solution results in. From an environmental point of view, CO2 emissions for the various transport solutions have been calculated. Implementing the entire road haulage with high capacity transport has resulted in a reduction of CO2 emissions by 30,6 percent compared to the current truck. On the other hand, the study showed that rail transport wins environmental benefits, as compared to the truck that carries the route today, emissions reduce by 80,4 percent. However, CO2 emissions in the intermodal railroad solution show that road transport contributes 98 percent of emissions.This study weighs strengths and weaknesses towards each other for the implementation of high capacity transport on the designated route. High capacity transport is also set against the railroad with pros and cons of transporting container goods. The study also contributes to strengthening and reflecting the theory from previously published studies within the subject.

The unavailability of crash-related data has been a long lasting challenge in Brazil. In addition to the poor implementation and follow-up of road safety strategies, this drawback has hampered the development of studies that could contribute to national goals toward road safety. In contrast, developed countries have built their effective strategies on solid data basis, therefore, investing a considerable time and money in obtaining and creating pertinent information. In this research, we aim to assess the potential impacts of supplementary data on spatial model performance and the suitability of different spatial modeling approaches on crash prediction. The intention is to notify the authorities in Brazil and other developing countries, about the importance of having appropriate data. In this thesis, we set two specific objectives: (I) to investigate the spatial model prediction accuracy at unsampled subzones; (II) to evaluate the performance of spatial data analysis approaches on crash prediction. Firstly, we carry out a benchmarking based on Geographically Weighted Regression (GWR) models developed for Flanders, Belgium, and São Paulo, Brazil. Models are developed for two modes of transport: active (i.e. pedestrians and cyclists) and motorized transport (i.e. motorized vehicles occupants). Subsequently, we apply the repeated holdout method on the Flemish models, introducing two GWR validation approaches, named GWR holdout1 and GWR holdout2. While the former is based on the local coefficient estimates derived from the neighboring subzones and measures of the explanatory variables for the validation subzones, the latter uses the casualty estimates of the neighboring subzones directly to estimate outcomes for the missing subzones. Lastly, we compare the performance of GWR models with Mean Imputation (MEI), K-Nearest Neighbor (KNN) and Kriging with External Drift (KED). Findings showed that by adding the supplementary data, reductions of 20% and 25% for motorized transport, and 25% and 35% for active transport resulted in corrected Akaike Information Criterion (AICc) and Mean Squared Prediction Errors (MSPE), respectively. From a practical perspective, the results could help us identify hotspots and prioritize data collection strategies besides identify, implement and enforce appropriate countermeasures. Concerning the spatial approaches, GWR holdout2 out performed all other techniques and proved that GWR is an appropriate spatial technique for both prediction and impact analyses. Especially in countries where data availability has been an issue, this validation framework allows casualties or crash frequencies to be estimated while effectively capturing the spatial variation of the data. / A indisponibilidade de variáveis explicativas de acidentes de trânsito tem sido um desafio duradouro no Brasil. Além da má implementação e acompanhamento de estratégias de segurança viária, esse inconveniente tem dificultado o desenvolvimento de estudos que poderiam contribuir com as metas nacionais de segurança no trânsito. Em contraste, países desenvolvidos tem construído suas estratégias efetivas com base em dados sólidos, e portanto, investindo tempo e dinheiro consideráveis na obtenção e criação de informações pertinentes. O objetivo dessa pesquisa é avaliar os possíveis impactos de dados suplementares sobre o desempenho de modelos espaciais, e a adequação de diferentes abordagens de modelagem espacial na previsão de acidentes. A intenção é notificar as autoridades brasileiras e de outros países em desenvolvimento sobre a importância de dados adequados. Nesta tese, foram definidos dois objetivos específicos: (I) investigar a acurácia do modelo espacial em subzonas sem amostragem; (II) avaliar o desempenho de técnicas de análise espacial de dados na previsão de acidentes. Primeiramente, foi realizado um estudo comparativo, baseado em modelos desenvolvidos para Flandres (Bélgica) e São Paulo (Brasil), através do método de Regressão Geograficamente Ponderada (RGP). Os modelos foram desenvolvidos para dois modos de transporte: ativos (pedestres e ciclistas) e motorizados (ocupantes de veículos motorizados). Subsequentemente, foi aplicado o método de holdout repetido nos modelos Flamengos, introduzindo duas abordagens de validação para GWR, denominados RGP holdout1 e RGP holdout2. Enquanto o primeiro é baseado nas estimativas de coeficientes locais derivados das subzonas vizinhas e medidas das variáveis explicativas para as subzonas de validação, o último usa as estimativas de acidentes das subzonas vizinhas, diretamente, para estimar os resultados para as subzonas ausentes. Por fim, foi comparado o desempenho de modelos RGP e outras abordagens, tais como Imputação pela Média de dados faltantes (IM), K-vizinhos mais próximos (KNN) e Krigagem com Deriva Externa (KDE). Os resultados mostraram que, adicionando os dados suplementares, reduções de 20% e 25% para o transporte motorizado, e 25% e 35% para o transporte ativo, foram resultantes em termos de Critério de Informação de Akaike corrigido (AICc) e Erro Quadrático Médio da Predição (EQMP), respectivamente. Do ponto de vista prático, os resultados poderiam ajudar a identificar hotspots e priorizar estratégias de coleta de dados, além de identificar, implementar e aplicar contramedidas adequadas. No que diz respeito às abordagens espaciais, RGP holdout2 teve melhor desempenho em relação a todas as outras técnicas e, provou que a RGP é uma técnica espacial apropriada para ambas as análises de previsão e impactos. Especialmente em países onde a disponibilidade de dados tem sido um problema, essa estrutura de validação permite que as acidentes sejam estimados enquanto, capturando efetivamente a variação espacial dos dados.

L’accidentalité des conducteurs de 2RM est au cœur des préoccupations de sécurité routière. Ce travail de thèse vise à apporter des connaissances sur les différences de sexe et de conformité aux stéréotypes de sexe au sein de cette communauté très masculine. Il a été montré que les femmes motocyclistes avaient des taux d’accidents corporels ou mortels bien inférieurs à ceux des hommes. Les résultats montrent que les comportements à risque accidentels des hommes et des femmes motocyclistes sont proches, hormis pour les comportements les plus risqués. Il a aussi été montré que le type de motocyclette avait un effet plus important que le sexe sur les comportements à risques accidentels. De plus, il a été montré que les motocyclistes se conforment aux stéréotypes de sexe qui leurs sont associés et que la masculinité renforce la prise de risque et la transgression des règles, cette relation étant expliquée par les motivations à conduire un 2RM. Enfin, ce travail a montré que les stéréotypes de sexe associés à la conduite d’une motocyclette existent déjà chez les adolescents dès 11 ans. Ce travail de thèse montre donc des différences significatives entre hommes et femmes conducteurs de 2RM, autant au niveau de leur accidentalité que de leurs prises de risque. De plus, des stéréotypes de sexe existent bien pour cette population spécifique d’usagers de la route. Ces travaux permettent d’avoir une connaissance plus fine des comportements des conducteurs de 2RM, et d’enrichir la réflexion sur des actions possibles en matière d’éducation routière, en ciblant les sous-populations les plus à risque chez les conducteurs de 2RM. / The powered two-wheelers (PTW) riders’ accidentality is at the heart of road safety issues. This PhD thesis is part of a comprehensive approach to generate knowledge on sex differences and sex stereotypes conformity within this very masculine stereotyped community. It first showed that female motorcyclists have much lower injury crashes and fatalities rates than males. A first study showed that males declared more intentional risky behaviors and female more non-intentional risky behaviors. A second study showed that the PTW type chosen had a more important effect on the aberrant behaviors than sex. A third study showed that individuals who conformed to masculine stereotypes declared more violations than those who conformed to feminine stereotypes (declared more lapses), whatever their sex. Motivations to ride a PTW explaining this relation. Finally, this work showed that sex stereotypes associated with motorcycle riding already existed on the adolescent population, from the age of 11, even if they are themselves moped riders or if at least one of their parents is a rider. As a conclusion, significant differences between male and female PTW riders were revealed, in terms of accidentology and risk taking. These differences are linked to the riders’ conformity to their sex group and to the effects of this conformity on their motivations to ride a PTW. Moreover, it also showed that sex stereotypes exist for this specific road users’ population. This work led to a more detailed understanding of PTW riders’ aberrant behaviors, and to enrich the thinking for actions with regard to road safety education and prevention.

Neste trabalho foram estabelecidos os fundamentos metodológicos baseados na identificação dos fatores de risco, fatores de risco identificados, necessários na elaboração de campanhas de segurança viária, orientados pelas diretrizes estabelecidas no projeto CAST (Campaigns and Awareness-raising Strategies in Traffic Safety), o qual define a identificação do comportamento-problema gerador de acidentes como elemento essencial na definição do objetivo de cada campanha. Não havendo dados estatísticos consistentes e confiáveis dos quais seja possível extrair esta informação, foi necessário identificar os fatores comportamentais que estão na origem de atos inseguros e que devem ser prioritários na elaboração de campanhas de segurança viária. Para o efeito, foi necessário estabelecer uma metodologia que envolveu a utilização de diferentes instrumentos de coleta de dados com o objetivo de identificar os comportamentos-problema, assim como outros fatores relacionados à infraestrutura e ao meio ambiente que poderiam induzir esses comportamentos. Esta metodologia envolveu a aplicação de Grupos Focais que orientaram a elaboração de um questionário voltado para a identificação dos fatores de risco do sistema viário, comportamentos de risco e opiniões relacionados com os tipos de mensagens de campanhas. Este questionário foi posteriormente aplicado a uma amostra de condutores, juntamente com o DBQ – Driver Behaviour Questionnaire, este orientado para a identificação de comportamentos de risco de cada inquirido. A associação desses dois instrumentos permitiu a identificação de fatores de risco no sistema viário, assim como comportamentais de risco habituais e as preferências dos inquiridos sobre tipos e mensagens de campanhas, bem como meios de mídia. A análise estatística dos dados coletados permitiu identificar os seguintes fatores de risco relacionados à via: (1) más condições da infraestrutura da via, (2) más condições de sinalização, (3) más condições dos veículos que trafegam pelas vias, e (4) interação principalmente com pedestres. Foram ainda identificadas as suas opiniões relativamente a comportamentos-problema dos demais usuários do sistema viário, destacando-se os seguintes fatores: (1) sono, (2) bebidas, (3) drogas e (4) mensagens de texto. Quanto à importância das campanhas, os inquiridos consideram que elas devem: (1) ser dirigidas aos fatores de risco identificados, (2) estar associadas a outras ações como fiscalização e (3) seus resultados devem ser divulgados pelos órgãos de governo. Finalmente, acerca das mensagens das campanhas e tipos de mídia utilizados, os inquiridos consideraram que as mensagens devem: (1) utilizar laços afetivos, (2) ser mais impactantes, (3) conter cenas reais, (4) ter finalidade educativa; e as mídias mais adequadas seriam: (1) televisão, (2) internet (com ênfase às redes sociais) e (3) painéis de mensagens variáveis. Quanto ao questionário DBQ, foram identificados sete fatores comportamentais dos próprios inquiridos com maior peso e relação com atos inseguros: (1) Baixo nível de aptidão, (2) Infração, (3) Distração/Inatenção, (4) Condução agressiva, (5) Baixa consciência da situação, (6) Estresse e (7) Adaptação Comportamental. Esses fatores representam os tópicos prioritários que devem ser trabalhados em futuras campanhas. / This research established the methodological principles based on the identification of risk factors necessary for elaboration road safety campaigns, following the guidelines recommended by the CAST (Campaigns and Awareness-raising Strategies in Traffic Safety) project, which defines the identification of main problem behaviors generating accidents as an essential element in the definition of each campaign goal. Since there is no consistent and reliable statistical data proveding this information, it was necessary to identify the behavioral factors that originate unsafe acts and that should be priority in road safety campaigns. In order to achieve this result, it was necessary to establish a methodology involving the use of different data collection tools with the intention of identifying the problem behaviors, as well as other factors related to the infrastructure and environment capable to induce these behaviors. This methodology involved Focus Groups application that directed the elaboration of a questionnaire aiming at a identifing road system risk factors, risky behaviors and opinions regarding the types of campaigns messages. This questionnaire was further applied to a drivers\' sample, together with the DBQ Driver Behavior Questionnaire, which was directed to the identification of risky behaviors of each respondent. The association between these two instruments enabled the identification of road system risk factors, as well as common risky behaviors and the respondents\' preferences about the types of campaign messages and media means. The statistical analysis of the collected data enabled the identification of the following risk factors related to the road system: (1) poor road infrastructure conditions, (2) poor signing conditions, (3) poor vehicle conditions, and (4) interaction mainly with pedestrians. Opinions about problem-behaviors were also identified, stressing the following factors: (1) drowsiness, (2) drink and drive, (3) drugs and medication effects (4) texting driving. Concerning the campaigns importance, the respondents considered that they should: (1) be directed to identified risk factors, (2) be associated to other actions such as enforcement and (3) have its results published by the government entities. Finally, regarding the campaign messages and the used media types, the respondents considered that the messages should: (1) use affective anchors, (2) be more impactful, (3) contain real scenes, (4) have educative purpose; and the most adequate media would be: (1) television, (2) internet (with emphasis to the social networks) and (3) variable message panels. With respect to the DBQ questionnaire, seven behavioral factors based on the respondents own experience were identified as more important and related to unsafe attitudes: (1) poor fitness to drive, (2) violations, (3) distraction/inattention, (4) aggressive driving, (5) poor situation awareness, (6) stress and (7) behavioural adaptation. These factors represented the priority topics that should be handled in future campaigns.

The intense economic growth that Brazil has experienced in recent decades and its consequent explosive motorization process have resulted in an undesirable impact: the continuously increasing trend in traffic fatality numbers. This study presents a research on indicators and indexes with the objective of delivering both overall and disaggregated evidence about the road safety performance and targets in fatality reduction in Brazil at the state level taking the exposure into account. The intention is to support road safety strategic analysis in the country and to contribute to improve this critical scene. The methodological structure of this thesis consists of the following three main parts: (I) diagnosing the road safety situation at the state level using final outcome related information, in particular traffic fatality risk data; (II) setting a target number of traffic fatalities based on the relationship between the exposure level and the number of traffic fatalities in each state; and (III) suggesting domains for improvements based on the research of safety performance indicators representing three domains (road user, environment and vehicle) throughout the states. From a benchmarking point of view, we divided the Brazilian states into three separate clusters in order to provide more realistic state performance comparisons. After a data collection and indicators selection step, Data Envelopment Analysis (DEA) was the method used for executing the different steps, with the application of four different types of models specially developed for the identified research purposes. In addition, by means of bootstrapping the DEA scores we measured the sensitivity of the results to possible variations in the input data, for example concerning data quality and availability. As a result, we provided a road safety diagnosis per state as well as traffic fatality targets according to different perspectives: the entire group of road users (motorized and nonmotorized ones), motor vehicle occupants, and finally a disaggregated performance evaluation by running four separate DEA models (for motorcycle, car, truck and bus). Moreover, the SPI research including a hierarchy of 27 safety performance indicators expressed the states relative performance on the main road safety domains. Lastly, state profiles compiling all this information summarized the \"per state\" findings. / O intenso crescimento econômico que o Brasil tem experimentado nas últimas décadas e seu consequente explosivo processo de motorização resultaram em um impacto indesejado: a tendência contínua do aumento do número de mortes no trânsito. Este estudo apresenta uma pesquisa acerca de índices e indicadores com o objetivo de fornecer evidências gerais e desagregadas sobre o desempenho da segurança viária e metas de redução no número de mortes no Brasil no âmbito estadual, levando a exposição em consideração. A intenção é embasar uma análise estratégica da segurança viária no país e contribuir para melhorar este cenário crítico. A estrutura metodológica desta tese consiste das seguintes três partes principais: (I) diagnóstico da situação da segurança viária no nível estadual utilizando informações relacionadas ao resultado final, em particular dados de risco de morte no trânsito; (II) estabelecer uma meta para o número de mortes no trânsito para cada estado; e (III) sugerir domínios para melhorias baseado em pesquisa de indicadores de desempenho da segurança viária voltada a três domínios (usuário da via, ambiente e veículo). Sob a ótica do benchmarking, dividiram-se os estados brasileiros em três clusters para proporcionar comparações mais realistas dos desempenhos estaduais. Após uma etapa de coleta e seleção de indicadores, utilizou-se o método de Data Envelopment Analyis (DEA) para executar as diferentes etapas, com a aplicação de quatro tipos distintos de modelos especialmente desenvolvidos para os propósitos da pesquisa. Além disso, por meio de bootstrapping dos escores obtidos com a DEA, mediu-se a sensibilidade dos resultados a possíveis variações nos dados de entrada, no que diz respeito a, por exemplo, qualidade e disponibilidade dos dados. Como resultado, propicia-se, a partir de diferentes perspectivas, um diagnóstico da segurança viária por estado, assim como metas no número de mortes: para todo o grupo de usuários (motorizados e não-motorizados), ocupantes de veículos motorizados, e finalmente uma avaliação desagregada por meio de quatro modelos separados (para motocicletas, automóveis, caminhões e ônibus). Adicionalmente, a pesquisa de indicadores de desempenho da segurança considerando a hierarquia de 27 indicadores expressou os desempenhos relativos dos estados nos principais domínios da segurança viária. Por fim, perfis estaduais compilando todas estas informações resumem os resultados para os estados.

The unavailability of crash-related data has been a long lasting challenge in Brazil. In addition to the poor implementation and follow-up of road safety strategies, this drawback has hampered the development of studies that could contribute to national goals toward road safety. In contrast, developed countries have built their effective strategies on solid data basis, therefore, investing a considerable time and money in obtaining and creating pertinent information. In this research, we aim to assess the potential impacts of supplementary data on spatial model performance and the suitability of different spatial modeling approaches on crash prediction. The intention is to notify the authorities in Brazil and other developing countries, about the importance of having appropriate data. In this thesis, we set two specific objectives: (I) to investigate the spatial model prediction accuracy at unsampled subzones; (II) to evaluate the performance of spatial data analysis approaches on crash prediction. Firstly, we carry out a benchmarking based on Geographically Weighted Regression (GWR) models developed for Flanders, Belgium, and São Paulo, Brazil. Models are developed for two modes of transport: active (i.e. pedestrians and cyclists) and motorized transport (i.e. motorized vehicles occupants). Subsequently, we apply the repeated holdout method on the Flemish models, introducing two GWR validation approaches, named GWR holdout1 and GWR holdout2. While the former is based on the local coefficient estimates derived from the neighboring subzones and measures of the explanatory variables for the validation subzones, the latter uses the casualty estimates of the neighboring subzones directly to estimate outcomes for the missing subzones. Lastly, we compare the performance of GWR models with Mean Imputation (MEI), K-Nearest Neighbor (KNN) and Kriging with External Drift (KED). Findings showed that by adding the supplementary data, reductions of 20% and 25% for motorized transport, and 25% and 35% for active transport resulted in corrected Akaike Information Criterion (AICc) and Mean Squared Prediction Errors (MSPE), respectively. From a practical perspective, the results could help us identify hotspots and prioritize data collection strategies besides identify, implement and enforce appropriate countermeasures. Concerning the spatial approaches, GWR holdout2 out performed all other techniques and proved that GWR is an appropriate spatial technique for both prediction and impact analyses. Especially in countries where data availability has been an issue, this validation framework allows casualties or crash frequencies to be estimated while effectively capturing the spatial variation of the data. / A indisponibilidade de variáveis explicativas de acidentes de trânsito tem sido um desafio duradouro no Brasil. Além da má implementação e acompanhamento de estratégias de segurança viária, esse inconveniente tem dificultado o desenvolvimento de estudos que poderiam contribuir com as metas nacionais de segurança no trânsito. Em contraste, países desenvolvidos tem construído suas estratégias efetivas com base em dados sólidos, e portanto, investindo tempo e dinheiro consideráveis na obtenção e criação de informações pertinentes. O objetivo dessa pesquisa é avaliar os possíveis impactos de dados suplementares sobre o desempenho de modelos espaciais, e a adequação de diferentes abordagens de modelagem espacial na previsão de acidentes. A intenção é notificar as autoridades brasileiras e de outros países em desenvolvimento sobre a importância de dados adequados. Nesta tese, foram definidos dois objetivos específicos: (I) investigar a acurácia do modelo espacial em subzonas sem amostragem; (II) avaliar o desempenho de técnicas de análise espacial de dados na previsão de acidentes. Primeiramente, foi realizado um estudo comparativo, baseado em modelos desenvolvidos para Flandres (Bélgica) e São Paulo (Brasil), através do método de Regressão Geograficamente Ponderada (RGP). Os modelos foram desenvolvidos para dois modos de transporte: ativos (pedestres e ciclistas) e motorizados (ocupantes de veículos motorizados). Subsequentemente, foi aplicado o método de holdout repetido nos modelos Flamengos, introduzindo duas abordagens de validação para GWR, denominados RGP holdout1 e RGP holdout2. Enquanto o primeiro é baseado nas estimativas de coeficientes locais derivados das subzonas vizinhas e medidas das variáveis explicativas para as subzonas de validação, o último usa as estimativas de acidentes das subzonas vizinhas, diretamente, para estimar os resultados para as subzonas ausentes. Por fim, foi comparado o desempenho de modelos RGP e outras abordagens, tais como Imputação pela Média de dados faltantes (IM), K-vizinhos mais próximos (KNN) e Krigagem com Deriva Externa (KDE). Os resultados mostraram que, adicionando os dados suplementares, reduções de 20% e 25% para o transporte motorizado, e 25% e 35% para o transporte ativo, foram resultantes em termos de Critério de Informação de Akaike corrigido (AICc) e Erro Quadrático Médio da Predição (EQMP), respectivamente. Do ponto de vista prático, os resultados poderiam ajudar a identificar hotspots e priorizar estratégias de coleta de dados, além de identificar, implementar e aplicar contramedidas adequadas. No que diz respeito às abordagens espaciais, RGP holdout2 teve melhor desempenho em relação a todas as outras técnicas e, provou que a RGP é uma técnica espacial apropriada para ambas as análises de previsão e impactos. Especialmente em países onde a disponibilidade de dados tem sido um problema, essa estrutura de validação permite que as acidentes sejam estimados enquanto, capturando efetivamente a variação espacial dos dados.