Les travaux menés dans cette thèse CIFRE s’inscrivent dans le cadre d’une collaboration pérenne entre le CRAN et l'EDF R&D dont un des résultats majeurs a été le développement d'une méthodologie d’analyse de risques, appelée Analyse Intégrée des Risques (AiDR). Cette méthodologie traite des systèmes sociotechniques sous les angles technique, humain et organisationnel et dont les équipements sont soumis à des actions de maintenance et/ou de conduite. La thèse a pour objet ainsi de proposer une évolution du modèle dit de « barrière humaine » développé dans l'AiDR pour évaluer l'efficacité de ces actions humaines prises leur contexte organisationnel. Nos contributions majeures s'organisent autour de 3 axes : 1. Une amélioration de la structure préexistante du modèle de barrière humaine afin d’aboutir à un modèle basé sur des facteurs de forme appelés performance shaping factors (PSF) fournis par les méthodes d’Évaluation Probabiliste de la Fiabilité Humaine (EPFH) ;2. L’intégration de la résilience et la modélisation de l’interaction entre mécanismes résilients et pathogènes impactant l'efficacité des actions dans les relations causales probabilistes ;3. Un traitement global des jugements d’expert cohérent avec la structure mathématique du modèle proposé permettant d’estimer d’une manière objective les paramètres du modèle. Ce traitement se fonde sur la construction d’un questionnaire permettant de "guider" l’expert vers l’évaluation d’effets conjoints issus de l’interaction entre mécanismes pathogènes et résilients. L’ensemble des contributions proposées a été validé sur un cas d’application portant sur une barrière humaine mise en place dans un cas d’inondation externe d’une unité de production d’électricité d’EDF / The work carried out in this CIFRE PhD thesis is part of a long-term collaboration between CRAN and EDF R&D, one of the major results of which was the development of a risk analysis methodology called Integrated Risk Analysis (AiDR). This methodology deals with sociotechnical systems from technical, human and organizational points of view and whose equipment is subjected to maintenance and/or operation activities. This thesis aims to propose an evolution of the so-called "human barrier" model developed in the AiDR in order to evaluate the effectiveness of these human actions taken their organizational context. Our major contributions are organized around 3 axes: 1. Improvement of the pre-existing structure of the human barrier model to achieve a model based on performance shaping factors (PSF) provided by the Human Reliability Assessment (HRA) methods; 2. Integration of resilience and modeling of the interaction between resilient and pathogenic mechanisms impacting the effectiveness of activities in a probabilistic causal framework; 3. A global treatment of the expert judgments consistent with the mathematical structure of the proposed model in order to objectively estimate the parameters of the model. This treatment is based on a questionnaire to guide experts towards the evaluation of joint effects resulting from the interaction between pathogenic and resilient mechanisms. All of the proposed contributions have been validated on an application case involving a human barrier put in place during an external flooding occurring at an EDF power plant

<p></p><p>Visual inspection is the primary means of ensuring the safety and functionality of in-service bridges in the United States and owners spend considerable resources on such inspections. While the Federal Highway Administration (FHWA) and many state departments of transportation have guidelines related to inspector qualification, training, and certification, an inspector’s actual capability to identify defects in the field under these guidelines is unknown. This research aimed to address the knowledge gap surrounding visual inspection performance for steel bridges in order to support future advances in inspection and design procedures. Focusing primarily on fatigue crack detection, this research also considered the ability of inspectors to accurately and consistently estimate section loss in steel bridge members. </p> <p> </p> <p>Inspection performance was evaluated through a series of simulated bridge inspections performed in representative in-situ conditions. First, this research describes the results from 30 hands-on, visual inspections performed on full size bridge specimens with known fatigue cracks. Probability of Detection (POD) curves were fit to the inspection results and the 50% and 90% detection rate crack lengths were determined. The variability in performance was large, and only a small amount of the variance could be explained by individual characteristics or environmental conditions. Based on the results, recommendations for improved training methods, inspection procedures, and equipment were developed. Above all, establishment of a performance based qualification system for bridge inspectors is recommended to confirm that a satisfactory level of performance is consistently achieved in the field. </p> <p> </p> <p>Long term, managing agencies may eschew traditional hands-on bridge inspection methods in favor of emerging technologies imagined to provide improved results and fewer logistical challenges. This research investigated the potential for unmanned aircraft system (UAS) assistance during visual inspection of steel bridges. Using the same specimens as in the hands-on inspections, four UAS-assisted field inspections and 19 UAS-assisted desk inspections were performed. A direct comparison was made between performance in the hands-on and UAS-assisted inspections, as well as between performance in the two types of UAS-assisted inspections. Again, significant variability was present in the results suggesting that human factors continue to have a substantial influence on inspection performance, regardless of inspection method. </p> <p> </p> <p>Finally, to expand the findings from the crack detection inspections, the lower chord from a deck truss was used to investigate variability in the inspection of severely corroded steel tension members. Five inspectors performed a hands-on inspection of the specimen and four engineers calculated the load rating for the same specimen. Significant variability was observed in how inspectors recorded thickness measurements during the inspections and engineers interpreted the inspection reports and applied the code requirements. </p><br><p></p>

Obtaining an accurate estimate of a land-cover classifier's performance over a wide geographic area is a challenging problem due to the need to generate the ground truth that covers the entire area that may be thousands of square kilometers in size. The current best approach constructs a testing dataset by drawing samples randomly from the entire area --- with a human supplying the true label for each such sample --- with the hope that the selections thus made statistically capture all of the data diversity in the area. A major shortcoming of this approach is that it is difficult for a human to ensure that the information provided by the next data element chosen by the random sampler is non-redundant with respect to the data already collected. In order to reduce the annotation burden, it makes sense to remove any redundancies from the entire dataset before presenting its samples to a human for annotation. This dissertation presents a framework that uses a combination of clustering and compression to create a concise-set representation of the land-cover data for a large geographic area. Whereas clustering is achieved by applying Locality Sensitive Hashing (LSH) to the data elements, compression is achieved through choosing a single data element to represent a given cluster. This framework reduces the annotation burden on the human and makes it more likely that the human would persevere during the annotation stage. We validate our framework experimentally by comparing it with the traditional random sampling approach using WorldView2 satellite imagery.

Made available in DSpace on 2014-06-12T17:40:46Z (GMT). No. of bitstreams: 2 arquivo723_1.pdf: 2509449 bytes, checksum: aee89d3fd6cd28c43ba22ccfaced1689 (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2009 / Companhia Hidro Elétrica do São Francisco / Esta dissertação enfoca a questão do erro humano nas atividades relacionadas com a manutenção dos equipamentos de proteção e automação numa grande empresa de transmissão de energia elétrica. A ênfase é no erro humano que tenha provocado o desligamento acidental de alguma função de transmissão. A justificativa do estudo cita ações preventivas aos erros humanos na manutenção, apresenta dados básicos de desligamentos, resume os cenários antes e depois da resolução normativa que alterou a remuneração do setor de transmissão, ressalta os impactos econômicos com a nova regulamentação e aborda aspectos da cultura organizacional. A pesquisa sobre as causas de acidentes no trabalho resume teorias com ênfase no homem (da fadiga; da propensão para acidentes, do puro acaso) e teorias com visão sistêmica (modelos Skill-Rule-Knowledge e Generic Error-Modelling System). Aborda ainda aplicações das teorias de análise e classificação de erros humanos que resultaram no modelo Human Factors Analysis and Classification System. Conclui destacando a dificuldade e a necessidade das empresas abandonarem as análises de acidentes com foco nos erros dos homens para considerarem os fatores organizacionais, apresenta fatores psicológicos que levam a buscar culpados e propostas para alcançar excelentes níveis de segurança. O método proposto para a análise e classificação de erros humanos, composto de nove fases, estabelece análise ampla e participativa, apresenta fluxogramas e diagramas de apoio e um processo de validação das recomendações de modo a obter o comprometimento dos responsáveis pela implantação. Apresenta a visão da análise de acidentes no mundo, alerta para o perigo da tendência em supervalorizar o organizacional , insiste que as organizações não podem perder de vista os atos inseguros. Na aplicação é descrita a avaliação da influência da mudança na tecnologia dos equipamentos de proteção e automação em relação aos erros humanos sendo constatado forte efeito sobre a atividade de cálculo e emissão de ajustes dos equipamentos digitais. A conclusão ressalta as bases teóricas do método proposto, destaca os aplausos às empresas que conseguem evitar análises de acidentes com foco excessivo no humano e sugere estudos envolvendo psicologia, engenharia, psiquiatria, estatística, administração e pedagogia

There are many pieces of flight deck research on general use of written English language technical information and problem solving using technical documentation. Contributory causes of aircraft accidents have been due to misunderstandings of crew alerts and procedural divergence by English as-a-second language flight crewmembers (ESL). Research was conducted to understand impact of written English language technical information on ESL flight crewmembers’ performance. Two types of systems were evaluated, technical documentation and crew alerting systems that contain technical information, with respect to their impact on ESL flight crewmember performance. Preliminary analysis results indicated written English language technical information can be confusing, difficult to read and interpret, and leads to misunderstandings by ESL flight crewmembers during aircraft nonnormal conditions. English as-a-second language flight crewmembers indicated they often experience problems executing written English language technical procedures after outset of crew alerts. Conversely, experimental trials revealed ESL flight crewmembers did not experience many cognitive performance issues with use of crew alerting systems and technical information designed with an English language emphasis. English as-a second language flight crewmembers’ English language proficiency, background knowledge, and use of use of metacognitive strategies to read and comprehend written English language on crew alerting and information systems, indicated they utilized written English technical information with ease. Particularly, ESL flight crewmembers’ workload was low, they had fast response times to system faults, and they experienced minimal procedural deviations. On the contrary, when ESL flight crewmembers utilized written English language technical procedures translated into their native language during non-normal conditions, they experienced several cognitive performance challenges. English as-a second language flight crewmembers’ background knowledge of written English language technical information translated into their native language, use of metacognitive strategies to read and comprehend written English language translated into their native language, indicated they experienced difficulties with reading and comprehending translated technical information on information systems. Particularly, ESL flight crewmembers were challenged cognitively when they responded to crew alerts through execution of decision-making processes. They indicated translation of written English language technical information into their native language was a pre-cursor to procedural deviation, long response times to system issues, as well as high workload during experimental trials. It is recommended that further research focus on design and use of written English language technical documentation by ESL flight crewmembers during non-normal conditions. It is also recommended that if deemed practical by the aviation industry, further research should focus on design, integration, and utilization of technical documentation in a language(s) other than English, and measurement of ESL flight crewmembers performance on the flight deck.

Descreve o panorama do surgimento da aviação e das teorias de investigação de acidentes aeronáuticos. Objetivo - Analisar a investigação dos acidentes e incidentes aeronáuticos ocorridos na Aviação Geral no Estado de São Paulo no período de 2000 a 2005 e verificar sua associação a fatores organizacionais. Método - Foram comparadas as atuações das agências de investigação de acidentes aeronáuticos dos Estados Unidos, União Européia e Brasil, identificado os métodos de investigação de acidentes empregados. Foi realizado o levantamento das ocorrências com as aeronaves classificadas na categoria de aviação geral no estado de São Paulo no período 2000 a 2005 e feita a análise comparativa de 36 relatórios finais de acidentes utilizados pelo Centro de Investigação e Prevenção de Acidentes (CENIPA) com o Human Factors Analysis and Classification System (HFACS). Resultados Identificou-se que as agências de investigação dos Estados Unidos e União Européia atuam de forma sistêmica e a brasileira atua isoladamente. Foi constatado que houve 636 ocorrências com aeronaves da aviação geral, sendo que 92 por cento foram incidentes, que não foram investigados. Dos acidentes, 5,5 por cento possuíam relatórios finais concluídos. A análise dos relatórios finais, segundo o CENIPA, apontou 163 fatores contribuintes, sendo que o principal fator foi o deficiente julgamento presente em 80,5 por cento dos acidentes. A utilização do modelo HFACS identificou a presença de 370 fatores contribuintes e os erros de habilidade, de decisão e planejamento inadequado foram os principais fatores contribuintes com 86,1 por cento . Considerações Finais O estudo possibilitou a identificação da falta de integração de diversos órgãos governamentais na investigação dos acidentes aeronáuticos. A ausência de investigação de incidentes aeronáuticos dificultando a prevenção. A análise do CENIPA não contempla de forma adequada os fatores organizacionais. O HFACS não deve ser uma ferramenta apenas quantificadora das causas de acidentes aeronáuticos / Introduction Describes the birth of aviation and aeronautical accidents factors. investigation theories. Objective Analyze the investigation of aeronautical accidents and incidents occurring in the General Aviation in the State of São Paulo during the period of 2000 through 2005 and verify their association with organizational factors. Method The air accident agencies of the United States, European Union and Brazil were studied, identifying and comparing the accident investigation methods used. It was registered the number of events with aircrafts in the State of Sao Paulo from 2000 to 2005, and performed the comparative analysis of 36 accident final reports used by the Centro de Investigação e Prevenção de Acidentes (CENIPA; Accident Prevention and Investigation Center) using the Human Factors Analysis and Classification System (HFACS). Results It was observed that the investigation agencies of the United States and the European Union work in a systemic way, and that the Brazilian one works alone. It was observed that there were 636 events with aircrafts of the general aviation, of which 92 per cent were not-investigated incidents. Of the accidents, 5.5 per cent had their final reports finished. The analysis of the final reports according to CENIPA pointed out 163 contributing factors, being the main factor \"inadequate evaluation\", present in 80.5 per cent of the accidents. The HFACS model identified 370 contributing factors, and the \"skill errors\", \"decision\" and \"inadequate planning\" were the main contributing factors, being present in 86.1 per cent of the cases. Final Considerations The study allowed identifying the lack of integration of several administration agencies when investigating air accidents. The non investigation of air incidents hinders the prevention. The CENIPA analysis does not study properly the organizational factors. HFACS must not be just a quantifying tool of the air accidents causes.

Submitted by admin tede (tede@pucgoias.edu.br) on 2018-05-29T13:32:18Z No. of bitstreams: 1 CLEITON JOSÉ COSTA SANTOS 1.pdf: 1693805 bytes, checksum: 73acf521014b6d3136d8e759744648f7 (MD5) / Made available in DSpace on 2018-05-29T13:32:18Z (GMT). No. of bitstreams: 1 CLEITON JOSÉ COSTA SANTOS 1.pdf: 1693805 bytes, checksum: 73acf521014b6d3136d8e759744648f7 (MD5) Previous issue date: 2018-03-14 / Recently the World Health Organization reclassified the snakebites as Neglected Tropical Disease. This problem mainly affects economically active age groups of rural populations in poor or in developing countries in tropical regions. Despite the low mortality rate associated with this type of accident in Brazil, one of the largest estimates of occurrence was attributed to the country. The epidemiological profile is widely known in several Brazilian regions, but little was inferred about factors associated with the occurrence of snakebites in the country. In order to analyze the spatial distribution of snakebites caused by snakes of the Elapidae and Viperidae families in the Brazilian municipalities between 2007 and 2015, we associate the number of snakebites with variables related to human factors (Area of the Municipality, Population, Agricultural Production, Municipal Human Development Index and Gross Domestic Product) and environmental factors (Species Richness, Remnant Area of Native Vegetation, Altitude and Climate). The records of notifications were obtained from the database of the Ministry of Health's Information and Notification System of Injuries. The average annual incidence rate for the time series analyzed was calculated. The association between the number of snakebites and factors analyzed was tested through Multiple Linear Regression Analysis by an OLS (Ordinary Least Square) model, evaluating the spatial autocorrelation in the model residues by the Moran Index. The variance partition was performed to evaluate the isolated effect of the different factors (environmental, human and spatial) on the variation of the accident data. Data were analyzed at national and regional scales. The average annual incidence for the period was 12.1/100,000 inhabitants. The North and Midwest regions presented the highest incidence, followed by the Northeast, South and Southeast regions, respectively. The OLS model indicated association more than 50% between snakebites data and variables tested, with the highest contribution of human factors followed the spatial effect and lower contribution of environmental factors in all scales. The model showed higher suitability especially for regions with the highest incidence rate, indicating differences in the composition of the most important variables in each region. However, it is possible to infer that in the analysis period, human factors contributed most heavily to the occurrence of snakebites in Brazil, followed by the spatial effect on the distribution of these snakebites and minor contribution of environmental factors. Clearly, the demographic density linked to type of occupation and climatic conditions may favor the increase of snakebites in the country, and different regional characteristics can clarify the disparity in the incidence rate among the Brazilian regions. / Recentemente os acidentes ofídicos foram reclassificados pela Organização Mundial da Saúde como Doença Tropical Negligenciada. Esse problema atinge principalmente grupos etários economicamente ativos de populações rurais em países pobres ou em desenvolvimento nas regiões tropicais. Apesar do baixo índice de mortalidade associado a esse tipo de acidente no Brasil, uma das maiores estimativas de ocorrência foi atribuída ao país. O perfil epidemiológico é amplamente conhecido nas diversas regiões brasileiras, porém, pouco se inferiu sobre fatores associados à ocorrência de acidentes ofídicos no país. Assim, com o objetivo de analisar a distribuição espacial dos acidentes causados por serpentes das famílias Elapidae e Viperidae, registrados para os municípios brasileiros entre os anos de 2007 e 2015, associamos o número de acidentes ofídicos, obtidos na base de dados do Sistema de Informação e Notificação de Agravos do Ministério da Saúde, com variáveis ligadas a fatores humanos e ambientais considerando o efeito espacial na distribuição dos dados. A taxa de incidência média anual foi calculada para elaboração do mapa de distribuição nacional e regional da incidência no período analisado. A associação entre o número de acidentes e fatores analisados foi testada através de Análise de Regressão Linear Múltipla OLS (Ordinary Least Square), avaliando a autocorrelação espacial nos resíduos do modelo pelo Índice de Moran. A partição de variância foi realizada para avaliar o efeito isolado dos diferentes fatores sobre a variação do número de acidentes. A incidência média anual para o período foi de 12,1/100.000hab. As regiões Norte e Centro-Oeste apresentaram as maiores incidências, seguidas pelas regiões Nordestes, Sul e Sudeste, respectivamente. O modelo OLS indicou associação superior a 50% entre dados de acidentes e variáveis testadas, com maior contribuição dos fatores humanos seguidos do efeito espacial e menor contribuição dos fatores ambientais em todas as escalas analisadas. O modelo mostrou maior adequação principalmente para as regiões com maior taxa de incidência, indicando diferenças na composição das variáveis mais importantes em cada região. Contudo, é possível inferir que no período analisado, fatores humanos contribuíram mais fortemente para a ocorrência de acidentes ofídicos no Brasil, seguidos do efeito espacial sobre a distribuição desses acidentes e menor contribuição dos fatores ambientais. Especialmente, o adensamento demográfico ligado ao tipo de ocupação e condições climáticas podem favorecer o aumento de acidentes ofídicos no país e diferentes características regionais podem esclarecer a disparidade na taxa de incidência entre as regiões brasileiras.

abstract: Highly automated vehicles require drivers to remain aware enough to takeover during critical events. Driver distraction is a key factor that prevents drivers from reacting adequately, and thus there is need for an alert to help drivers regain situational awareness and be able to act quickly and successfully should a critical event arise. This study examines two aspects of alerts that could help facilitate driver takeover: mode (auditory and tactile) and direction (towards and away). Auditory alerts appear to be somewhat more effective than tactile alerts, though both modes produce significantly faster reaction times than no alert. Alerts moving towards the driver also appear to be more effective than alerts moving away from the driver. Future research should examine how multimodal alerts differ from single mode, and see if higher fidelity alerts influence takeover times. / Dissertation/Thesis / Masters Thesis Human Systems Engineering 2018

Este trabalho apresenta uma proposta de modelo de análise de risco para embarcações, incorporando em sua formulação a influência das interações entre os sistemas de alarmes e os fatores humanos e organizacionais presentes a bordo. De acordo com estatísticas de sociedades classificadoras, o homem é o maior responsável por acidentes marítimos, e, desta forma, é um dos principais componentes da segurança de uma embarcação. A relevância do elemento humano se dá pelo fato de que ações e decisões humanas estão relacionadas tanto à causa, quanto à prevenção de acidentes. O sistema de alarmes é um equipamento de segurança obrigatório, e com direta participação na dinâmica de um acidente, fornecendo informação para a tomada de decisão das ações por parte da tripulação na resolução do problema. O estudo da interação entre este equipamento e o elemento humano a bordo permite a gestores e armadores elaborar diretrizes para investimentos em sistemas de segurança e políticas que influenciam o desempenho humano, e, portanto, a segurança a bordo. O modelo proposto, inspirado em relatório de acidentes marítimos, tem como ponto de partida uma estrutura sequencial do acidente e leva em consideração uma sucessão usual e simplificada dos eventos ocorridos, iniciando a partir de uma falha no sistema físico. O elemento humano, então, é incorporado à análise de risco através das técnicas de análise de confiabilidade humana, as quais têm como princípio a visão do homem como mais um componente do sistema, ou o liveware interagindo com software e hardware. Sob este ponto de vista, uma abordagem sócio-técnica é aplicada, ou seja, considera-se que um navio é composto não apenas de sua estrutura e máquinas, mas também de toda a tripulação. Com o objetivo de exemplificar passos e hipóteses de uma aplicação do modelo proposto, é apresentada a aplicação ao caso do acidente da embarcação Maersk Doha, ocorrido em outubro de 2006 nos Estados Unidos, cujo relatório de investigação do acidente é de acesso público via Internet, no site da Marine Accident Investigation Branch MAIB. / This work presents a risk analysis model for ships, focusing on the scenarios where the crew interacts with the alarm and monitoring system. According to statistics of classification societies, humans are largely responsible for accidents on board and, therefore, are considered a major component of the safety of vessels. The relevance of the human element is given by the fact that human decisions and actions are related to the cause of accidents, either being the direct causative factor of failure or influencing the probability of failure, and the prevention of accidents or mitigation of the consequences. The alarm system is a mandatory component of certified vessels, with direct participation in an accident. It supplies information for the decision making process of the crew, considering their actions to recover the system. The study of the interactions between this automation equipment and the human element on board provides guidelines to managers and owners to invest in proper security systems and policies that influence human behavior, and therefore the safety on board. The model, inspired in accident reports, has as starting point a sequential structure of the accident, and takes into account a typical and simplified sequence of events, starting from a failure in the physical system. The human element is incorporated into the risk analysis through techniques of human reliability analysis, which place man as another component of the system, or the \"liveware\" interacting with software and hardware. From this point of view, a socio-technical approach is applied, considering that a ship is composed of not only its structure and machinery, but also of the entire crew. In order to illustrate the steps and assumptions to be done by an analyst applying the proposed model, the accident of the vessel Maersk Doha, occurred in October of 2006 in the United States, is analyzed. The report on the investigation of this accident is public and accessible via the Internet site of the Marine Accident Investigation Branch MAIB.

Traffic simulation has become an invaluable part of the traffic engineering toolbox. However, the majority of driver models are designed to recreate traffic performance based on interactions among vehicles. In keeping with this pursuit, most are fundamentally built to avoid collisions. This limits the applicability of using these models for addressing safety concerns, especially those regarding pedestrian safety performance. However, by explicitly including some of the sources of human error, these limitations can, in theory, be overcome. While much work has been done toward including these human factors in simulation platforms, one key aspect of human behavior has been largely ignored: driver distraction. This work presents a novel approach to inclusion of driver distraction in a microsimulation or agent-based model. Distributions of distraction events and inter-distraction periods are derived from eye-glance data collected during naturalistic driving studies. The developed model of distraction is implemented -- along with perception errors, visual obstructions, and driver reaction times -- in a simulated mid-block pedestrian crossing. The results of this simulation demonstrate that excluding any of these human factors from the implemented driver model significantly alters conflict rates observed in the simulation. This finding suggests that inclusion of human factors is important in any microsimulation platforms used to analyze pedestrian safety performance.