En la presente tesis se ha desarrollado la evaluación estructural de cinco reservorios circulares del tipo apoyado, construidos entre los años 1977 y 1997, ubicados en zonas de alto riesgo sísmico en Lima Metropolitana y ubicados en suelos medianamente rígidos, con el objetivo de evidenciar si estas estructuras continúan conservando un diseño sísmico adecuado en base a los requerimientos sísmicos actuales, y por ende si serán capaces de resistir un evento sísmico severo y continuar con el servicio. Para poder modelar y determinar la respuesta de los se reservorios se empleó el modelo equivalente de Housner, obteniendo así la masa impulsiva y convectiva, modelado en el programa SAP2000 con ayuda de las normas ACI 350.3-06 y E.030. En cuanto a la determinación de las fuerzas resistentes, para poder realizar la evaluación estructural correspondiente, se utilizó la norma peruana E.060-2009 Concreto Armado, con la cual se obtuvo dichas fuerzas y se realizaron las verificaciones estructurales. Con respecto a los resultados de las verificaciones realizadas, se observó que los reservorios en estudio no mantienen un diseño estructural adecuado en cuanto a las solicitaciones sísmicas actuales. Estas deficiencias se plasman en déficit de refuerzo horizontal por corte en muros, cuantía mínima vertical por corte en muros, refuerzo en la base del muro por momento tangencial, armadura requerida en la viga collarín, y refuerzo en el extremo de la cúpula por tracción radial; por lo que estas estructuras, ante la presencia de un evento sísmico severo, se encuentran expuestas a presentar fallas estructurales. / In the present thesis has been carried out the structural assessment of five round ground concrete tanks, built between 1977 and 1997, and located in high seismic risk areas in Lima Metropolitana in moderately rigid soils, with the objective of demonstrating if these structures still preserve an adequate structural design base on the current standards and consequently, if they will be able to withstand a severe seismic event and, hence, continue with their service. In order to model and determine the response of the tanks, the Housner’s rigid equivalent model was used, obtaining this way the impulsive and convective masses, which were modeled in the software SAP2000 with the ACI 350.3-06 standard and the E.030 Peruvian standard. Regarding on the determination of the resistant forces, in order to carry out the corresponding structural evaluation, the Peruvian standard “Concreto Armado E.060” was utilized. With regard to the results of the verifications carried out, it was observed that the reservoirs under study do not maintain an adequate structural design in terms of the current seismic solicitations. These deficiencies are reflected in horizontal reinforcement deficit by shear force on the walls, minimum amount of vertical rebar by shear on the walls, reinforcement in the base of the wall by tangential bending moment, rebar required in the beam by radial tensile force, and rebar in the end of the dome by radial traction; so these structures, in the presence of a severe seismic event, are exposed to structural failures. / Tesis

Le risque sismique est un sujet d’étude pluridisciplinaire qui fait l’objet de nombreux travaux de recherches. Pendant longtemps, il a été étudié sous l’aspect de l’aléa et ce n’est qu’au milieu du XXe siècle que nous nous sommes intéressés à la vulnérabilité des éléments exposés. Malgré la multiplicité des études sur le risque sismique, aucune d’entre elles n’adopte une démarche globale en utilisant la règlementation parasismique. Dans le cadre de cette thèse, nous soutenons l’hypothèse selon laquelle il est possible d’évaluer la vulnérabilité des lieux d’habitation à l’échelle de plusieurs bâtiments en utilisant la norme Européenne, l’Eurocode 8. Utiliser cette règlementation a l’avantage de réduire les temps d’étudede la vulnérabilité physique puisque l’on évalue la résistance sismique d’un unique bâtiment dont ce dernier représente une population de plusieurs immeubles à usage d’habitation collective. La méthodologie proposée, illustrée sur l’exemple de la conurbation Mulhouse-Bâle, se compose de deux phases. La première consiste à étudier l’aléa sismique de la zone urbaine de Mulhouse et de Bâle à travers les études bibliographiques de quelques auteurs. Cette phase consiste aussi à examiner la compatibilité de la règlementation parasismique européenne et helvétique. En dernier lieu, un diagnostic du bâti existant et de la population est réalisé pour évaluer la vulnérabilité de ces deux territoires urbains, après un découpage des deux villes en secteurs historico géographiques. Une seconde phase consiste à proposer un modèle simplifié d’évaluation déterministe et probabiliste de la vulnérabilité du bâti, Celui-ci est fondé à partir de la nouvelle règlementation Européenne et de la mécanique des structures, pour évaluer la résistance sismique des bâtiments. L’aspect probabiliste a permis d’affiner le modèle proposé afin d’intégrer certaines incertitudes. Une étude de cas simulant un séisme important de magnitude Mw égale à 6 sur l’échelle de Richter, intégrant les phénomènes d’effets de site comme le préconise l’Eurocode 8, a permis de valider l’application du modèle envisagé. Le modèle d’évaluation proposé a pour intérêt de fournir un outil permettant d’évaluer la vulnérabilité du bâti sans effectuer de calcul mécanique. Il se veut donc accessible à tous(géographes, ingénieurs, sismologues, etc.…). Plus généralement, ce modèle pour objectif de fournir un outil d’aide à la décision dans la démarche de prévention que doivent les autorités publiques à la population, puisqu’ils permettent de déterminer la plus ou moins grande vulnérabilité des zones étudiées. / The seismic risk is a subject of multidisciplinary study which is the object of numerous research works. For a long time, it was studied in terms of hazard and it is only in the middle of the 20th century that we became interested in the vulnerability of the exposed elements. In spite of the multiplicity of the studies on the seismic risk, none of them adopts a global approach by using the earthquake-resistant regulations. Within the framework of thesis, we support the hypothesis that it possible to estimate the vulnerability of dwellings on the scale of several buildings by using the European standard, Eurocode 8. Using these regulations has the advantage reducing the time to study physical vulnerability by assessing the seismic resistance of a single building, where the latter represents a population of several buildings used as collective dwellings. The proposed methodology, illustrated on the example of the Mulhouse-Basel conurbation, consists of two phases. The first one consists in studying the seismic hazard of the urban area of Mulhouse and Basel through the bibliographical studies of some authors. This phase also consists in examining the compatibility of the European and Helvetian seismic regulations. Finally, a diagnosis of the existing structures and of the population is made to assess the vulnerability of these two urban territories, after a division of both cities into historic-geographical sectors. A second phase consists in proposing a simplified model of deterministic and probabilistic assessment of the vulnerability of the built, based on the new European regulation and the mechanics of the structures, to evaluate the seismic resistance of buildings. The probability aspect allowed to refine the proposed model to integrate certain uncertainties. A case study feigning an important earthquake of magnitude Mw equal to 6 on the Richter scale, integrating the phenomena of site effects as recommended by Eurocode 8, validated the application of the envisaged model. The proposed evaluation model is intended to provide a tool for assessing the vulnerability of the built without performing mechanical calculations. Thus, it aims to be accessible to all (geographers, engineers, seismologists, etc…). More generally, this model aims to provide a decision-making tool in the approach of prevention which the public authorities owe to the population, because they allow to determine the more or less big vulnerability of the studied areas.

Nuclear power plays a crucial role in energy supply in the world: around 15% of the electricity generated worldwide is provided from nuclear stations avoiding around 2.5 billion tonnes of CO2 emissions. As of January 2016, 442 reactors that generated 380+ GW were in operation and 66 new reactors were under construction. The seismic design of new nuclear power plants (NPPs) has gained much interest after the high-profile Fukushima Dai-ichi accident. In the UK, a tectonically stable continental region that possesses medium-to-low seismic activity, strong earthquakes capable of jeopardising the structural integrity of NPPs, although infrequent, can still occur. Despite that no NPP has been built in Great Britain after 1995, a New Build Programme intended to build 16 GW of new nuclear capacity by 2030 is currently under way. This PhD project provides a state-of-the-art framework for seismic probabilistic safety assessment and risk control of NPPs in Northwest Europe with particular application to the British Isles. It includes three progressive levels: (i) seismic input, (ii) seismic risk analysis, and (iii) seismic risk control. For seismic input, a suitable model to rationally define inputs in the context of risk assessments is proposed. Such a model is based on the stochastic simulation of accelerograms that are compatible with seismic scenarios defined by magnitude 4 < Mw < 6.5, epicentral distance 10 km < Repi < 100 km, and different types of soil (rock, stiff soil and soft soil). It was found to be a rational approach that streamlines the simulation of accelerograms to conduct nonlinear dynamic analyses for safety assessments. The model is a function of a few variables customarily known in structural engineering projects. In terms of PGA, PGV and spectral accelerations, the simulated accelerograms were validated by GMPEs calibrated for the UK, Europe and the Middle East, and other stable continental regions. For seismic risk analysis, a straightforward and logical approach to probabilistically assess the risk of NPPs based on the stochastic simulation of accelerograms is studied. It effectively simplifies traditional approaches: for seismic inputs, it avoids the use of selecting/scaling procedures and GMPEs; for structural outputs, it does not use Monte Carlo algorithms to simulate the damage state. However, it demands more expensive computational resources as a large number of nonlinear dynamic analyses are needed. For seismic risk control, strategies to control the risk using seismic protection systems are analysed. This is based on recent experience reported elsewhere of seismically protected nuclear reactor buildings in other areas of medium-to-low seismic activity. Finally, a scenario-based incremental dynamic analysis (IDA) is proposed aimed at the generation of surfaces for unacceptable performance of NPPs as function of earthquake magnitude and distance. It was found that viscous-based devices are more efficient than hysteretic-based devices in controlling the seismic risk of NPPs in the UK. Finally, using the proposed scenario-based IDA, it was found that when considering all controlling scenarios for a representative UK nuclear site, the risk is significantly reduced ranging from 3 to 5 orders of magnitude when using viscous-based devices.

La evolución es intrínseca en el ser humano y con nosotros evoluciona nuestro entorno. La industria de la construcción a manos de la necesidad del hombre, busca no solo un crecimiento horizontal territorial, sino también uno vertical. Sin embargo, esta proyección en las alturas se limita a una gestión plenamente espacial con poco interés en la capacidad de tránsito y zonas de refugio del agente humano durante un evento sísmico. Esta investigación realiza una evaluación de un escenario real existente del tramo de la avenida en estudio y el análisis de escenarios hipotéticos con variación del ancho de la zona de refugio. De este modo, se propone valores límite de altura y densidad poblacional de edificaciones para una Gestión Prospectiva del Riesgo Sísmico teniendo como ejemplo base un tramo de la Av. San Felipe en Lima-Perú. Se realizó el análisis de una evacuación eficiente de las edificaciones hacia la zona de refugio con modelos basados en el agente con el software Pathfinder, modelos tridimensionales (BIM), se analizó tiempos de evacuación, la influencia de las edificaciones, comportamiento humano y los parámetros urbanísticos zonales. Los resultados del presente trabajo establecen gráficamente la pertinente relación directa de la altura máxima permisible de los edificios proyectados con el ancho de la zona de refugio para una correcta evacuación. Como ejemplo se tiene un tramo de la Av. San Felipe con 7m de ancho de la zona de refugio correspondiente a 18 pisos como altura límite en los 10 edificios adyacentes considerando una longitud frontal total de 144m. / Evolution is intrinsic to human beings and our environment evolves with us. The construction industry at the hands of man's need, seeks not only a territorial horizontal growth, but also a vertical one. However, this projection in the heights is limited to a fully spatial management with little interest in the transit capacity and refuge areas of the human agent during a seismic event. This research carries out an evaluation of a real existing scenario of the section of the avenue under study and the analysis of hypothetical scenarios with variation in the width of the refuge area. In this way, limit values ​​of height and population density of buildings are proposed for a Prospective Management of Seismic Risk, taking as a base example a section of Av. San Felipe in Lima-Peru. The analysis of an efficient evacuation of the buildings towards the refuge area was carried out with models based on the agent with the Pathfinder software, three-dimensional models (BIM), evacuation times, the influence of the buildings, human behavior and the parameters were analyzed. zonal urban planning. The results of the present work graphically establish the pertinent direct relationship of the maximum permissible height of the projected buildings with the width of the refuge area for a correct evacuation. As an example, there is a section of Av. San Felipe with a 7m width of the refuge area corresponding to 18 floors as the limit height in the 10 adjacent buildings considering a total frontal length of 144m. / Tesis

Perú se localiza en una zona de alta sismicidad, debido a que se encuentra encima del área de subducción entre la placa tectónica de Nazca y Sudamericana, perteneciente al cinturón de fuego del Pacífico. Perú es un país en vía de desarrollo, por lo que es de suma importancia estar preparados para auxiliar a los miles de damnificados que pueda haber ante un evento sísmico importante. La evaluación del riesgo sísmico de edificaciones esenciales como colegios y hospitales es necesario para trabajos de reforzamiento estructural en este tipo de infraestructura. En el presente artículo científico, se presenta cuán vulnerables son los colegios públicos del distrito de San Juan de Miraflores en la ciudad de Lima ante un evento sísmico. Para ello, se utilizó la metodología de Inspección Visual Rápida del FEMA P-154. Además, se analizó de forma cuantitativa el pabellón 780 Pre, un módulo educativo estandarizado y construido en los años noventa cuya presencia es frecuente en dicho distrito. Para ello, se realizó un análisis no-lineal estático y no-lineal dinámico. Los resultados de la investigación concluyen que la mayoría de las edificaciones educativas presentan una alta vulnerabilidad sísmica y no cumplen con los requerimientos de uso post-sismo como se exige en la norma sismorresistente; así como también se verificó la deficiencia del módulo 780 Pre frente a un sismo severo cuando este fue sometido a los análisis no-lineales. / Peru is located in a high seismicity zone because it is set above the subduction area between the Nazca and South American tectonic plates, both belonging to the Pacific’s Ring of Fire. Being a developing country, it is of utmost importance to be prepared to help the thousands of victims that may be in the face of a major seismic event. The assessment of the seismic vulnerability of essential buildings —such as schools and hospitals— is necessary for structural reinforcement procedures in this type of infrastructure if needed. In this thesis, it is presented how vulnerable are the public schools of the district of San Juan de Miraflores in the city of Lima to a seismic event. For this, the FEMA P-154 Rapid Visual Screening methodology was used. In addition, the “780 Pre” public school building, a standardized educational building built in the 1990s and whose presence is frequent in that district, was analyzed quantitatively. For this, a static nonlinear and dynamic nonlinear analysis were performed. The results of the investigation show that most of the educational buildings present high seismic vulnerability and do not meet the requirements of post-earthquake use as required by the Peruvian seismic design provisions. Also, the deficiency of the 780 Pre building against a severe earthquake when it was subjected to non-linear analyzes was verified. / Tesis

En la investigación se analiza el estado de las viviendas de albañilería confinada construidas sin asistencia profesional en zonas con alto riesgo sísmico del distrito de Villa El Salvador. Se estudian los errores constructivos mediante la inspección visual de las casas autoconstruidas, las cuales presentan una alta vulnerabilidad frente a un terremoto de gran magnitud, en los sectores I (zona 1) y III (zona 2) del distrito. El error constructivo más frecuente que se ha podido identificar es el uso de ladrillos tubulares (ladrillo pandereta) con un fin estructural. Frente a este problema, la propuesta para mejorar la resistencia a la compresión y fuerzas cortantes de los muros de albañilería de pandereta es reforzarlos con malla de alambre galvanizado recubierta con mortero. Lo que se busca es proporcionar, por lo menos, un área segura dentro de las viviendas para que las familias puedan salvaguardar la vida de sus ocupantes frente a la ocurrencia de movimientos telúricos. En ese sentido, mediante ensayos de laboratorio se verifica que el reforzamiento con malla galvanizada mejora, en gran medida, la resistencia a la compresión y al corte de los muros. Además, su implementación resulta ser más económica frente a otras técnicas de reforzamiento como el refuerzo con mallas electrosoldadas. Esto es importante, porque la población beneficiaria pertenece a los niveles socioeconómicos C y D (bajo nivel socioeconómico). / The research analyzes the state of confined masonry houses built without professional assistance in areas with high seismic risk in the district of Villa El Salvador. Constructive errors are studied by visual inspection of self – constructed houses, which have a high vulnerability in the face of a large earthquake, in sectors I (Zone 1) and III (Zone 2) of the district. The most frequent constructive error that has been identified is the use of tubular bricks (pandereta brick) with a structural purpose. In the face of this problem, the proposal to improve the compression resistance and cutting forces of the pandereta masonry walls is to reinforce them with galvanized wire mesh coated with mortar. What is sought is to provide, at least, a safe area within homes so that families can safeguard the lives of their occupants in the face of the occurrence of telluric movements. In this sense, through laboratory tests it is verified that the reinforcement with galvanized mesh improves, to a large extent the resistance to the compression and the cutting of the walls. In addition, its implementation turns out to be more economical compared to other reinforcement techniques such as reinforcing with electro welded meshes. This is important, because the beneficiary population belongs to the socioeconomic levels C and D (low socioeconomic level). / Tesis

La presente investigación se realizó para evaluar el desempeño estructural y funcional del sistema de evacuación, de un edificio educativo superior durante un sismo severo, en atención a lo cual, se analizó mediante el método basado en Agent Based Modelling y Análisis Tiempo-Historia no Lineal. No obstante, el análisis no se limita solo a edificaciones esenciales; se puede aplicar para edificios importantes y comunes que concentran gran cantidad de personas, ya que el lineamiento que se plantea en esta tesis busca una alternativa más que se debe tener en cuenta en los proyectos de la construcción civil. La información estructural y geotécnica del edificio en estudio, se recopila y se introduce en una base de datos para su análisis. Se realiza un procedimiento similar para la información relacionada de los ocupantes. Usando estas informaciones: a) Se determinará la fragilidad estructural y el colapso localizado, b) se establecerá la interacción de la persona con el colapso focalizado. Para el primer aspecto, se utilizará el historial de tiempo no lineal; para el segundo, el modelado basado en agentes se empleará para recrear la reacción de las personas que enfrentan el micro colapso. Los resultados importantes de estas evaluaciones son: 1) Se ha localizado vigas, columnas y muros de corte colapsadas que afectan la evacuación; 2) Se ha ubicado de cuellos de botella, donde presentan gran concentración de personas durante la evacuación; y 3) se ha cuantificado a las personas afectadas, en términos de individuos atrapadas en el edificio que no pudieron evacuar. / The present investigation was carried out to evaluate the structural and functional performance of the evacuation system of a high-rise higher education building during a severe earthquake, which was analyzed using the Agent Based Modeling and Non-Linear Time-History Analysis method. However, the analysis is not limited only to essential buildings; it can be applied to important and common buildings that concentrate a large number of people, such as: Health establishments, airports, shopping centers, stadiums, condominiums, etc., since the guideline proposed in this thesis seeks one more alternative to be taken into account in civil construction projects. The structural and geotechnical information of the building under study is compiled and entered into a database for analysis. A similar procedure is performed for the related occupant information. Using this information: a) the structural fragility and localized collapse will be determined, b) the interaction of the person with the focused collapse will be established. For the first aspect, nonlinear time history will be used; for the second, agent-based modeling will be used to recreate the reaction of people facing the micro collapse. The important results of these evaluations are: 1) collapsed beams, columns and shear walls have been located, which make evacuation routes inoperable; 2) bottleneck areas have been located, where they present high concentration of people during evacuation; and 3) affected people have been quantified, in terms of individuals trapped in the building who could not evacuate. / Tesis

Recent research interests in studying the performance of different seismic force resisting systems (SFRS) have been shifting from component- (individual walls) to system-level (complete building) studies. Although there is wealth of knowledge on component-level performance of reinforced masonry shear walls (RMSW) under seismic loading, a gap still exists in understanding the response of these components within a complete system. Consequently, this study’s main objective is to investigate the influence of the diaphragm’s out-of-plane stiffness on the seismic response of RMSW buildings. In addition, the study aims to synthesize how this influence can be implemented in different seismic design approaches and assessment frameworks. To meet these objectives a two-story scaled asymmetrical RMSW building was tested under quasi-static cyclic loading. The analysis of the test results showed that the floor diaphragms’ out-of-plane stiffness played an important role in flexurally coupling the RMSW aligned along the loading direction with those walls orthogonal to it. This system-level aspect affected not only the different wall strength and displacement demands but also the failure mechanism sequence and the building twist response. The results of the study also showed that neglecting diaphragm flexural coupling influence on the RMSW at the system-level may result in unconservative designs and possibly undesirable failure modes. To address these findings, an analytical model was developed that can account for the aforementioned influences, in which, simplified load-displacement relationships were developed to predict RMSW component- and system-level responses under lateral seismic loads. This model is expected to give better predictions of the system response which can be implemented, within the model limitations, in forced- and displacement-based seismic design approaches. In addition, and in order to adapt to the increasing interest in more resilient buildings, this study presents an approach to calculate the system robustness based on the experimental data. Finally, literature shows that the vast majority of the loss models available for RMSW systems were based on individual component testing and/or engineering judgment. Consequently, this study proposes system damage states in lieu of component damage states in order to enhance the prediction capabilities of such models. The current dissertation highlights the significant influence of the diaphragm out-of-plane stiffness on the system-level response that may alter the RMSW response to seismic events; an issue that need to be addressed in design codes and standards. / Dissertation / Doctor of Philosophy (PhD)

Le problème de la dimension stochastique élevée est récurrent dans les analyses probabilistes des structures. Il correspond à l’augmentation exponentielle du nombre d’évaluations du modèle mécanique lorsque le nombre de paramètres incertains est élevé. Afin de pallier cette difficulté, nous avons proposé dans cette thèse, une approche à deux étapes. La première consiste à déterminer la dimension stochastique efficace, en se basant sur une hiérarchisation des paramètres incertains en utilisant les méthodes de criblage. Une fois les paramètres prépondérants sur la variabilité de la réponse du modèle identifiés, ils sont modélisés par des variables aléatoires et le reste des paramètres est fixé à leurs valeurs moyennes respectives, dans le calcul stochastique proprement dit. Cette tâche fut la deuxième étape de l’approche proposée, dans laquelle la méthode de décomposition de la dimension est utilisée pour caractériser l’aléa de la réponse du modèle, par l’estimation des moments statistiques et la construction de la densité de probabilité. Cette approche permet d’économiser jusqu’à 90% du temps de calcul demandé par les méthodes de calcul stochastique classiques. Elle est ensuite utilisée dans l’évaluation de l’intégrité d’une toiture à ossature bois d’une habitation individuelle installée sur un site d’aléa sismique fort. Dans ce contexte, l’analyse du comportement de la structure est basée sur un modèle éléments finis, dans lequel les assemblages en bois sont modélisés par une loi anisotrope avec hystérésis et l’action sismique est représentée par huit accélérogrammes naturels fournis par le BRGM. Ces accélérogrammes permettent de représenter différents types de sols selon en se référant à la classification de l’Eurocode 8. La défaillance de la toiture est définie par l’atteinte de l’endommagement, enregistré dans les assemblages situés sur les éléments de contreventement et les éléments d’anti-flambement, d’un niveau critique fixé à l’aide des résultats des essais. Des analyses déterministes du modèle éléments finis ont montré que la toiture résiste à l’aléa sismique de la ville du Moule en Guadeloupe. Les analyses probabilistes ont montré que parmi les 134 variables aléatoires représentant l’aléa dans le comportement non linéaire des assemblages, 15 seulement contribuent effectivement à la variabilité de la réponse mécanique ce qui a permis de réduire la dimension stochastique dans le calcul des moments statistiques. En s’appuyant sur les estimations de la moyenne et de l’écart-type on a montré que la variabilité de l’endommagement dans les assemblages situés dans les éléments de contreventement est plus importante que celle de l’endommagement sur les assemblages situés sur les éléments d’anti-flambement. De plus, elle est plus significative pour les signaux les plus nocifs sur la structure. / The problem of the curse of dimensionality is frequently encountered in practical applications. It can be defined as the significant increase of the number of mechanical model calls with the number of uncertain parameters. To overcome this difficulty, a two-steps stochastic approach has been developed in this work. The first step of this approach consists in calculating the stochastic effective dimension by the means of Morris screening method. Once the most significant uncertain parameters on the variability of the mechanical responses are identified, they are modeled as random variables and the remaining parameters are fixed to their respective mean values. This allows us to reduce significantly the stochastic dimension of the problem in the second step of the approach where the decomposition method is used to estimate the statistical characteristics of the mechanical responses. The efficiency and the accuracy of this approach are evaluated through an academic problem dealing with the assessment of the integrity of a three-span five-story frame structure subjected to horizontal loads. We have demonstrate that we can reduce about 90% of the computation time required by the classical stochastic methods. Then, the proposed approach is used to the analysis of the integrity of timber roofs under seismic loading. The behaviour of this structure is described through a finite element model where the timber joints are modeled by anisotropic hysteresis law, and the seismic action is represented by eight real earthquake ground motion records. These accelerograms provided by the French institution involved in geosciences BRGM allow us to take into account different soil types according to the classification provided by the europeen design code dealing with seismic events Eurocode 8. The failure of timber roofs is reached when the damage levels in the timbers joints localized on the buckling and bracing members reach the critical value. It is shown, through a deterministic analysis, that the structure resists the seismic hazard representing the city of Le Moule in Guadeloupe. The stochastic analysis has shown that, among the 134 random variables representing the uncertainty in the nonlinear behaviour of the timber joints, only 15 have a significant effect on the variability of the structural response, which allow us to reduce the stochastic dimension in the computation of the statistical moments. According to the estimates of the mean and the standard deviation, we have shown that the variability of bracing members damage is greater than the variability of buckling members damage. Moreover, the variability of the bracing members damage is more significant for the earthquake ground motion records having the lowest collapse PGA.