Repositorio Institucional Universidad EAFIT ::Examinando por Materia "Buildings"

Examinando por Materia "Buildings"

Mostrando 1 - 9 de 9

Automatic detection of building typology using deep learning methods on street level images
(PERGAMON-ELSEVIER SCIENCE LTD, 2020-03-20) Duque, J.; Gonzalez, D.; Rueda Plata, Diego; Acevedo, A.; Ramos, R.; Betancourt, A.; García, S.; Universidad EAFIT. Departamento de Economía y Finanzas; Research in Spatial Economics (RISE)
An exposure model is a key component for assessing potential human and economic losses from natural disasters. An exposure model consists of a spatially disaggregated description of the infrastructure and population of a region under study. Depending on the size of the settlement area, developing such models can be a costly and time-consuming task. In this paper we use a manually annotated dataset consisting of approximately 10,000 photos acquired at street level in the urban area of Medellín to explore the potential for using a convolutional neural network (CNN) to automatically detect building materials and types of lateral-load resisting systems, which are attributes that define a building's structural typology (which is a key issue in exposure models for seismic risk assessment). The results of the developed model achieved a precision of 93% and a recall of 95% when identifying nonductile buildings, which are the buildings most likely to be damaged in an earthquake. Identifying fine-grained material typology is more difficult, because many visual clues are physically hidden, but our model matches expert level performances, achieving a recall of 85% and accuracy scores ranging from 60% to 82% on the three most common building typologies, which account for 91% of the total building population in Medellín. Overall, this study shows that a CNN can make a substantial contribution to developing cost-effective exposure models. © 2020 Elsevier Ltd
Automatic detection of building typology using deep learning methods on street level images
(PERGAMON-ELSEVIER SCIENCE LTD, 2020-03-20) Duque, J.; Gonzalez, D.; Rueda Plata, Diego; Acevedo, A.; Ramos, R.; Betancourt, A.; García, S.; Mecánica Aplicada
An exposure model is a key component for assessing potential human and economic losses from natural disasters. An exposure model consists of a spatially disaggregated description of the infrastructure and population of a region under study. Depending on the size of the settlement area, developing such models can be a costly and time-consuming task. In this paper we use a manually annotated dataset consisting of approximately 10,000 photos acquired at street level in the urban area of Medellín to explore the potential for using a convolutional neural network (CNN) to automatically detect building materials and types of lateral-load resisting systems, which are attributes that define a building's structural typology (which is a key issue in exposure models for seismic risk assessment). The results of the developed model achieved a precision of 93% and a recall of 95% when identifying nonductile buildings, which are the buildings most likely to be damaged in an earthquake. Identifying fine-grained material typology is more difficult, because many visual clues are physically hidden, but our model matches expert level performances, achieving a recall of 85% and accuracy scores ranging from 60% to 82% on the three most common building typologies, which account for 91% of the total building population in Medellín. Overall, this study shows that a CNN can make a substantial contribution to developing cost-effective exposure models. © 2020 Elsevier Ltd
Automatic detection of building typology using deep learning methods on street level images
(PERGAMON-ELSEVIER SCIENCE LTD, 2020-03-20) Duque, J.; Gonzalez, D.; Rueda Plata, Diego; Acevedo, A.; Ramos, R.; Betancourt, A.; García, S.; Universidad EAFIT. Departamento de Ingeniería de Producción; Materiales de Ingeniería
An exposure model is a key component for assessing potential human and economic losses from natural disasters. An exposure model consists of a spatially disaggregated description of the infrastructure and population of a region under study. Depending on the size of the settlement area, developing such models can be a costly and time-consuming task. In this paper we use a manually annotated dataset consisting of approximately 10,000 photos acquired at street level in the urban area of Medellín to explore the potential for using a convolutional neural network (CNN) to automatically detect building materials and types of lateral-load resisting systems, which are attributes that define a building's structural typology (which is a key issue in exposure models for seismic risk assessment). The results of the developed model achieved a precision of 93% and a recall of 95% when identifying nonductile buildings, which are the buildings most likely to be damaged in an earthquake. Identifying fine-grained material typology is more difficult, because many visual clues are physically hidden, but our model matches expert level performances, achieving a recall of 85% and accuracy scores ranging from 60% to 82% on the three most common building typologies, which account for 91% of the total building population in Medellín. Overall, this study shows that a CNN can make a substantial contribution to developing cost-effective exposure models. © 2020 Elsevier Ltd
Building monitoring during the partial implosion stage
(Universidad EAFIT, 2015) Garcés Lengua, Julio; Botero, J.C.; Muriá Vila, David
The partial collapse of a building in Colombia caused severe damage to its structural components -- An implosion was realized to induce the collapse of 50% of the deteriorated building -- To evaluate the influence of the implosion on the remaining structure, a monitoring survey was realized using triaxial accelerometers -- Time signals associated with ambient, seismic and forced vibration were obtained -- A study of the records in the time and the frequency domain was made -- The analysis of the information allowed determining some structural properties that were useful to calibrate the analytical model of the structure
Development of structural debris flow fragility curves (debris flow buildings resistance) using momentum flux rate as a hazard parameter
(Elsevier B.V., 2018-05-18) Prieto, Jorge Alonso; Journeay, Murray; Acevedo A.B.; Arbelaez, Juan; Ulmi, Malaika; Mecánica Aplicada
Societal risks associated with debris flow hazards are significant and likely to escalate due to global population growth trends and the compounding effects of climate change. Quantitative risk assessment methods (QRA) provide a means of anticipating the likely impacts and consequences of settlement in areas susceptible to landslide activity and are increasingly being used to inform land use decisions that seek to increase disaster resilience through mitigation and/or adaptation. Current QRA methods for debris flow hazards are based primarily on empirical vulnerability functions that relate hazard intensity (depth, velocity, etc.) to expected levels of loss for a given asset of concern, i.e. most of current methods are dedicated to loss-intensity relations. Though grounded in observed cause-effect relationships, empirical vulnerability functions are not designed to predict the capacity of a building to withstand the physical impacts of a debris flow event, or the related uncertainties associated with modelling building performance as a function of variable debris flow parameters. This paper describes a methodology for developing functions that relate hazard intensity to probability of structural damage, i.e., fragility functions, rather than vulnerability functions, based on the combined hydrodynamic forces of a debris flow event (hazard level) and the inherent structural resistance of building typologies that are common in rural mountainous settings (building performance). Hazard level includes a hydrodynamic force variable (FDF), which accounts for the combined effects of debris flow depth and velocity, i.e. momentum flux (hv2), material density (?) and related flow characteristics including drag (Cd) and impact coefficient (Kd). Building performance is measured in terms of yield strength (Ay), ultimate lateral capacity (AU) and weight to breadth ratios (W/B) defined for a portfolio building types that are common in mountain settlements. Collectively, these model parameters are combined using probabilistic methods to produce building-specific fragility functions that describe the probability of reaching or exceeding successive thresholds of structural damage over a range of hazard intensity values, expressed in terms of momentum flux. Validation of the proposed fragility model is based on a comparison between model outputs and observed cause-effect relationships for recent debris flow events in South Korea and in Colombia. Debris flow impact momentum fluxes, capable of resulting in complete damage to unreinforced masonry buildings (URM) in those regions are estimated to be on the order of 24 m3/s2, consistent with field-based observations. Results of our study offer additional capabilities for assessing risks associated with urban growth and development in areas exposed to debris flow hazards. © 2018 Elsevier B.V.
Development of structural debris flow fragility curves (debris flow buildings resistance) using momentum flux rate as a hazard parameter
(Elsevier B.V., 2018-05-18) Prieto, Jorge Alonso; Journeay, Murray; Acevedo A.B.; Arbelaez, Juan; Ulmi, Malaika; Universidad EAFIT. Departamento de Ingeniería de Producción; Materiales de Ingeniería
Societal risks associated with debris flow hazards are significant and likely to escalate due to global population growth trends and the compounding effects of climate change. Quantitative risk assessment methods (QRA) provide a means of anticipating the likely impacts and consequences of settlement in areas susceptible to landslide activity and are increasingly being used to inform land use decisions that seek to increase disaster resilience through mitigation and/or adaptation. Current QRA methods for debris flow hazards are based primarily on empirical vulnerability functions that relate hazard intensity (depth, velocity, etc.) to expected levels of loss for a given asset of concern, i.e. most of current methods are dedicated to loss-intensity relations. Though grounded in observed cause-effect relationships, empirical vulnerability functions are not designed to predict the capacity of a building to withstand the physical impacts of a debris flow event, or the related uncertainties associated with modelling building performance as a function of variable debris flow parameters. This paper describes a methodology for developing functions that relate hazard intensity to probability of structural damage, i.e., fragility functions, rather than vulnerability functions, based on the combined hydrodynamic forces of a debris flow event (hazard level) and the inherent structural resistance of building typologies that are common in rural mountainous settings (building performance). Hazard level includes a hydrodynamic force variable (FDF), which accounts for the combined effects of debris flow depth and velocity, i.e. momentum flux (hv2), material density (?) and related flow characteristics including drag (Cd) and impact coefficient (Kd). Building performance is measured in terms of yield strength (Ay), ultimate lateral capacity (AU) and weight to breadth ratios (W/B) defined for a portfolio building types that are common in mountain settlements. Collectively, these model parameters are combined using probabilistic methods to produce building-specific fragility functions that describe the probability of reaching or exceeding successive thresholds of structural damage over a range of hazard intensity values, expressed in terms of momentum flux. Validation of the proposed fragility model is based on a comparison between model outputs and observed cause-effect relationships for recent debris flow events in South Korea and in Colombia. Debris flow impact momentum fluxes, capable of resulting in complete damage to unreinforced masonry buildings (URM) in those regions are estimated to be on the order of 24 m3/s2, consistent with field-based observations. Results of our study offer additional capabilities for assessing risks associated with urban growth and development in areas exposed to debris flow hazards. © 2018 Elsevier B.V.
Diseño de un sistema de indicadores de sostenibilidad como herramienta en la toma de decisiones para la gestión de proyectos de infraestructura en Colombia
(Universidad EAFIT, 2013) Gaviria Gaviria, Paula Andrea; Botero Botero, Luis Fernando
Desde el uso del concepto de desarrollo sostenible y su aplicación en las diversas áreas de la planeación urbana y a los proyectos de construcción en general, un contado número de indicadores de sostenibilidad han aparecido. Este trabajo analiza la necesidad de establecer un sistema formado por un grupo de indicadores que incluya a todos los partícipes involucrados en el ciclo de vida de un proyecto de Infraestructura, para hallar criterios de registro y evaluación de la sostenibilidad. Por lo tanto, se identifica, clasifica y prioriza las diferentes variables de la sostenibilidad que conforman un instrumento de gran utilidad dentro de los mecanismo de toma de decisiones o como herramienta de gestión integral de la obra civil, aplicado a las áreas ambiental, social, económico, institucional y tecnología/innovación. La aplicación de este método para proyectos de infraestructura en Colombia es el primer paso para controlar los procesos y su evaluación en diferentes ámbitos de la sostenibilidad de una obra de infraestructura en vía de mejorar el sector de la construcción y el medio ambiente construido.
Mejoramiento de los procesos de planificación de obras a partir de la introducción de conceptos de gestión logística soportados en TIC, para el sector de la construcción en Colombia
(Universidad EAFIT, 2011) Fonseca Arias, Cristian Guillermo; Botero Botero, Luis Fernando; Colciencias
Some of the concepts of logistics management is needed in the construction industry in Colombia with the intention to strengthen the processes involved in production and reduce waste due to lack of logistical planning, which are evident in excessive waiting times for provisioning materials on site, poor scheduling of materials, little or no tracking materials, inofficious stock accumulation and general waste of resources by inefficient control of logistics flows. The present study will address the tools and concepts of logistics management needed to deal the construction industry under criteria that are compatible with existing production models are tested and validated in other production sectors, Additionally, that information will be supplemented by adaptation TIC a tool to the local context, which will support the changes introduced with the new criteria for logistics planning of the works.
¿Quién responde por los daños causados con la construcción de edificios que se arruinan?
(Universidad EAFIT, 2019) Posada Arango, Mateo; Aramburo Calle, Maximiliano Alberto
In Colombia, civil liability within construction can be contractual (breach of contract) or non-contractual (tort) and may arise from complications during the construction, as well as damages caused ten years after completion of the works, due to its ruin, ruin threat or breach of contractual arrangements. Nonetheless, Colombian regulations do not clearly identify who is responsible for the damages caused by construction work. This paper makes an analysis of the role of each one of the individuals that intervene in the construction work, in order to determine if they are liable for the damages caused within the works, and therefore, if they have the obligation to repair said damages.

Examinando por Materia "Buildings"

Resultados por página

Opciones de ordenación