Kim, Minju; Lee, Dongmin; Kim, Taehoon; Oh, Sangmin; Cho, Hunhee. (2023). Automated Extraction of Geometric Primitives with Solid Lines from Unstructured Point Clouds for Creating Digital Buildings Models. Automation In Construction, 145.
View Publication
Abstract
Point clouds produced by laser scanners are an invaluable source of data for reconstructing multi-dimensional digital models that reflect the as-is conditions of built facilities. However, previous studies aimed to reconstruct models by overlaying the dataset on top of ground-truth reference models to manually adjust the accuracy of the output. Therefore, this paper describes the extraction of geometric primitives with solid lines—the simplest form of objectified data that computer-aided design systems can handle—from unorganized data points and creation of digital models of built facilities in a form of floor plan. The geometric primitives are extracted from 3D points by hybridizing machine learning algorithms, which are mean-shift clustering, non-convex hull, and random sample and consensus (RANSAC). This paper provides a solution for creating a new form of as-built model with high accuracy and robustness from scratch without the involvement of ground-truth solutions or manual adjustments. © 2022 Elsevier B.V.
Keywords
Computer Aided Design; Geometry; Laser Applications; Learning Algorithms; Machine Learning; Scanning; As-build Model Creation; Build Facility; From-point-to-line; Geometric Primitives; Laserscanners; Model Creation; Outline Extractions; Point-clouds; Point-to-line; Solid Lines
Ruoniu (Vince) Wang is an Assistant Professor in the Runstad Department of Real Estate in the College of Built Environments at the University of Washington. He studies spatial justice and inclusive communities, including their impacts reflected in the built environment, human behaviors, and policy interventions. Vince joined the University of Washington after serving six years as the research manager and director in a national non-profit organization Grounded Solutions Network. He has designed and conducted a U.S. Census of inclusionary housing policies, a U.S. census of community land trusts, and a national performance evaluation of shared equity homeownership programs. His research expands to policy evaluation for the two largest federal assisted housing rental programs in the U.S.: the Low-Income Housing Tax Credit program and the Housing Choice Voucher program. Vince grounds his research with applied tools to democratize data for low-income communities.
Lingzi Wu is an Assistant Professor with the Department of Construction Management (CM) at the University of Washington (UW). Prior to joining UW in September 2022, Dr. Wu served as a postdoctoral fellow in the Department of Civil and Environmental Engineering at University of Alberta, where she received her MSc and PhD in Construction Engineering and Management in 2013 and 2020 respectively. Prior to her PhD, Dr. Wu worked in the industrial construction sector as a project coordinator with PCL Industrial Management from 2013 to 2017.
An interdisciplinary scholar focused on advancing digital transformation in construction, Dr. Wu’s current research interests include (1) integration of advanced data analytics and complex system modeling to enhance construction practices and (2) development of human-in-the-loop decision support systems to improve construction performance (e.g., sustainability and safety). Dr. Wu has published 10 papers in top journals and conference proceedings, including the Journal of Construction Engineering and Management, Journal of Computing in Civil Engineering, and Automation in Construction. Her research and academic excellence has received notable recognition, including a “Best Paper Award” at the 17th International Conference on Modeling and Applied Simulation, and the outstanding reviewer award from the Journal of Construction Engineering and Management.
As an educator and mentor, Dr. Wu aims to create an inclusive, innovative, and interactive learning environment where students develop personal, technical, and transferable skills to grow today, tomorrow, and into the future.
Celina Balderas Guzmán, PhD, is Assistant Professor in the Department of Landscape Architecture. Dr. Balderas’ research spans environmental planning, design, and science and focuses on climate adaptation to sea level rise on the coast and urban stormwater inland. On the coast, her work demonstrates specific ways that the climate adaptation actions of humans and adaptation of ecosystems are interdependent. Her work explores how these interdependencies can be maladaptive by shifting vulnerabilities to other humans or non-humans, or synergistic. Using ecological modeling, she has explored these interdependencies focusing on coastal wetlands as nature-based solutions. Her work informs cross-sectoral adaptation planning at a regional scale.
Inland, Dr. Balderas studies urban stormwater through a social-ecological lens. Using data science and case studies, her work investigates the relationship between stormwater pollution and the social, urban form, and land cover characteristics of watersheds. In past research, she developed new typologies of stormwater wetlands based on lab testing in collaboration with environmental engineers. The designs closely integrated hydraulic performance, ecological potential, and recreational opportunities into one form.
Her research has been funded by major institutions such as the National Science Foundation, National Socio-Environmental Synthesis Center, UC Berkeley, and the MIT Abdul Latif Jameel Water and Food Systems Lab. She has a PhD in the Department of Landscape Architecture and Environmental Planning from the University of California, Berkeley. Previously, she obtained masters degrees in urban planning and urban design, as well as an undergraduate degree in architecture all from MIT.
Hong, Jinhyun; Shen, Qing; Zhang, Lei. (2014). How Do Built-Environment Factors Affect Travel Behavior? A Spatial Analysis at Different Geographic Scales. Transportation, 41(3), 419 – 440.
View Publication
Abstract
Much of the literature shows that a compact city with well-mixed land use tends to produce lower vehicle miles traveled (VMT), and consequently lower energy consumption and less emissions. However, a significant portion of the literature indicates that the built environment only generates some minor-if any-influence on travel behavior. Through the literature review, we identify four major methodological problems that may have resulted in these conflicting conclusions: self-selection, spatial autocorrelation, inter-trip dependency, and geographic scale. Various approaches have been developed to resolve each of these issues separately, but few efforts have been made to reexamine the built environment-travel behavior relationship by considering these methodological issues simultaneously. The objective of this paper is twofold: (1) to better understand the existing methodological gaps, and (2) to reexamine the effects of built-environment factors on transportation by employing a framework that incorporates recently developed methodological approaches. Using the Seattle metropolitan region as our study area, the 2006 Household Activity Survey and the 2005 parcel and building data are used in our analysis. The research employs Bayesian hierarchical models with built-environment factors measured at different geographic scales. Spatial random effects based on a conditional autoregressive specification are incorporated in the hierarchical model framework to account for spatial contiguity among Traffic Analysis Zones. Our findings indicate that land use factors have highly significant effects on VMT even after controlling for travel attitude and spatial autocorrelation. In addition, our analyses suggest that some of these effects may translate into different empirical results depending on geographic scales and tour types.
Keywords
Land-use; Urban Form; Multilevel Models; Physical-activity; Neighborhood; Choice; Impact; Specification; Accessibility; Causation; Built Environment; Travel Behavior; Self-selection; Spatial Autocorrelation; Bayesian Hierarchical Model
Liang, Huakang; Lin, Ken-yu; Zhang, Shoujian; Su, Yikun. (2018). The Impact Of Coworkers’ Safety Violations On An Individual Worker: A Social Contagion Effect Within The Construction Crew. International Journal Of Environmental Research And Public Health, 15(4).
View Publication
Abstract
This research developed and tested a model of the social contagion effect of coworkers' safety violations on individual workers within construction crews. Both situational and routine safety violations were considered in this model. Empirical data were collected from 345 construction workers in China using a detailed questionnaire. The results showed that both types of safety violations made by coworkers were significantly related to individuals' perceived social support and production pressure. Individuals' attitudinal ambivalence toward safety compliance mediated the relationships between perceived social support and production pressure and both types of individuals' safety violations. However, safety motivation only mediated the effects of perceived social support and production pressure on individuals' situational safety violations. Further, this research supported the differences between situational and routine safety violations. Specifically, we found that individuals were more likely to imitate coworkers' routine safety violations than their situational safety violations. Coworkers' situational safety violations had an indirect effect on individuals' situational safety violations mainly through perceived social support and safety motivation. By contrast, coworkers' routine safety violations had an indirect effect on individuals' routine safety violations mainly through perceived production pressure and attitudinal ambivalence. Finally, the theoretical and practical implications, research limitations, and future directions were discussed.
Keywords
Health-care Settings; Job Demands; Attitudinal Ambivalence; Industry Development; Workplace Safety; Behavior; Climate; Model; Risk; Employee; Social Contagion; Situational Safety Violations; Routine Safety Violations; Social Learning; Social Information Processing
Kim, Taehoon; Kim, Yong-woo; Lee, Dongmin; Kim, Minju. (2022). Reinforcement Learning Approach To Scheduling Of Precast Concrete Production. Journal Of Cleaner Production, 336.
View Publication
Abstract
The production scheduling of precast concrete (PC) is essential for successfully completing PC construction projects. The dispatching rules, widely used in practice, have the limitation that the best rule differs according to the shop conditions. In addition, mathematical programming and the metaheuristic approach, which would improve performance, entail more computational time with increasing problem size, let alone its models being revised as the problem size changes. This study proposes a PC production scheduling model based on a reinforcement learning approach, which has the advantages of a general capacity to solve various problem conditions with fast computation time and good performance in real-time. The experimental study shows that the proposed model outperformed other methods by 4-12% of the total tardiness and showed an average winning rate of 77.0%. The proposed model could contribute to the successful completion of off-site construction projects by supporting the stable progress of PC construction.
Keywords
Precast Concrete; Reinforcement Learning; Deep Q -network; Production Scheduling; Minimize; Model
Migliaccio, G. C.; Bogus, Susan M.; Cordova-Alvidrez, A. A. (2014). Continuous Quality Improvement Techniques for Data Collection in Asset Management Systems. Journal Of Construction Engineering And Management, 140(4).
View Publication
Abstract
Transportation infrastructure assets are among the largest investments made by governmental agencies. These agencies use data on asset conditions to make decisions regarding the timing of maintenance activities, the type of treatment, and the resources to employ. To collect and record these data, agencies often utilize trained evaluators who assess the asset either on site or by analyzing photos and/or videos. These visual assessments are widely used to evaluate conditions of various assets, including pavement surface distresses. This paper describes a Data Quality Assessment & Improvement Framework (DQAIF) to measure and improve the performance of multiple evaluators of pavement distresses by controlling for subjective judgment by the individual evaluators. The DQAIF is based on a continuous quality improvement cyclic process that is based on the following main components: (1)assessment of the consistency over timeperformed using linear regression analysis; (2)assessment of the agreement between evaluatorsperformed using inter-rater agreement analysis; and (3)implementation of management practices to improve the results shown by the assessments. A large and comprehensive case study was employed to describe, refine, and validate the framework. When the DQAIF is applied to pavement distress data collected on site by different evaluators, the results show that it is an effective method for quickly identifying and solving data collection issues. The benefit of this framework is that the analyses employed produce performance measures during the data collection process, thus minimizing the risk of subjectivity and suggesting timely corrective actions. The DQAIF can be used as part of an asset management program, or in any engineering program in which the data collected are subjected to the judgment of the individuals performing the evaluation. The process could also be adapted for assessing performance of automated distress data acquisition systems.
Keywords
Asset Management; Civil Engineering Computing; Data Acquisition; Decision Making; Inspection; Maintenance Engineering; Quality Control; Regression Analysis; Roads; Transportation; Continuous Quality Improvement Techniques; Asset Management System; Governmental Agencies; Transportation Infrastructure Assets; Maintenance Activities; Visual Assessment; Pavement Surface Distresses; Data Quality Assessment & Improvement Framework; Dqaif; Linear Regression Analysis; Interrater Agreement Analysis; Data Collection Process; Automated Distress Data Acquisition System; Manual Pavement Distress; Pavement Management; Quantitative Analysis; Data Collection; Assets; Reliability; Case Studies
Stewart, Orion T.; Moudon, Anne Vernez; Littman, Alyson J.; Seto, Edmund; Saelens, Brian E. (2018). The Association between Park Facilities and Duration of Physical Activity During Active Park Visits. Journal Of Urban Health, 95(6), 869 – 880.
View Publication
Abstract
Public parks provide places for urban residents to obtain physical activity (PA), which is associated with numerous health benefits. Adding facilities to existing parks could be a cost-effective approach to increase the duration of PA that occurs during park visits. Using objectively measured PA and comprehensively measured park visit data among an urban community-dwelling sample of adults, we tested the association between the variety of park facilities that directly support PA and the duration of PA during park visits where any PA occurred. Cross-classified multilevel models were used to account for the clustering of park visits (n=1553) within individuals (n=372) and parks (n=233). Each additional different PA facility at a park was independently associated with a 6.8% longer duration of PA bouts that included light-intensity activity, and an 8.7% longer duration of moderate to vigorous PA time. Findings from this study are consistent with the hypothesis that more PA facilities increase the amount of PA that visitors obtain while already active at a park.
Keywords
Park Facilities; Physical Activity; Park Use; Recreation; Built Environment; Global Positioning System; Accelerometer; Gis; Gps; Accelerometer Data; United-states; Adults; Proximity; Features; Walking; Size; Attractiveness; Improvements; Environment; Parks & Recreation Areas; Parks; Luminous Intensity; Clustering; Urban Areas
Migliaccio, G. C.; Bogus, Susan M.; Cordova-Alvidrez, A. A. (2014). Techniques for Continuous Improvement of Quality of Data Collection in Systems of Capital Infrastructure Management. Journal Of Construction Engineering And Management, 140(4).
View Publication
Abstract
oLa infraestructura del transporte es una de las mas grandes inversiones que realizan los gobiernos. Las agencias gubernamentales de transporte administran este capital y utilizan la informacion de las condiciones de este para decidir la programacion y tipo de mantenimiento y recursos a ejercer. Para recolectar la informacion pertinente, las agencias emplean evaluadores adiestrados para evaluar la infraestructura, ya sea en sitio o analizando fotografias y/o videos. Las evaluaciones visuales son empleadas para inspeccionar las condiciones de la infraestructura, incluyendo el desgaste de la superficie de los caminos y carreteras. Este articulo describe un Data Quality Assessment & Improvement Framework (DQAIF) (Sistema de Evaluacion y Mejora de la Calidad de la Informacion) para medir y controlar los datos de los evaluadores del deterioro de carreteras, al controlar el criterio de estos. El DQAIF es en un proceso ciclico de Mejora Continua de Calidad compuesto por: a)la evaluacion del nivel de acuerdo entre evaluadores -por medio del analisis estadistico (inter-rater agreement analysis), b)la evaluacion de la consistencia a traves del tiempo -mediante analisis de regresion lineal, y c)la implementacion de practicas gerenciales para mejorar los resultados mostrados en las evaluaciones anteriores. Se llevo a cabo un estudio de caso para validar el sistema propuesto. Los resultados mostraron que el DQAIF es efectivo para identificar y resolver problemas de la calidad de los datos obtenidos en las inspecciones de infraestructura. Con este sistema se garantiza la reduccion del riesgo de la subjetividad y asi aplicar acciones de mantenimiento mas oportunas. El DQAIF puede ser empleado en un programa de gerencia de infraestructura o en cualquier programa de ingenieria en donde la informacion esta sujeta al juicio o criterio personal de los individuos que realizan la evaluacion. Este proceso puede ser adaptado, incluso, para evaluar el desempeno de sistemas automatizados de evaluacion de pavimentos.
Keywords
Manual Pavement Distress; Quality Control; Pavement Management; Inspection; Quantitative Analysis; Data Collection; Assets; Reliability; Construction Materials And Methods