Computing in Civil Engineering 2017: Sensing, Simulation, and Visualization

Virtual reality (VR) has attracted increasing attention of the architecture, engineering, construction and facility management (AEC/FM) industry in recent years. VR has the potential to improve workflow efficiency through enhanced common understanding. However, the current workflow, which usually contains manually converting design files to VR application, is quite cumbersome and usually, inefficient. The lack of automated and efficient data exchange approach and protocol could potentially cause designers and other users to abandon VR after the first few design cycles. In this paper, we introduce a BIM-VR real-time data exchange system. The system is based on an innovative Cloud-based BIM metadata interpretation and communication method. With the function of allowing users to update BIM model changes in VR headsets automatically and simultaneously, the system achieves high user stickiness and reduces the complexity of workflow. We tested the system in a variety of design change scenarios including changing object dimensions and changing object types. Results confirmed the usability and efficiency.

With the future becoming more technologically advanced with the modeling of all types of information across varying domains, it is imperative that the information contained in the models can be shared and exchanged seamlessly and effortlessly by the end user. The purpose of this research is to develop a methodology that would enable the interoperability of multi-disciplinary information models. Current methods of producing information models have been mostly dependent on proprietary software programs, non-compatible program languages, or standards specific to only a single domain. Therefore, the objective is to have a methodology that applies across the heterogeneous landscape of information modeling. Although the approach is extendable to other domains, the scope for this paper is for bridge information modeling (BrIM) and the architecture, engineering, construction, and operations (AECO) industry. This research is motivated by the fundamental issues of interoperability, such as semantics, logic, and software issues. The use of an ontology, the formal naming and definition of the types, properties, and interrelationships of the entities in a domain, has been shown to facilitate the sharing of knowledge. This paper builds on the assertion that a well-established taxonomy, a formal hierarchical classification of terms, is the imperative first step in defining an ontology. Therefore, the focuses of this paper is a discussion of defining and organizing a taxonomy. An ongoing case study with a joint industry subcommittee of the American Association of State Highway and Transportation Officials (AASHTO) and the National Steel Bridge Alliance (NSBA) is used to evaluate and validate the methodology. Preliminary results of taxonomy being developed by the case study is presented to show the importance of defining the semantic and syntactic information in a natural language to promote interoperability. By reducing semantic issues and maintaining consistency of terminology at the software level, reliable, automatic, and standardized electronic exchange of data can be ultimately achieved.

Every structural design incorporates biases from the professional experiences of engineers involved in the design process, resulting in choices that do not necessarily lead to optimal designs. And, while the design codes provide a methodology for safe design, optimal design is not explicitly considered. In response, artificial intelligence and advanced computation techniques are emerging avenues for guiding engineers towards optimal designs. This research explores this concept through a hybridization of topology optimization (TO) and evolutionary algorithms (EA) for design optimization. TO and EA have both been studied for use in structural optimization, however each of these techniques is inadequate individually. Since TO generates monolithic components, the designs need to be heavily post-processed to generate practical structural assemblies. The challenge of using EA is that the performance of the algorithm is dependent on a set of specified initial conditions. This paper presents a new technique for transforming TO generated designs into structural assemblies suitable for EA. A combination of clustering and mathematical morphology provides the basis for this algorithm. Presented herein is a demonstration and analysis of the efficacy of the proposed algorithm using three benchmark trusses. The results highlight the potential of this algorithm, and future work looks to improve the consistency of the algorithm across a diverse set of inputs.

Educational simulation games and serious games in various disciplines are being adopted for their ability to engage and educate students. The purpose of this study was to understand the level on enjoyment while playing a construction simulation game, the Virtual Construction Simulator 4. For this study, the researchers were interested in the users’ rating of their experience while developing the construction sequence for the three different modules. The game was implemented in a fourth-year construction engineering course, where 34 students participated in the study. The users were given a survey to rate their experience while playing the game at the end of a series of interventions. Based on the results there is evidence to conclude that the students’ experience was significantly higher that the neutral level of 3.5. With the presented results, the researchers have provided supporting evidence of the engaging nature of educational simulation games, as well as invaluable suggestions for future development. Lastly, this paper discusses the difficulty in striking a balance between the engaging “fun” nature of games and their educational “serious” purpose.

The ultimate goal of this research endeavor is to enhance student learning of the advanced decision-making techniques and technologies in the architectural, engineering, and construction (AEC) disciplines. These subjects include, but are not limited to, simulation, statistical modeling, building information modeling, virtual construction, real-time sensing, and vision-based recognition. Aiming to meet the goal, the present study proposes a collaborative learning model designed to increase the experiential and collaborative practice of learning. This paper presents a survey on the current status of education of the subjects, the proposed model, and a case study. The results of the case study indicate that the model can help students overcome three significant limitations of the current learning approaches: fragmented learning, insufficient exposure, and lack of experiential learning.

Cities cover three percent of the world’s land, yet are responsible for up to eighty percent of global greenhouse gas emissions and seventy-five percent of natural resource consumption. They impact air and water quality, alter resource consumption patterns, and pose unique challenges for the environment, energy, and infrastructure. Addressing these challenges requires understanding the city as the nexus of environmental context, built infrastructure, and human communities, which can only be achieved by close collaboration among educated professionals from a wide range of related domains. This paper describes the experiences of a university course that was developed to prepare students from diverse academic backgrounds to understand the nature of and learn the required skills to address sustainable urbanization challenges. The course, which was partially online enabled by the latest information and communications technology (ICT), was jointly offered by three academic departments at Tsinghua University and Stanford University. The course used Beijing and the Bay Area as two case studies, which offered students a unique cross-cultural and cross-disciplinary educational experience. Within the context of this course, students formed multinational teams, collaborated remotely, and explored sustainability topics such as air quality, urban mobility and transport, and cultural heritage buildings. The teams developed creative solutions to their chosen topics, presented them in the Human Cities Expo at Stanford University, and tested their validity in the case cities in consultation with local experts, scholars, and non-government organizations community partners. Lessons learned during this teaching experience, in terms of the role of ICT infrastructure in supporting online teaching, the management of cross-disciplinary in-class collaboration, and the value of conducting cross-cultural comparative studies, are discussed at the end of the paper.

To help undergraduate students visualize stresses, deformation, and limit states for each steel connection type we have designed and developed an augmented reality (AR) interactive learning environment in which holographic images illustrating the connection behavior are overlaid on physical models of the 48 steel connection types. The physical models of the 48 connection types were produced by 3D printing a previously developed virtual model of the steel sculpture. Students apply forces to the connections, and observe the stresses in the connection, the deformations such as deflections, rotation, and curvatures, and the limit states–all without a mouse, joystick, smartphone or tablet device. Findings from an initial study with undergraduate students support the usability and usefulness of the AR environment.

Massive open online courses (MOOCs) have become more and more popular due to the advancement of technology that has brought up a convenient and flexible learning modal for various learners around the world. Based on our prior research of integrating a MOOC into a conventional course in 2015, the majority of students improved grades especially the ones who completed MOOCs course items on schedule. However, specific teaching and learning mechanism are still undiscovered. This year we further purpose a learning-enhanced model, named guided soft classroom, to provide a monitoring mechanism based on the analysis of progress data from MOOCs (Coursera) to enhance students self-regulated learning strategies. The monitor system is developed to identify students who need “guidance” throughout the course of the semester. Students behind expected weekly progress are asked to attend a guiding session provided by the instructor. According to the discussions and observations in the guided sessions, guided soft classroom delivers a flexible and minor pressured learning model for various learning patterns and individual situations. We expect students in guided soft classroom will perform better; moreover, we also discover other impact factors of self-regulated learning. This paper presents an ongoing research, validated questionnaires and focus group interview will be conducted and analyzed along with MOOCs data for this course to further demonstrate students learning outcomes, which will be presented in the time of conference.

As a technique for practice and learning, virtual simulation can be applied to many disciplines to replace and intensify the real-world experiences with guided ones. In this study, the researchers applied the technique to a material testing course, which used to be taught in a traditional engineering laboratory but is now being converted to a hybrid course by incorporating virtual labs. Ideally engineering laboratories are intended to deepen students’ learning of engineering concepts through hands-on practices. However, due to a variety of resource, safety, and support constraints, some students are not fully engaged and perform poorly. Virtual labs provide a risk-free environment for students to explore theoretical concepts with a wider selection of materials and specimen sizes. The duration of experiments can be shortened intentionally to allow more focus on the testing procedure and concepts. In this study, a web application was developed using flash, Node.js, HTML, and AngularJS technologies to provide virtual simulations for experiments such as tension/compression testing, torsion testing, shear testing, flexural testing, etc. Instead of generating theoretical results via built-in formulas, the researchers decided to use real data collected from actual lab experiments. Students thus would have a more realistic perception of field data and practice various techniques on data processing. The web application will be hosted on a full-fledged server. This paper will detail the course development and discuss its future potential applications.

The efficient utilization of construction equipment plays a significant role in engineering and building operations, since the utilization information is instrumental for construction equipment management and project cost control. Therefore, it is necessary to monitor the utilization rate of onsite equipment in order to maintain the efficiency of construction operations. In current practices, the monitoring of the utilization rate of onsite construction equipment mainly relies on the manual observations by field engineers, which is time-consuming and labor-intensive. The main objective of this paper is to present a vision-based method to automate the monitoring of the utilization rate of onsite construction equipment. Under this method, the equipment of interest is firstly localized and tracked in video frames. Also, the positions of the equipment in video frames are extracted. Then, the location information of equipment is analyzed in comparison with the location information of work zones at the jobsite. Such way, the utilization rate of construction equipment in each work zone can be measured. Accordingly, construction professionals can conduct the further analysis, in terms of productivity analysis and equipment rental expense calculation. The method has been tested on the construction site of Poste De Lorimier in Canada. The test results demonstrated the capability and effectiveness of the method to monitor the utilization rate of construction equipment in an automatic manner.

Building HVAC system is essentially set at nominal levels according to industry guidelines. Several studies have demonstrated that this conventional practice is unable to meet the thermal requirements of occupants in a single or multi-occupancy space due to occupants’ diverse preferences, activities and needs. To improve occupants’ health and well-being, this study develops and tests a smartphone application framework which is capable of dynamically determining the optimum room conditioning mode (mechanical conditioning or natural ventilation) and HVAC settings (e.g., thermostat setpoint) in single and multi-occupancy spaces. The “personalized” control framework integrates building indoor environment data (obtained from sensors) with human physiological and behavioral data (obtained from wearable devices, polling apps, and surveys) to develop occupant comfort models under both mechanical and natural ventilation modes. In the operation phase, occupants’ personalized thermal sensations and preferences are continuously predicted using the comfort model when determining the optimum HVAC control strategies. A case study is conducted to demonstrate the capabilities of the developed application framework to improve personal comfort in single occupancy spaces.

Conventional structural defect detection methods, such as visual inspection, in-situ testing and Bayesian model updating, etc., have been widely applied in the condition assessments of bridge structures. However, spatial data collection and processing required by these methods are time-consuming and error-prone, and subject to human experiences. To address these challenges, the paper presents an automated bridge condition assessment method based on 3D laser scanning technique, which can rapidly detail the 3D geometry of bridge structures. With a computer-assisted automatic data analysis algorithm, the geometric variations of bridges can be real-timely traced and quantitatively digitized. Then a structural system identification method is proposed to automatically reveal the developing trend of the collapse mechanism behind these measured geometric variations. By doing this, a timely warning can be given of the possible structural failure. A case study on single-column-pier bridges was given by tracing the variations of bending, torsional and lateral deformations of girders, and correlated deformations of bearings to demonstrate the 3D laser scanning technique for diagnosis of the structural condition and prediction of failure modes. This study concluded that the proposed method is of great potential to identification, prediction, and assessment for the possible defects of existing bridges.

Construction workers’ valence, which is an important dimension of emotions by representing intrinsic attractiveness and aversiveness, significantly influences their awareness, attention, motivation, etc. Recently, a wearable electroencephalogram (EEG) device has opened a door to measure and understand construction workers’ valence levels at the workplace. However, acquiring high quality EEG signals is very difficult at the field due to signal artefacts prevalent in construction sites. In this regard, a signal processing framework was previously developed by the authors to remove the most common artefacts recorded in the EEG signals. In this paper, we further demonstrate how construction workers’ valence can be identified by applying this framework. Significant differences in the valence levels were captured while subjects were working in various real work conditions (e.g., working at ground level, top of the ladder, and in confined space). The results show the feasibility of using a wearable EEG device to monitor a worker’s valence.

The throw-and-roll method of temporary pothole patching, which is mostly accomplished manually, is to fill packing material into the depression on pavement. However, to determine the required filling volume to patch every individual pothole depends on the workers’ experience. The inappropriate patching may cause potential danger to the traffic. For the purpose of avoiding inappropriate pothole patching, we proposed a method to measure the volume of a pothole. Our method applies photometric stereo technique by using only a camera and couples of spotlights to achieve the purpose. Comparing to other non-contact 3D scanning technology using lasers or x-ray equipment, our method only requires a readily available equipment, and is relatively low-cost. By processing a set of a pothole’s top view images taken with exact same viewing angle but different illumination angles, photometric stereo technique can be used to generate the 3D model of the pothole. After model reconstruction, the volume of the pothole is accessible. Result showed that models of potholes can be rebuilt by proposed method; however, it needs further improvement in its accuracy of volume measurement. Still, we consider that based on the known volume of a pothole, the patching process can be more accurate and simpler. Furthermore, our method may enhance the possibility of developing an autonomous inspection and repair system as well as a cost estimation method for pothole patching activities.

Laser scanning has become increasingly popular throughout the architecture, engineering, construction and operations (AECO) industry and is becoming an asset for various project management processes. Traditional phase-shift or time of flight laser scanning devices have been the standard in the industry and set the bar for data accuracy. However, these devices can be costly and time consuming to operate, which limits their availability for many projects. To counteract this concern, there has been a surge in emerging technologies that use infrared, lidar, photogrammetry or a combination of those solutions. The workflows of these emerging scanning techniques are vastly different from that of traditional laser scanning, which is the primary advantage these solutions rely on. Emerging point cloud generating technologies show great promise for the AECO industry, but need to be effectively evaluated to achieve widespread acceptance. The goal of this research was to assess one emerging reality capture technology and compare it to traditional laser scanning. A mobile lidar based technology was evaluated and compared to traditional laser scanning based on a number of factors including; accuracy, operation time, processing time, workflow and quality. A field study was conducted using both devices and associated processes. The data collected were used to compare and evaluate their performance. This research provides a baseline understanding for the comparison of a mobile lidar scanning solution and traditional laser scanning techniques in the AECO industry.

Dynamic workspace modeling is important for providing real-time assistance including hazard detection, collision avoidance and 3D visualization for operators of heavy equipment such as cranes. This information helps the operators to rapidly perceive the surrounding environment and carry out manipulative tasks safely and efficiently. 2D image data can be collected quickly but lacks depth information; on the other hand, point cloud data captures complete 3D information but requires a long period of time to acquire and process the data. To address these shortcomings, this study proposes a hybrid point cloud-vision approach to track the location of mobile assets (e.g., machinery, workers) and to build a dynamic 3D model of the crane workspace. The initial site geometry is captured in a point cloud format using a laser scanner, where mobile assets are represented by 3D oriented bounding boxes. A wide-angle camera mounted on the crane boom continuously monitors the locations of moving objects. The acquired information from the camera and laser scanner is then fused to incrementally update the 3D location of object bounding boxes. The proposed approach is validated in a series of realistic crane lift scenarios in the field.

The importance of efficient construction equipment management can never be over emphasized. However, in current practices, construction equipment management is reportedly inefficient. It is largely based on rule-of-thumb rather than scientific evidence derived from solid data analytics. It faces challenges such as ‘small’, erratic data. Building on previous studies on smart construction objects (SCOs), this paper aims to develop a SCO-enabled proactive big data management system to facilitate the data collection, data visualization and data analysis for construction equipment management. It does so by first analyzing the problems of data management in prevailing construction equipment management practices. Then, the architecture of the proposed system is presented, which consists of two key components. The first one is a highly customizable smart chip that integrates various sensing and communication modules for proactively collecting and exchanging big data with volume, velocity, and verity inflowing from daily equipment operations. The second one is the data analytics platform for data storage, visualization and analytics. The pilot of the system has just been started in a central police station renovation project in Hong Kong. Fruitful datasets and analytics are expected from the pilot application of the proposed system. It is envisaged that the proposed system can supplement the existing construction equipment management with more comprehensive and concurrent decision-making information.

The construction industry is one of the most hazardous industries based on the number of fatalities. Mobile cranes, as a central component of many construction operations, contribute to a large number of construction fatalities. While many of these fatalities are caused by inadequate planning, few of the mobile crane safety issues have been well addressed. Developments in information technologies, notably cyber-physical systems (CPS), are changing the way in which mobile crane planning can be done. With the capacity for bi-directional coordination between physical world components and their virtual representations, CPS offers advantages in monitoring mobile cranes in such a way as to facilitate effective planning, proactively monitor crane operations, provide rich multi-modal feedback to crane operators, and ultimately, to avoid mobile crane failures. This paper explores CPS applicability in planning mobile crane operations on construction sites. It reviews the key definitions and characteristics of CPS, current CPS applications in the built environment, potential benefits and barriers for CPS application to mobile cranes, and discusses how CPS application to the planning and operation of mobile crane uses can be achieved.

This paper presents a new visual production management system for proactive project controls on construction sites. Taking advantage from the unprecedented growth of visual data on construction sites, the system continuously reconstructs and visualizes reality in form of 4D point clouds and maps it directly within 4D BIM. By putting schedule tasks and project performance data in a visual context for the entire team and mapping Reality to Plan, the system communicates “who does what work in what location”, tracks actual progress and productivity at the weekly work plan and the look ahead schedule, and infers top locations at risk for potential delays. To validate the system and its impact on field reporting and project controls practices, a case study is conducted on a building construction project. Results demonstrate that such a system improves coordination and communication by providing transparency in project execution and helps project teams improve reliability of their short term plans.

Intelligent sensing or monitoring has emerged as one the most attractive technologies in smart buildings. Recently, wireless sensor network is increasingly popular in smart building applications. One big remaining challenge is how to integrate multiple sensors conveniently under the system constraints of low-cost and small form-factor. For owners or users of smart buildings, installing wireless sensor networks for each sensing function is cost prohibitive. An alternative approach is to assemble multiple sensing functions (e.g., humidity, temperature, occupancy) into one node. This study presents a design of miniature and affordable wireless sensor node prototype, which is versatile to support five sensing functions (i.e., temperature, carbon dioxide, room occupancy, lighting and fire detection) via three sensors (i.e., a temperature sensor, a light sensor and a CO₂ sensor). Experimental tests have been performed to verify the functionality and benefits of this proposed prototype.

Automated activity analysis is vital for efficient productivity management on construction sites. A vision-based method has received attention due to its efficiency in identification and tracking. Although existing vision studies showed applicability, they did not consider interactive operations between equipment, and thus deteriorated analysis performance. To tackle the limitation, this paper proposes a vision-based activity analysis framework considering interactive operations of construction equipment. The framework includes four main processes: detection, tracking, individual action recognition, and interaction analysis. For the feasibility analysis, the individual actions of excavators were analyzed automatically using tracking-learning-detection and bags-of-features. This framework was validated with video images collected from earthmoving construction sites, and the average precisions of detection, tracking and action recognition were 88.0%, 88.0%, and 83.6% respectively. The experimental results showed the developed approach was able to identify independent actions of equipment, but the interactive operations should be considered to be more reliable activity analysis.

In efforts to automate construction performance monitoring, past studies have worked on vision-based registration of image to BIM and 3D point clouds to BIM. The continuous development of simultaneous localization and mapping (SLAM) enabled real-time estimation of locations and orientations of a camera while incrementally reconstructing a 3D scene. However, it localizes a camera to an arbitrary local coordinate system and produces a low-resolution and noisy point cloud that is not suitable for quality assessment of a structure. For the architecture/engineering/construction industry, the better and realistic approach is to localize with respect to building information models (BIMs) in real-time and post-process 3D dense reconstruction. This approach will allow project management teams to better communicate quality and progress using visuals associated with locations shown with BIMs. Moreover, it will automate images-to-BIM and image-based point clouds-to-BIM registration, enhancing past studies that attempt to automate image-based progress detection and quality assessment. On the other hand, the current state-of-the-art method for registering an image-based point cloud to a BIM requires selection of the correspondences. To address these challenges and achieve automation, this paper presents a new localization method that aligns an image to a BIM by detecting and matching perspectives of the image and the BIM. The results demonstrate the potential for enabling automated visual data collection (as-built aligned with as-planned) for performance monitoring.

Considering the contribution of building systems in the energy and electricity consumption globally and in the US, facilities management and operational strategies that help improve energy efficiency have been the subject of several studies. Electricity desegregation through non-intrusive methods is one of these strategies that provide cost-effective solutions for energy monitoring with higher spatiotemporal resolution (i.e., at appliance level). Since these techniques use machine learning for signal processing algorithms on aggregate data from one sensing point to infer contribution of each load, their application commonly calls for algorithm configuration (i.e., training) in a new environment. In this study, we have proposed a novel feature extraction approach for lighting loads through spectral analysis of the light intensity signal in order to automate the training of algorithms in the non-intrusive load monitoring systems. The proposed feature identifies the artificial light variation in presence of natural light and improves the rate of detecting lighting loads operations with minimum number of light intensity sensor.

Cyber-physical systems offer integrated sensing, computing and communication capabilities which drive innovation and competition in all industries including construction. Highway construction projects are major projects that, as a result of their sequential activity progress, are usually more vulnerable to schedule overruns. The schedules of highway projects need to be monitored in real-time to ensure timely implementation of project deliverables and efficient management of sequentially used resources. This paper presents a cyber-physical systems-based approach for improving progress monitoring and control of road and highway projects. This is achieved through establishing effective bi-directional interactions between project tracking tools and physical construction components on the worksites. A case study is presented to demonstrate the efficacy of the proposed application in efficient progress monitoring and control of highway projects.

Providing users’ location information in indoor environments has great importance for the management of the built environment in energy conservation efforts, facilitating navigation in large and complex places, providing context-aware information, and addressing safety concerns in emergency situations. Various technologies have been proposed to address the problem of indoor positioning among which is the emerging positioning techniques based on LED lights. LED lighting has the capability for visible light communication (VLC), which in turn could enable visible light positioning (VLP). Recent research has shown that VLP can achieve highly accurate results compared to systems based on RF waves. In this paper, the scalability of using an LED- based indoor positioning system for a typical office environment is studied. A trilateration positioning method is applied to localize the user. The effect of using different number of lights and specific hardware parameters that can affect the communication range of the system has been investigated. The findings were presented in terms of accuracy at different points and attainable area for localization. The findings show that the scalability and performance of the positioning system rely on the selection of these parameters, and highly accurate positioning can be achieved in an office setting through adjusting the configuration of lights and parameter selection.

Although the vast majority of construction accidents are caused by workers’ unsafe behavior, the management of safety behavior remains inadequate. To reduce unsafe behaviors, construction managers have mainly relied on external formal controls (e.g., penalties) to prevent a violation of safety rules. However, a number of studies recently demonstrated that social factors such as group norms and social identification play a critical role in shaping workers’ safety behavior. Although previous studies have illuminated the social aspect of safety behavior, there is a noticeable paucity of research addressing the mechanism that underlies the link between cognitive process, social influence, and safety behavior. To address this issue, this study aims to understand the effect of workers’ socio-cognitive processes on their safety behavior using experimental analyses with computer simulations. An agent-based model was developed to perform experimental analyses by incorporating the theoretical cognitive model of safety behavior and empirical finding regarding social influence. Then, model experiments were conducted to explore the effect of the interaction between the socio-cognitive process and different management interventions on workers’ safety behaviors. The results indicate that frequent managerial feedback has a positive but limited influence on workers’ safety behavior. In addition, it was also found that workers’ safety behavior can be further improved by combining the strict managers’ risk acceptance and stimulation of workers’ project-based identity. This study contributes to the body of knowledge on construction safety by modeling the socio-cognitive mechanism of safety behavior and exploring how the mechanism interacts with different management interventions.

Naturally ventilated building, which utilizes wind force and stack effect to circulate air of the building, has gained increasing interest among building owners due to the reduction in energy cost and the healthier indoor environment it could provide. Building simulation, as an important engineering analysis technique to evaluate the future building performance, has been frequently utilized to ensure the usability of naturally ventilated buildings during the design phase. Nonetheless, many of these fully natural ventilation buildings or zones still can’t meet the expected performance in real operation. One of reasons for these unexpected outcomes could be that design decisions are typically made based on only deterministic simulation without sufficiently taking into account the variability spring from the building’s local micro-climate, physical properties and actual operations. In this paper, we propose to use uncertainty analysis in building simulation in order to thoroughly assess the risks of thermal comfort for naturally ventilated buildings. Uncertainty sources for thermal comfort analysis through building simulation are identified first. Then an uncertainty analysis workbench is employed to conduct full uncertainty analysis for a prototype office building. These results are later compared with the deterministic simulation result to show the potential risks if the only deterministic result is considered in design decision making. After comparison, the most important uncertainties to be considered in building simulation are identified through sensitivity analysis.

Large scale and high-rise buildings are often built because of functional, economical, or aesthetic needs, especially in metropolises. As buildings get larger and taller, traffic volume and time for evacuation increase when incidents occur. To ensure the safety of occupants in those buildings, a proper evaluation of the occupants’ evacuation is often achieved through simulations. However, the defined input parameters of the currently available evacuation simulation systems rarely consider the occupant differences among countries or races, which may impair the accuracy of the evacuation simulation if the simulation tool is used on another ethnic group. The objective of this paper is to obtain basic parameters required for an accurate evacuation simulation that targets the Korean population through real-world experiments. In this study, the authors conducted a mock evacuation experiment with 60 people in a 50-story apartment building located in Daegu, Korea. The occupants’ change of walking speed and behavior that occurred in the evacuation staircases was analyzed through a CCTV that recorded time and passage on each staircase. Both the horizontal and vertical walking speeds at stairs were obtained. The simulation system was examined through comparing the actual evacuation time of the experiment with the evacuation time of simulation. The paper contributes to the overall body of knowledge by providing insights into the basic parameters that the simulation systems need to take into account to enable a better evaluation of evacuation time that matches the Korean population.

Unstructured construction sites including incomplete structures and unsecured resources (e.g., materials, equipment, and temporary facilities) are among the most vulnerable environments to windstorms such as hurricanes. Wind-induced cascading damages cause substantial losses, disruption, and considerable schedule delays in construction projects. Moreover, this would negatively affect neighboring buildings and interdependent infrastructures (e.g., electric power transmission or transportation systems), which triggers serious economic losses in our community. Nonetheless, prior works on disaster management mainly focused on post-disaster assessment and reconstruction process of built environments, and thus predicting potential risks associated with expected disasters for proactive preparedness remain largely unknown. This paper presents a new Imaging-to-Simulation framework that can uncover potential risks of wind-induced cascading damages to construction projects and their negative impacts on neighboring communities. The outcomes are expected to benefit our society as it will enhance current windstorm preparedness and mitigation plans, which ultimately promote public safety, property loss reduction, insurance cost reduction, and raise awareness of ‘Culture of Preparedness’ for disasters.

Network clearance time is the time required to clear a network under evacuation; it can be interpreted as the time for the last person to leave the network for the place of safety as well. The minimization of network clearance time is a commonly used objective in evacuation problems, as it is critical when the surge of demand in a short period overwhelms the existing network capacity. However, it lacks mathematical formulation in contrast to other objectives, such as the minimization of total evacuation time (the system optimal flow) and maximization of arrivals to safe areas within a given time period (the earliest arrival flow). Hence, this study develops a dynamic programming approach to estimate network clearance time. Given the network structure and the distribution of evacuation demand, an algorithm is proposed to solve the evacuation problem backward from a super sink. The evacuation problem is decomposed into sub-problems of capacity assignment that can be sequentially traced back from the super sink. Accordingly, time-dependent bottlenecks are identified in each sub-problem, and the evacuation flow of minimum clearance time that best utilizes network capacity against these bottlenecks can be determined. Other than determining the evacuation flow, the identification of bottlenecks also provides the insights for network enhancement in the contexts of both evacuation planning and operation.

In practice, RSMeans (2016) and Caterpillar Performance Handbook (Caterpillar 2016) are the two most widely accepted sources of construction productivity and cost benchmark data in the heavy construction industry. In general, CAT Handbook provides detailed time data for equipment cycle and its main steps given particular categories of equipment and work conditions. CAT Handbook is deemed more precise than RSMeans which merely features overall average production rates. The foundation knowledge and recent research in earthworks estimating and simulation are first reviewed. Then the differential factor between “precise estimating” by CAT Handbook and “rough estimating” by RSMeans is characterized in a quantitatively reliable fashion through conducting a case study based on a real world site grading project. As the equipment cycle time data available in CAT Handbook can be sufficient to enable the application of simulation modeling for fleet balancing and productivity improvement, simulation models were established to cross check estimating on case studies.

Construction work zones impose mobility constraints and remain major contributors toward incidents on our highways. Pro-active hazard identification and risk assessment is necessary in order to remediate the occurrences of incidents. The existing framework and applications in literature are not adequate in assimilating the hazards posed by both the construction activities and the dynamic flow of traffic in daily safety planning. This paper proposes a novel framework that leverages the capabilities offered by models in both construction safety models and traffic engineering domains to address this problem. Specifically, this study systematically integrates the outcomes from activity-based construction risk assessment and simulation-based traffic conflicts assessment for quantifying daily risk in highway construction. The application of this framework is demonstrated by a numerical example analogous to practice. Results indicate that enhancing the risk assessment of construction activities by chronologically integrating them with traffic conflicts around work zones capture safety hazards more realistically and completely. The integrated safety assessment index developed in this study will assist in accurate representation of risky work zone periods to devise preventive measures. The results and insights from this study will provide a holistic framework to support both the traffic planners and the construction project engineers in their day-to-day safety planning for highway projects.

A building code sets the minimum standards of a building project in various aspects for architects, engineers, constructors, and other stakeholders to abide by, such as requirements for building accessibility, structural safety, and occupant health/comfort. Regarding accessibility, each public building must have an accessible design that assures the locations of the entrances and exits of the building are accessible by people with disabilities, in addition to other accessible requirements from the Americans with Disabilities Act (ADA). However, without a proper simulation, it is difficult to assess if a building design satisfies the actual access need of people with disabilities, even if the design met accessibility code requirements. The use of virtual reality (VR) simulation, on the other hand, has a great potential to help people assess building accessibility before its construction, in an intuitive and objective manner; because potential occupants would be able to experience the building design in an immersive way. To test out the idea, the authors developed a model for a one-story administration building design in Unity 3D game engine and visualized the simulation in a VR environment. The building has spaces designed according to wheelchair accessibility requirements in ADA regulations, and spaces designed without such considerations. The simulation was tested by 10 volunteers during which they were required to check the ADA compliance and noncompliance cases in the building model by navigating through it using a game simulation in a computer and in the VR environment. The majority of the volunteers easily identified the spaces accessible and inaccessible to people in wheelchairs. The results showed that although the simulations in a computer and in VR both fulfilled the compliance verification purpose, the VR environment was mostly preferred.

It is well known that higher consumption of energy generated by fossil fuel leads to more greenhouse gas emissions, known to cause global warming. Especially, recent studies show that occupant behavior is a major factor affecting the energy consumption of buildings. While building energy simulation software may predict the overall energy consumption of buildings, it is not capable of determining the temperature distribution of a space, which is critical for studying occupant behavior. Thus, the authors have developed a pilot application of computational fluid dynamic (CFD) simulations to the analysis of indoor temperature distribution (ITD) considering the impact of heat from various sources such as electrical devices and solar heat. The purpose for developing the CFD simulation is to better understand the balance between accuracy, calculation speed, and computational cost with the goal of reaching real-time analysis of ITD. The temperature distribution from the zero energy (ZOE) research laboratory at the University of North Texas (UNT) is used to validate the CFD simulation. A method for increasing analytical speed and reducing computational cost was proposed and discussed.

Portfolio management, in particular, the selection and prioritization of projects among a pool of known projects, has been studied extensively by researchers. However, a major gap in the work from a strategic planning perspective is modeling the stream of unknown future projects. This paper presents findings of an ongoing project that explores a solution to this need. A proof of concept case study is presented using 12 years of records from the Florida Department of Transportation (FDOT) project database. Specifically, this paper reports on the development and evaluation of a stochastic project stream generator that can predict future unknown project streams of FDOT projects in terms of their occurrence, cost, and duration. The generator uses correlations between variables, macroeconomic metrics, and construction indices within a Monte Carlo simulation framework to predict stochastically future available projects and their characteristics. The paper concludes with an identification of on-going and future work.

The philosophy of lean construction has provided a great insight into the complexity of construction projects. Based on the concept of flows, this philosophy provides a theoretical and practical basis to understand the variability in flows, which can cause deterioration of the per-formance of construction projects by causing an increase waste. However, many previous stud-ies conducted based on lean construction assumed the independence of flows despite the fact that the interrelationship between flows can make the global system behaviors of construction projects more complex. Furthermore, the discrete event simulation (DES) methods primarily used in previous studies have a limitation in modeling a whole construction process of con-struction projects, as the DES model can represent only a process that can be recognized by the intuition of modelers. Therefore, in this work, a theoretical framework is proposed to describe the primary components and flows of construction projects and their interrelationship resulting from shared resources. In addition, an agent-based simulation framework is also proposed for representing the complex behaviors of construction projects. Finally, the applicability of the proposed simulation framework is discussed, compared with DES methods. The achievements of this work will contribute to broaden the insights of researchers and to improve the manage-rial abilities of practitioners.

Aging infrastructure in the nation and limited available budget makes the role of strategic decision making more important in its maintenance. The challenge is to maintain the infrastructure asset systems in a state of good repair and to improve the efficiency and performance of the infrastructure systems. Decision makers are in need of a decision support system to effectively allocate funding and achieve the highest possible return on investment on their infrastructure. This paper proposes and validates a framework for such decisions. The main objective is to develop a decision support system for selecting and prioritizing necessary actions for MR&R (maintenance, repair, and rehabilitation) of multiple asset items in order for a roadway to function within an acceptable level of service, budget, and time. To achieve these outcomes, this model creates a framework for an infrastructure asset management via analyzing a multi-objective problem. The objectives of the problem are: (i) minimizing maintenance costs, (ii) minimizing maintenance time, and (iii) maximizing level of service improvement. The results of implementing the proposed model on a case study are presented for a sub-set of asset items. Results reveal the capability of the model in generation of optimal solutions for the selection of maintenance strategies. The model optimizes decision making process and benefits decision makers by providing a set of solutions for infrastructure asset management. In addition, it provides a tool to run sensitivity analysis to evaluate annual budget effects. Application of the proposed model is implemented on roadway asset items; however, the model is not limited to roadways and is applicable to other infrastructure assets.

Struck-by accidents are a major type of accidents which can occur on construction sites. Researchers have developed several methods for reducing the chance of struck-by accidents in construction, such as spatial conflict assessment and workspace analysis. However, scarce attention has been paid to the impact individual risk taking tendencies have on the occurrence of struck-by accidents. Findings in psychology clearly suggest that individuals have different risk-taking tendencies (e.g., risk takers vs. risk avoiders) in various situations, and that behavioral patterns affect the occurrence of certain types of accidents. Individual differences in risk taking tendency manifest under dangerous situations, and consequently affect the probability that an individual experiences a struck-by accident. Against this background, it is proposed that an agent-based simulation can be used to predict potential collisions between equipment and workers and to investigate the impact of different risk-taking tendencies on the occurrence of stuck-by accidents on construction sites. In this research, different risk taking tendencies are modeled into agent behavior rules, this simulation model generates moving trajectories of workers and equipment, which can be used to predict potential collisions and the consequences of different individual risk-taking tendencies under collision-possible situations. This paper demonstrates research opportunities associated with the proposed approach using an illustrative example.

Advances in computational mechanics and simulation platforms create new opportunities for modeling and analyzing complex civil engineering problems and for exploring nonlinear system response and limit states. Nevertheless, methodologies and numerical tools currently in use for the risk assessment of critical structural systems are typically still based on a number of simplifying assumptions such as linear-elastic behavior of the superstructures and the idealization of equivalent linear behavior of the supporting soils. These traditional tools can diminish the ability to realistically investigate inelastic regimes, compromise the interpretation and prediction of the actual structural performance, and limit the ability to accurately explore limit states in a complex facility. In a comprehensive nonlinear computational framework development being supported by the U.S. Department of Energy, ongoing research aims at creating a robust nonlinear time-domain simulation tool set capable of capturing and rigorously modeling the nonlinearities of soil-structure systems. This paper will focus on a general verification and validation (V&V) methodology employed for ascertaining the accuracy of the nonlinear simulation technologies implemented in a new numerical tool under development (RealESSI).

Modern remote sensing technologies such as three-dimensional (3D) laser scanning or photogrammetry are in increasing demand for civil engineering applications such as infrastructure maintenance and as-built construction verification. In order to leverage the resulting point clouds, reliable estimation of the normal vector for each point in a cloud is a fundamental task for applications such as point-based rendering, 3D surface reconstruction, feature detection, or object segmentation. However, 3D point clouds generated by the aforementioned techniques are prone to noise, outliers, missing/erroneous data and sampling anisotropy, all of which impede reliable normal vector estimation. To address these issues, the method presented herein utilizes robust multivariate statistical outlier analysis for normal estimation in complex 3D point clouds with varying noise levels and geometric complexities. Robust adaptation of the Mahalanobis distance, in which the mean vector and covariance matrix are derived from a high-breakdown robust multivariate estimator called deterministic MM-estimator (DetMM), is used to find and discard outlier points within each points k-nearest neighbors during local plane estimation. The proposed algorithm is capable of handling points on highly curved surfaces or with sharp planar transitions. This new method is then compared against approaches such as minimum covariance determinant (MCD) estimator, RANdom SAmple Consensus (RANSAC) and its maximum likelihood variant (MLESAC), as well as principal component analysis (PCA) in local plane fitting applications. The performance and robustness of the presented normal estimation method is also demonstrated on several challenging noisy 3D point clouds. Experimental results indicate that this normal estimation method can improve the accuracy of surface mesh reconstruction and region growing point cloud segmentation while preserving the complex geometry of the reconstructed physical object.

Energy prediction methods and tools play a crucial role in the lifecycle of energy-efficient buildings. Such tools and methods could be leveraged at different stages of a project by different actors including but not limited to: 1) owners and facility managers for evaluating cost of ownership, 2) architects and engineers for selecting the best-fit system for target energy performance, and 3) construction managers for making more informed value-engineering decisions. There are two major approaches for energy modeling: forward modeling and inverse modeling. The former approach is the most common approach used in industry due to its quick analysis process. In this approach, provides a prediction of energy use by building system design parameters and environmental drivers using energy models. On contrary, inverse modeling approach is used to select building parameters such as energy conservation measures from energy use and drivers by developing statistical models with the use of available data. However, the gap between predicted and actual is larger compared to the forward modeling approach due to the differences in detailing and interoperability issue between the design authoring and analysis tools. Even though some energy analysis tools include an authoring module, the modeling details are limited. There is a variety of design authoring and energy simulation tools that could communicate with each other using open file formats such as industry foundation classes (IFC). This paper investigated two different energy analysis workflows and selection of parameters to compare them with the metered data available for an educational facility building located in 5A climate region. These two workflows include: 1) creating a physical model in a BIM authoring tool and transferring the model to an energy analysis model by means of IFC file format, 2) creating an energy model directly in an energy analysis tool. In the light of this case study, this research contributes to the body of knowledge by providing 1) justification of testing criteria for the selection of simulation tool, workflow, and level of development (LOD) for models, 2) parameters for modeling, and 3) metered conditions for the basecase to compare with two different energy analysis building information modeling (BIM) workflow by using same design authoring and energy simulation tools.

Barcodes are extensively used in inventory management to identify and track products and components. While the current prevalent method of reading barcodes with handheld barcode readers is suitable for retail checkout lanes and small-scale inventory management, it presents significant work efficiency and worker safety challenges for large-scale inventory management where products are usually stored on multistory racks in large warehouses or storage facilities. Such facilities present significant opportunities to improve logistics by automating the inventory management process. To ease inventory tracking while mitigating the identified issues, this research takes advantage of computer vision based barcode readers to replace traditional barcode readers and proposes three techniques to facilitate barcode extraction from video scan data. First, Harris corner detector and Hough transformation are proposed to work in unison to estimate the direction of a region including a single barcode and rotate it to an ideal state for the benefit of existing decoding algorithms. Then, an algorithm based on barcode region connectivity and geometry property is proposed to find multiple barcodes in a single image to avoid brute force searching of valid barcodes. In addition, a histogram difference based key frame selection method is also proposed to eliminate redundant information between sequential frames which helps to improve efficiency by using fewer frames. Finally, these techniques are applied to video data collected in a large logistics warehouse and the overall performance is evaluated. The experimental results show that our algorithm can achieve both satisfactory recognition rates and efficiency, and thus offers significant promise for broad deployment in automated large-scale inventory management using terrestrial and aerial robotic platforms.

Although construction is often seen as an outdoor activity, much of the work is completed indoors. The indoor environment presents many challenges for the automatic capture of data during progress monitoring, quality control, quantity verification, and overall project documentation. This paper examines the potential of using robots, and more specifically, unmanned aerial vehicles (UAV) and unmanned aerial systems (UAS), to monitor indoor construction sites. UAVs have become ubiquitous due to their commercialization as an off-the-shelf tool. However, their potential to facilitate the automation of many mundane and challenging activities in indoor construction is very high. Unfortunately, they also come with some important challenges. This paper explores the potential applications for UAV in indoor construction sites, their benefits, and their challenges. A sample of the benefits of this technology include their in-flight agility, capacity to hold an array of sensors, and automation potential. Challenges include potential safety risks if they distract or interfere with workers, navigation within ever changing surroundings, battery capacity limits, and impacts of automation. Finally, challenges of a multi-sensory smart UAS with respect to emerging technologies such as internet of things (IoT) and possible solutions are discussed.

For years, numerous studies have been done to apply advanced technologies to avoid workers from injuries. Recently with the development of robotics and automation technologies, the application of assistive wearable robotics has got more attention from construction industry after its successful applications in medical and military fields. However, there is little information about exoskeleton’s application in construction area since the technology is not fully adopted by construction industry yet. Thus, the objective of this paper is to review the state-of-art of exoskeleton and its applications in different fields, illustrate how exoskeleton can benefit construction workers, and discuss the challenges and opportunities.

Earthquakes may induce severe fire disasters in urban regions. Different fire following earthquake (FFE) models have been proposed and developing worldwide. However, the high-fidelity visualization of conflagration in an urban region is rarely mentioned in the existing models. The decision makers in government departments or fire departments are generally not professionals with respect to earthquake engineering. As a result, a high-fidelity visualization of conflagration is more understandable for such nonprofessional users. In this work, a high-fidelity visualization for FFE is presented for the simulation of fire spread and smoke effects, based on the OpenSceneGraph (OSG) 3D graphics toolkit and fire dynamic simulator (FDS) software. Specifically, the development of building burning states and fire spread in the entire region can be clearly shown via changing the color contours of the building, using OSG toolkit. The motions of smoke particles can be directly computed by FDS using the large-eddy simulation, which follows the laws of physics and generates a realistic smoke diffusion. A case study of downtown Taiyuan City with 44,152 buildings is performed. The results show that the FFE scene can be displayed realistically. The proposed high-fidelity visualization can help decision makers understand the overall situation of fire spread area and direction, estimate the sub-regions with high risks of FFE, and make decisions regarding fire rescue and fire protection planning. It can also serve as important technical supports for virtual reality-based fire disaster training.

People spend more than 90% of their time indoors, which makes it crucial to assess the relation between the interior environment and human experience. Architecture, psychology, and neuroscience researchers have been trying to relate different design features to human experience. However, the extent of how we feel and experience in a space has not been fully quantified yet. One of the challenges has been the technical limitations in generating architectural configurations similar to real-world settings. Technologies such as virtual reality and biometric tools provide opportunities for quantification of physiological and emotional conditions of humans as they interact with different architecture spaces. This paper provides an overview of the method and results of the initial experiments. Findings showed that emotional response of people change based on different architectural design features configurations. Findings will help practitioners in the AEC industry to improve the design process for achieving better human experience in spaces.

This study proposes a novel research technique that tackles one of the major causes of uncertainty in building performance—application of immersive virtual environment (IVE) in occupant energy related behavioral studies. The concept upon which occupant energy-use related behavior and thermal comfort is based, obviously encompasses both psychological and physiological parameters. The authors try to determine whether the virtual replica of the lab environment, as a proposed test-bed for studies of occupancy behavior, would have a significant impact on subjects’ perceiver thermal comfort. Given that, the main focus of this study is to gain insight into individuals’ thermal comfort and their physiological reactions in a wide range of energy expenditure modes of bodily activity, in IVE from what is typical in the physical environment, in-situ. To that end, pairwise comparisons across two comparable experimental settings were performed for heart rate (HR), oxygen uptake (VO₂), respiratory exchange ratio (RER), tympanic temperature (T_t), and skin temperature (T_sk) responses. The data revealed no statistically significant differences in the mean of the measured physiological variables between in-situ and IVE interventions except for the RER (p=0.025). This finding supports the ability of IVE to replicate mundane realism in occupant thermal comfort and subsequently occupant energy related behavior studies.

A clear better understanding of occupancy behaviors in built environments is a salient criterion that can help in satisfying building performance. In fact, even the most carefully validated simulations have discrepancies from reality. The purpose of this study is to establish a procedure for evaluating the impact of discrepancies between lighting simulations in immersive virtual environment (IVE) and actual lighting conditions on human subjects’ visual perception in IVE. To that end, an IVE which is aggregated with lighting simulation software, namely RADIANCE, will be developed with similar lighting in working spaces and in-situ. The perceived qualitative features (e.g. illuminance levels, color, and glare) of the lighting and sensation will be measured by using a subjective method and the difference of illuminance between the IVE and in-situ setting will account for the aforementioned discrepancies across the two experimental settings. In the long run, this study aims at determining the extent to which these discrepancies will impact occupant lighting sensation in the virtual scene and subsequently its potential to impact their lighting-use behavior in IVE.

In a construction project, cranes are usually involved in the mobilization of materials and equipment. Therefore, a well-planned crane movement on a construction job site increases the efficiency of construction operations. This is especially the case when a crane is being operated from the material storage place to the workspace in a highway construction project, because the time allowed for such movement is typically tight, as constrained by the vehicular traffic on the highway. To explore the use of construction simulation in facilitating people’s understanding of construction processes, this paper presents a crane movement simulation from the site storage location to the beam erection location in a highway construction project on the Interstate 94. This simulation was modeled in the Unity 3D game engine and visualized in a CAVE virtual reality environment. CAVE is an immersive virtual reality environment where projections according to different view angles of a scene are projected onto different walls of a space. The purpose of this simulation is to help people plan crane routes during the construction planning phase. The objectives of planning the crane routes are minimizing crane trip length and trip time so as to minimize interruptions to the highway traffic, save construction costs, and achieve more sustainable construction operations. The simulation was tried by 72 volunteers. A survey of these volunteers showed that 95% of them understood crane operations after trying the simulation, compared to 35% before trying the simulation, and 80% of the responses agreed that such simulation helps people conduct construction planning.

The ability of workers to recognize and manage construction hazards is essential for effective safety management. However, studies have unanimously demonstrated that a large proportion of construction hazards remain unrecognized in dynamic work environments. Such poor hazard recognition levels have been partly attributed to the pervasive use of unengaging and ineffective training practices within construction. To improve training effectiveness, recent efforts have focused on assessing the learning needs of particular workers, and customizing training experiences accordingly to maximize training outcomes. This paper builds upon the previous research by developing an immersive safety training environment that provide a more effective personalized training experience for workers. After development, the degree of realism and immersive experience offered by the training environment was measured and found to be 73% of the real environment. The findings of this study will be useful to practicing professionals seeking to improve training efforts are safety training outcomes.