Construction Research Congress 2022: Computer Applications, Automation, and Data Analytics

Measurement of construction labor productivity involves various subjective factors (e.g., motivation, stress, and fatigue). Most measurement approaches for subjective factors in productivity applications require manual data collection (e.g., questionnaires, interviews, and observations); therefore, research gaps exist regarding how to (1) directly measure subjective factors using data that reflect workers’ real performance at single points in time, and (2) integrate these factors into existing or new models in labor productivity applications. This paper proposes an empirical framework for integrating real-time data from multiple sensors for directly measuring subjective factors affecting labor productivity. The proposed framework, which was designed, built, and evaluated using design science research methodology, contributes to the body of knowledge as part of a longer-term study proposing an empirical framework for triangulating data from a multi-sensor system to simultaneously measure multiple subjective factors affecting labor productivity. Study outcomes will complement existing artificial intelligence, simulation, and statistical models for construction productivity applications.

This study reviews the concept of Digital Twins (DTs) and related studies in the construction industry and identifies three key factors that is missing from the current practices. The missing factors are: (1) inadequate consideration of occupants in DT models, (2) lack of the inclusion of unstructured data, and (3) absence of Linked Data technologies. To address these issues, architecture for the design of DTs is proposed and partially implemented in a case study. The proposed architecture utilizes semantic web technologies and proposes a linked data approach to integrate different data sources of a DT. Further, the architecture leverages machine learning approaches to dynamically update and enrich the linked data platform and automate its maintenance. The case study takes the first step to integrate BIM and unstructured data generated by occupants (as work orders) using a linked-data approach. The research sets the path for future works in the domain of building DTs.

Recognized as a best practice, risk management offers an effective method to ensure project success under uncertain conditions. Particularly true for major transportation projects, where technical and institutional complexity increases the risk and challenges for collaboration, practitioners implement assessment and tracking. The vast majority of research in the field of risk identification focuses on the expertise, views, and judgments of Subject Matter Experts (SMEs). However, public agencies with limited experience in major transportation projects find themselves in a challenging situation in performing and auditing risk assessments for accuracy and robustness. The need for a comprehensive risk template to facilitate risk assessment grows rapidly. This paper presents a novel data-driven framework for developing typical risk templates for major transportation. The method bases its techniques on a large risk database of more than 70 US major transportation projects. Using natural language processing and deep learning techniques, the model uses different terminology to detect common risks in various projects and provides a summary of the risk consequences based on probability, cost, and schedule impacts. The advantage of this informative risk template not only captures important risks based on the frequency of occurrence in actual project data but also considers the potential consequences of each risk item. Also, the novel data-driven approach customizes risk registers according to specific project characteristics.

Researchers have developed various automated methods to recognize rebar using ground penetrating radar (GPR). These methods can be classified as either pattern-based or machine-learning-based methods. However, these methods are limited by their low robustness in noisy cases or heavy dependence on training large-size datasets. Thus, this paper proposes a rebar recognition method based on the ShearLet Transform, which can distinguish signal from different scales and directions. The proposed method includes two sequential phases: (1) hyperbola decomposition that decomposes rebar signal into a specific scale and direction; and (2) hyperbola reconstruction that reassigns the decomposition components to form new hyperbola without noise. A concrete building is selected to validate our method. The results revealed that: (1) the proposed method can achieve F1 score with 0.9649 on the collected dataset; and (2) it is a robust method that can discriminate strong noise, separate interlaced rebars, and remove cross rebar signals and direct wave.

The challenges of climate change, water and food scarcity have created the need for planning tools that assist in the agricultural decision-making process. This study aims to propose a multi-objective automated optimization model that uses an embedded comprehensive database to maximize the economic return whilst ensuring minimal water consumption. The suggested model capitalizes on big data in farming and greenhouses to filter all viable scenarios for a given soil, climate properties, and water availability. Evolutionary and genetic algorithms are employed to assess the long-term production, profitability, and water consumption of the scenarios. This study describes how the automated model operates through applying it to a case study to optimize the use of a land plot in Giza, Egypt, for farming purposes. The research opens the door for further application of the proposed model in different contexts both regionally and internationally, thus playing a vital role in the water and food nexus.

Big Data (BD) can revolutionize the way that the architecture, engineering, and construction (AEC) industry operates by improving decision-making processes and improving design and construction decisions. However, there is a fundamental lack of understanding about how to best employ available data into AEC processes. After providing a brief summary of the state-of-the-practice and identifying key sources of BD in the construction industry, this paper considers how BD could be employed in decisions to change the flow of information. Recognizing that the first step to filling the identified void is an improved understanding is what data are actually generated and available to AEC project teams, and when, this research identifies this information and presents a framework of data flow throughout the project’s lifecycle. Ultimately, this information will improve core knowledge around how BD can and should be used by engineers during their decision-making processes throughout the project to allow for a better project outcome, faster delivery rates, and overall higher quality. As we conclude the paper, we then poise a future path forward for this research.

In recent years, emerging technologies such as the Internet of Things (IoT) and Artificial Intelligence (AI) have broadened the horizon for innovative applications of smart buildings, supporting people with cognitive impairment to live more safely and independently. Currently, researchers and industry professionals need a state-of-the-art overview of the smart building innovations that can be used to facilitate the independence of the cognitively impaired population. This paper aims to evaluate and summarize the existing and potential smart building applications that can improve the welfare of cognitively impaired occupants by offering them supports in their daily lives. This review finds that most contributions have been toward data analytics for behavioral recognition to identify assistive tasks and to provide guidance. More smart innovations of the built environment are in need to provide assistive services and compensate for the intelligence shortfall of occupants who are suffering from cognitive-related dysfunctions. Especially, user-centered designs are needed to improve the smart buildings’ acceptability and usability.

This study aims to realize construction hazard warning by determining critical characteristics of hazards and using network of network (NoN) analysis. First, text mining is adopted to analyze the hazard text records. Word segmentation, part-of-speech tagging, and word frequency statistics are performed to extract the high-frequency and meaningful nouns as hazard characteristics. Then, the co-occurrence matrix between characteristics is calculated as the data basis of characteristic hierarchical clustering to determine characteristic indicators. After that, the NoN is established with hazard network as the bottom layer, hazard characteristic network as the middle layer, and characteristic indicator network as the top layer. Finally, the critical characteristics and hazards are determined by considering the correlation strength and network index. The results show that the correlation between each network layer is close and complex. And the layer of hazard characteristic network has more advantages than hazard network layer with tighter connections and higher quality nodes. The proposed NoN based on hazard characteristic may assist to focus on critical hazards and characteristic for improving the efficiency of hazard identification.

Accurate forecasting of field production is an essential ingredient for effective planning, management, and control of construction resources. Traditional forecasting methods often rely on historical data and on the subjective experience of project managers that fail to account for the dynamic and unique nature of construction operations. As such, this paper presents a generic framework to collect, mine, and analyze ongoing field production data. It incorporates developing a time series machine learning based model to forecast production for the upcoming days. The framework was tested on excavation activities of an infrastructure project for eight months. The developed machine learning model showed a satisfactory performance with a good model fit R² of 0.94 and an RMSE on the training set of 548 cubic meters, 563 on the validation set, and 652 on the test set. Therefore, the proposed framework can assist project managers to accurately forecast field production in order to make data-driven resource allocation decisions.

In today’s technology-driven world, data have become an integral part of all industries. Businesses are embracing technologies that can analyze data to recognize past, present, and future trends. Data analytics can help industries uncover historical data patterns to make informed decisions for a successful future. The traditional construction industry generates enormous amounts of data from the project’s inception to completion. Despite the availability of massive project data, the use of data analytics in the construction industry is still in nascent stages when compared to other industries. This is because the potential uses of data analytics in the construction industry have not been studied adequately. This paper investigates the current state of adoption and future potential applications of data analytics across multiple spheres in the construction industry. Following a review of the literature, semi-structured interviews with data analytics professionals in the construction industry are conducted. The outcome includes key data analytics applications in preconstruction, quality, safety, financial management, and decision making, while cost implications, data quality, and company culture are hindrances to wider adoption. Understanding the applications of data analytics helps contractors in gaining new insights to achieve improved efficiencies.

Construction work zones are a major cause of traffic disruptions and delays on roadways. Thus, accurate estimation of traffic states within work zones would be beneficial to both road users and transportation agencies. To this end, a deep neural network has been calibrated based on the hourly data points collected from 80 projects completed in Utah from 2013 to 2020. Reported results show that the proposed model outperforms its counterparts from the literature while achieving the R score of 0.97, RMSE of 185, and MAE of 108. Comparing the study results with the Highway Capacity Manual 2016 (HCM) shows that the proposed model is a good alternative for work zone capacity estimation. Future studies could leverage the probe vehicle data to improve the model’s performance by decreasing the RMSE and MAE values.

This study explores the use of satellite imagery as a rapid and cost-effective alternative to evaluate road surface condition. Spectral and texture information derived from 30-cm panchromatic and 1.2-m spaceborne optical imagery were used to assess the capabilities of satellite images in describing pavement distresses. The spectral analysis of the multispectral images indicates that pavements become brighter across all the bands as the roads become rougher. The pixel brightness values decreased when pavements were classified based on deteriorating rutting condition. While this may be useful in characterizing individual distresses, this opposing behavior of the pixel brightness values limits the capabilities of spectral indices approach in analyzing overall pavement condition when multiple distresses are present in the same location. All the texture metrics were found to follow a logical relation with pavement conditions, indicating strong evidence that satellite data have the potential to be an alternative method to evaluate pavement condition.

The use of Machine Learning (ML), Deep Learning, and Artificial Intelligence (AI) in building construction has been gaining traction since the mid 2000s. The ability to process the ever-increasing construction data, identify patterns, and predict future values has encouraged researchers to develop informed decision-making applications. This provides solutions to the challenges faced in the different areas of construction. This paper provides a literature review to identify, review, and categorize the existing body of knowledge involving the research and implementation of AI and ML in building construction. Related papers from journals were searched and identified using Scopus and Web of Science databases. Domain areas reviewed included clash detection, construction contract, cost, documents, equipment, labor, material, monitoring, planning, and scheduling productivity, risk, safety, and waste. Research studies reviewed are summarized and analyzed to identify gaps, describe the utilized algorithms and their applications, and aid in subsequent work by the research team.

The demand for safe, reliable, and higher-quality infrastructure systems often increases the complexity of transportation construction projects and necessitates more comprehensive evaluation methods leading to the incorporation of technologies. With a focus on effective technology implementation, this research stretches beyond simply evaluating technologies to include investigating the technology integration of the technology with personnel and policy at Departments of Transportation (DOTs). Drawing from a literature review and a survey of DOT personnel, this study applies and evaluates maturity models for people, processes, and technologies combined with a technology-task fit model. The findings support the need for understanding people, process, and technology maturity to improve technology implementation. This paper presents a preliminary assessment of technology maturity in the context of readiness for DOT use and move toward an assessment tool and an approach for DOTs to improve the likelihood of implementation success by moving to higher maturity levels. The study includes analysis of Automatic Identification and Data Capture (AIDC) technologies, Radio Frequency Identification (RFID), barcoding and readers, object-recognition devices, Global Positioning System (GPS), or Global Navigation Satellite System (GNSS), Geographic Information System (GIS), Unmanned Aerial Systems or Vehicles (UAS/UAV), Ground Penetrating Radar (GPR), Light Detection and Ranging (LiDAR), Electronic ticketing, and Infrared Sensing (IR). The future objective of the research will be to develop guidelines for the successful implementation of these technologies for highway construction and infrastructure asset management. This paper introduces three models, and their possible future uses as presented through preliminary survey results.

COVID-19 has spread rampantly in 2020 and people in different countries experienced unprecedented challenges to respond immediately. It has deeply shaken different industries and the overall economy. Many businesses were closed during this disease outbreak, but the construction industry continued to operate while others struggled. As the pandemic spread, the intensity of social media interactions dramatically increased as people started maintaining physical distances and staying at home. Recent studies have shown that social media serves the purpose of disseminating information in major crises and such social interactions correlate with what happens in the real world. Unlike traditional media, social media data also offered sufficient statistical power to understand risk communication patterns. The construction industry adopted innovative strategies to continue its operations and activities while responding to a number of concerns associated with health risks of the workers on site. This study aims at identifying the emerging communication patterns of construction stakeholders on social media (Twitter) during COVID-19. Around 12 million tweets were collected in the early days of the ongoing pandemic (March, April, and May 2020) by Twitter Application Programming Interface (API), in the region of North America; then filtered this data based on keywords pertaining to COVID-19 and construction activities. Such information was processed by applying several machine learning and natural language processing techniques. The spatiotemporal patterns of construction risk communication revealed in this study would support policymakers, businesses, and other industries to take more efficient measures and be more resilient both in the current and future pandemics.

Construction hazards and workers’ unsafe behavior are two critical factors contributing to fatalities and injuries at construction sites. Current hazard identification practice mostly relies on safety managers’ prior knowledge and experience, which leaves numerous hazards unidentifed. In addition, many workers engage in risk-taking behaviors, although they correctly recognize hazards, which can lead to severe accidents. Therefore, it is important to help workers better identify hazards and engage in safe behavior for proactive and preventive safety management. Previous studies revealed that analyzing workers’ electroencephalogram (EEG) signals provides an opportunity to identify dynamic construction hazards, and behavior intervention can modify workers’ unsafe behavior. However, it still remains unclear how the identified relationship between EEG signals and hazards can be incorporated into proactive and preventive safety management. This paper proposes an EEG-based ubiquitous hazard identification and proactive safety management framework that comprises the following three steps: (1) development of immersed EEG hazard classifier, (2) multi-sensor, real-time hazard mapping, and (3) behavior intervention. The feasibility of the proposed framework was demonstrated by focusing on the first component (immersed EEG hazard classifier) based on indoor laboratory experiments. It was found that the binary classifier classified the hazard-related EEG signals with 93.7% accuracy, and the multiclass classifier was able to classify the EEG signals into five different hazard types with 79.3% accuracy. It is expected that the use of the framework can advance the current safety management practice by providing a toolbox that can better identify construction hazards.

The majority of the currently available roadway network data is presented in the format of centerlines. This format has its unique advantages, but other important attributes of a roadway network such as width and area are commonly missing from roadway management agencies’ database. In recent years, airborne LiDAR and high-spatial resolution aerial photos are routinely collected by government agencies and provided to the public for free access. Coupled with advanced image processing techniques, these data hold the potential to permit the identification of transportation infrastructure assets with unprecedented accuracy and speed, and ultimately, the extraction of drivable space with higher accuracy than has traditionally been possible by ground survey. This research explored the utility of airborne LiDAR and high-spatial resolution aerial photos for extracting drivable space information from a complex urban area. Subsequently, roadway width information was derived from the extracted drivable space data. Research results revealed that the accuracy of extracting roadway network width information from airborne LiDAR data and high-spatial resolution aerial photos are acceptable for city planning purposes, but may not be used for engineering design purposes.

Residential buildings consume a significant portion of energy and electricity in the US. The residential housing stock in rural areas of the US specifically is more likely to be older and less energy-efficient than those homes built in more urban areas. Improving the energy efficiency of such residential buildings includes benefits such as overall reduction in emissions resulting from building operations, and reduction in costs to the building occupants. In order to accomplish these goals, understanding the energy performance characteristics of buildings in these rural areas is important. This can help to better define which buildings are in greatest need of energy efficiency retrofits, and establish a framework by which to target such buildings for performance improvements through efforts such as utility-supported rebate programs. However, to date, the study of rural residential buildings has been limited, in part, due to the lack of available data and detailed information on buildings. In this research, through collaboration with the communities of Ames, Bloomfield, and Cedar Falls, Iowa, US, building energy use data and building characteristic data were used to assess the overall characteristics and energy performance of rural residential buildings. Using statistical analysis and data mining techniques, this study investigates the differential changes in energy use intensity (EUI) relative to the average built area, the building age, among others. Such evaluations can be used to design retrofit and rebate policies and programs that benefit both individual building occupants, and also address absolute levels of consumption now and moving forward.

Virtual reality systems are making significant headway in the construction industry for design, training, planning, and management applications. With research continuing to show the virtues of adopting virtual reality in blue-collar occupational environments, the industry is making rapid and substantial investments to facilitate institutional innovation that supports adoption. Despite a push from both the research and professional community, the widespread adoption of virtual reality remains hindered because details on practical application remains unclear. This study used a social constructionist approach to capture the perspectives of industry professionals on the challenges and opportunities associated with use of virtual reality for construction safety applications. The interviews with professionals revealed that the application of virtual reality in the workplace is currently in its infancy stage and lacks standardization primarily due to scalability and quality concerns. Additionally, the cost associated with large-scale adoption also remains prohibitive to date. The experiential learnings of professionals are used in this paper to determine where future research and practice efforts need to focus to pave the way for future of virtual reality within the construction industry.

Most building energy-saving strategies are based on thermal comfort standards defined by regulatory bodies (e.g., ASHRAE), which do not account for individual differences or preferences of the building occupants. To address this gap, this paper proposes a non-intrusive automated method to capture the individual thermal discomfort states of occupants over time based on their thermal discomfort cues. The method aims to support the identification of personalized energy-saving strategies toward improved occupant thermal comfort and reduced building energy consumption. It uses a deep learning model to recognize occupant cues from videos. To implement and test the proposed approach, participants were video recorded in uncontrolled office settings while performing work-related activities. The deep learning model was used to recognize the thermal discomfort cues, and a thermal comfort profile that captures the discomfort states of occupants over time was developed. The method achieved a weighted recall and precision of 84% and 96%, respectively, in recognizing building occupant thermal discomfort cues.

The selection of high-energy performance facade systems could play a pivotal role in enhancing energy efficiency and promoting sustainability in buildings. However, many decisions on the selection of building facade systems are made at the early building design stage when no detailed building information is available. Therefore, an intelligent advisory system is needed to support designers and architects in selecting high-performance facade systems at the early stage of building design. The main objective of this paper is to develop an advisory system that extracts underlying structures from the thermal behavior of façade systems and expresses them as recommendations to support architects and designers in the early stages of building design for the selection of high-performance and energy-efficient façade systems. The proposed advisory system extracts useful rules from data using the association rule mining technique and expresses them as recommendations to support the selection of façade systems. As an illustration, the proposed system was examined and validated on an ultra-high-performance fiber-reinforced-concrete (UHP-FRC) façade panel. The results showed that the proposed advisory system facilitates decision-making in selecting high-energy performance facade systems in the early building design stage by providing insights into the performance of facade panels in various building contexts and weather conditions.

Collaborative robots have recently shown the potential to assist workers in performing dexterous and physically demanding construction tasks. However, due to unparalleled diligence, a lack of social intelligence, and a lack of functional perception, a collaborative robot poses problems for the mental and emotional states of workers when interacting with them. These effects can lead to mental health problems, such as depression and anxiety. To ensure the mental well-being of workers during human-robot collaboration (HRC), this study proposes a psychologically stimulated motion planner built into collaborative robots. This capacity to adapt will be based on the robots’ recognition of workers’ mental states. Through it, first, the cognitive load of workers will be decoded from their brainwave signals via the Multilayer Perceptron Classifier. Second, the decoded results will activate a robotic control mechanism to achieve human-friendly interaction. To examine the safety and feasibility of the robot motion planner, an HRC materials-delivery experiment was designed using a 3D robotic-simulation environment (i.e., Gazebo). The planner allowed robots to assess subjects’ cognitive load with an 86.4% accuracy. Once subjects experienced an undesirable cognitive load, the motion planner adjusted the movements of the robots to deliver materials at a safer distance and more comfortable speed. The findings demonstrate the potential of establishing the psych physiologically based HRC solution, opening new avenues to implementing collaborative robots at construction sites.

Work-related musculoskeletal disorder (WMSD) is a critical occupational hazard and among the leading causes of nonfatal injuries in construction. Rapid ergonomic assessment is important to proactively detect and prevent WMSD-related hazards. This study proposes a novel deep learning framework for ergonomic assessment from construction videos. First, continuous skeleton postures of workers are extracted using a deep-learning-based pose tracking algorithm. Second, a long short-term memory (LSTM) based multi-task learning (MTL) model is created to simultaneously classify various ergonomic poses in different body parts using time-series skeleton postures. Finally, Ovako working posture analysis system (OWAS) is applied to assess the ergonomic risk from the identified poses in different body parts. Real-world construction videos are used to demonstrate the efficacy of the proposed method. Compared with existing vision-based ergonomic assessment methods, the novelty and contribution of this study is: (1) this study leverages LSTM network to exploit the temporal dependency among time-series skeleton postures, which effectively mitigates the errors associated with a single-frame posture and improves the accuracy, and (2) MTL is adopted to learn a unified classifier for multiple body parts leveraging the commonality in human pose, leading to improved performance and computational efficiency.

Recent advances in computer vision and camera-equipped unmanned aerial systems (UASs) for 3D modeling enabled UAS-based surveys with high spatial–temporal resolutions. Although there have been many studies on various factors that affect the accuracy of UAS-based photogrammetry-based geospatial data, there is little to no quantitative method that identifies the relationships between flight configuration factors and point cloud accuracy. This paper seeks to assess the degrees of influences that these factors affect accuracy. This study assesses five impact factors using the ANOVA method to understand which factors have higher or lower impacts on the accuracy. The influence factors include flight height, average image quality, image overlap, ground control point (GCP) quantities, and focal lengths of cameras. The results show that the number of GCPs has the largest impact on the vertical and horizontal accuracies than other impact factors. The findings of this study can help surveyors better design flight configurations given different site conditions and constraints.

With the advances in Augmented Reality (AR) technology, head-mounted displays (HMDs) provide users with various information presentation options, which combine the physical world with virtual objects through a 2D or 3D display. Compared with traditional blueprint documentation, AR presents the design in a more perceptual and intuitive graphic language. However, it is necessary to understand how different information displays through HMD affect user’s comprehending performance in terms of both efficiency and cognitive status. In this paper, we describe a study measuring the impact of information display types in AR HMD on the user’s comprehension of complex construction information. Subjects from construction engineering background are using both 3D conformal model and 2D tag-along image displayed through an AR HMD to observe a roof deck design. The evaluation of each method includes both mental workload and comprehension performance. Significant differences were captured in the performance evaluation using electroencephalogram (EEG) analysis. The results show the advantage of 3D display in reducing mental workload and improving comprehending performance.

Artificial intelligence is rapidly expanding into energy infrastructure for its implementation in smart grid integration, energy management, and electric vehicle production. However, research indicates AI techniques are often biased against racial, gender, and socioeconomic minorities. This paper is a preliminary review of current AI techniques for predictive modeling and applications in energy engineering and construction. This work was completed through the qualitative coding of 71 publications relating to artificial intelligence, energy infrastructure, and bias. Government researchers, construction leaders, scholars, and technologists rely on AI systems daily to analyze energy consumption, enhance power distribution, and ensure safe transportation. Therefore, in order to construct an accurate and inclusive AI technology for energy infrastructure this discrimination cycle must be identified, addressed, and eliminated. The literature review examines the application of AI primarily in smart buildings, smart homes, and smart grid technology. This work highlights the insufficient attention to social and economic inequities in energy automation.

Vision-based 3D pose estimation of construction workers has drawn attention for its usefulness in occupational ergonomics, safety, and productivity analysis. However, it is still challenging to develop an extensive training image dataset, which is essential for deep neural network-powered approaches, thus inhibiting the maximum potential of vision-based 3D pose estimation. To address this issue, we built a synthetic training image dataset and validated its effectiveness for 3D pose estimation. We trained and tested a state-of-the-art 3D pose estimation architecture using these synthetic images. The results show that the synthetic data-trained model can estimate 3D poses of construction workers with a Mean Per-Joint Position Error of 50.24 mm—comparable to real-data-trained model (46.5 mm). This finding indicates that synthesized construction images are effective in training a 3D pose estimation model, thus enabling the development of more accurate and scalable 3D pose estimation and alleviating the shortage of real-world construction training data.

Reliable communication is a critical component of a successful emergency management operation in resilient cities. However, large-scale extreme events may demolish telecommunication infrastructure. The current state of practice to maintain communication when telecom infrastructure are not available heavily depend on expensive equipment (e.g., transportable telecom towers and unmanned aerial networks) that may not be readily available everywhere. To address this issue, this study aims to develop theoretical foundations for a new framework that provides a fully decentralized infrastructure-less communication system using ad hoc mobile networks. The proposed framework uses device-to-device (D2D) wireless connection, distributed ledgers adopted from Blockchain technology, and a new all-to-all network broadcasting algorithm. The proposed infrastructure-less system collects critical spatial data from available cellphone devices in the affected area and shares it among all of them. At the end of this procedure, each device will have spatial information of all available devices. When emergency teams find a device through a D2D connection, they will actually have access to spatial data of all devices and can locate all people in one step.

The world is urbanizing rapidly. By 2050, 68% of the world population will reside in urban areas, posing unprecedented pressures on the livability, equity, and sustainability of city systems. While the concept of smart city holds great promise to release these pressures, the planning decision making toward smart cities is challenging because of the enormous breadth of smart city dimensions. As a result, decision makers are often lost in where and how to make the best investments to address the most pressing issues at hand. In recent years, many smart city initiatives and programs have been implemented, which offers an important source of smart city knowledge for decision makers to learn from. To this end, this paper proposes a smart city ontology, as a shared conceptualization of smart city knowledge, to facilitate the sharing and reuse of the knowledge for informed planning decision making. At the highest level of abstraction, the ontology semantically models three key elements of smart city knowledge, including visions, challenges, and solutions. This paper focuses on presenting the proposed ontology and its evaluation. The evaluation results showed the promise of the proposed ontology in supporting smart city planning decision making.

A concept of “digital twin” as a model to bridge between a real-world and a virtual twin has emerged in manufacturing and product management. In the context of civil and infrastructure domains, the term “digital twin” has been redefined as up-to-date digital representations of built environments. This paper proposes a real-time participatory sensing-driven computational framework to model the up-to-date state of built environments in a virtual environment. In the proposed framework, crowdsourced visual data are obtained from citizens’ participation to automatically identify the up-to-date states of infrastructure that would negatively impact the community resilience in extreme weather conditions. Then, to update the associated built environment information, real-time participatory sensing data are processed by using deep learning algorithms. Finally, the identified geometric and geospatial information of infrastructure is fed into the virtual environment toward a digital twin city model. Case studies in the context of power distribution infrastructure systems were conducted in Houston, TX, and it was demonstrated that the proposed method robustly updates the up-to-date condition of infrastructure into the digital twin city model. The proposed approach for updating virtual models based on participatory sensing-based real-time information has a great potential to facilitate data-driven decision-making for urban planning and infrastructure management in a smart city digital twin.

Information extraction provides an opportunity to automatically extract safety requirements from construction safety regulations to support automated safety compliance checking for detecting field non-compliances with these regulations. However, previous efforts on automating the safety compliance checking process fall short in their scalability and ability to automatically extract safety requirements, due to the complexity in unstructured text. Therefore, this paper proposes a deep learning-based information extraction method for extracting relations that link fall protection-related entities extracted from construction safety regulations for supporting automated field compliance checking. The proposed method uses an attention-based convolutional neural network model for recognizing and classifying relations. The proposed method was implemented and tested on two selected Occupational Safety and Health Administration (OSHA) sections related to fall protection. It has achieved a weighted precision, recall, and F-1 measure of 82.7%, 81.1%, and 81.3%, respectively, which indicates good relation extraction performance.

3D reconstruction has become a common method to obtain digital representations of the built environment. Meanwhile, Scan-to-BIM technologies are investigated to extract semantic information from raw point clouds data. To improve the accuracy and reduce the time in object segmentation, Deep Learning (DL) techniques have been proposed to process 3D point clouds and proved as a promising approach in unmanned-vehicle driving. However, due to the little available dataset specified for construction projects, implementing deep learning in Scan-to-BIM applications is limited. Therefore, a point cloud dataset including the common scenes in construction is urgently desired. This paper is a part of a research project aiming to develop a comprehensive 3D point clouds dataset integrating the Red/Green/Blue (RGB), XYZ, intensity, and thermal data collected using digital cameras, terrestrial laser scanners, and infrared cameras. The collected data were first transformed into panorama (PAN) form to ensure enough and flexible field of view (FoV). Then, a registration method is proposed to fuse these data from different sources and to establish an integrated dataset. Two case studies were carried out in both indoor and outdoor environment to investigate the feasibility of the proposed methodology. The preliminary results show that feature-based registration method provides a reliable alignment between different data.

Highway work zones are hazardous not only for motorists but also for the laborers who work on and nearby them. Because of the frequent use of heavy equipment, rapidly changing conditions, and proximity to fast moving cars, highway workers suffer from higher rates of injury and death than the rest of the construction industry. The rate of injuries and accidents during construction on highway projects is a concern for the public and private sectors. Thus, new technologies must be experimented with in order to overcome this disparity. The highway work environment requires a holistic safety analysis and continuous monitoring of changing conditions. The combination of wearable technology and Unmanned Arial Vehicles (UAVs) is able to create a robust framework for pre-construction safety planning, hazardous identification during construction, and continuous monitoring of potential hazards throughout the construction phase. This study developed a safety protocol that helps transportation agenesis and contractors enhance safety practices on highway projects. The proposed protocol consists of (1) construction site analysis, (2) digital data collection before and during construction, (3) pre-construction safety analysis based on photogrammetry data, and (4) digital site safety plan. The proposed system is a proactive approach to improving safety planning and practices of complex projects such as highway projects. This inevitability will reduce potential hazards before construction commences and allow safety officers to monitor site changes.

Cost overruns and schedule delays in US transit projects have been of growing concern for years. Nevertheless, the data availability and sample size have restricted quantitative analysis toward investigating the risks leading to overruns. Innovative data sources and collection methods need to be identified in addition to traditional surveys and case studies. News articles report on issues and risk events leading to overruns as projects progress but have not yet been explored in the construction domain. The difficulty lies in data compilation and analysis. To fill this gap, the paper tested combinations of different natural language processing (NLP) and machine learning methods to automatically identify risk narratives from news articles. The risk sentences are classified into 5 categories and 26 subcategories through a content analysis approach. Then the risks are ranked and analyzed using an appropriate co-occurrence network. The research demonstrates the possibility of integrating NLP and network analysis for exploring publicly available textual documents to explain project performance issues. The approach serves as a baseline for future studies to develop more intelligent models to examine a wide range of media data and other textual reports in the construction domain.

Residents have full control over home systems and place a significant influence on energy consumption. Despite advanced building technologies and energy-efficient appliances, energy consumption in residential buildings remains high in the US and around the world. This can be explained by the interaction effects between building technology and occupant behavior. Given an increasing number of studies on occupant behavior, residents’ routine daily energy use activities remain unclear in the literature. The goal of this study is to identify the components that influence the similarity and differences of energy usage-related activities. To achieve the goal, this study aims to (1) compare the components of routine occupant behaviors using the US national behavior data by region, and (2) identify if geographical location affects the characteristics of activities using GIS. The findings inform that duration, start time, and end time have more influences on the differences in energy usage-related activities, and watching TV, washing and grooming, and cooking and food preparation are more different by geographical location. The result can be used to provide more reliable information regarding energy and behavior to the occupants in residential buildings. Also, the result can be applied to the new energy and behavior strategies and policies about residential building energy plans.

The data-analytics solutions are currently more needed to be implemented regarding construction firms to allow for more flexible operations and higher productivity. Yet these data analytical applications are difficult to undertake because of low-level awareness and control of data quality. Through literature review, this paper identifies that the integrated applications that fuse multiple data sources and apply advanced algorithms become a research trend. However, an approach to assessing and controlling data quality is missing in these applications. This paper conceptualizes a novel framework for Data Quality Assessing and Controlling (DQAC), based on the cycle of Total Data Quality Management (TDQM). Then, this paper designs the architecture of DQAC and elaborates the functionalities and main components, which could be the theoretical foundation for building a data-quality-aware analytical application. The data quality measuring engine translates queries and retrieving quality information from the data wrapper, also it generates a data quality profile. Quality control strategies complete the TDQM cycle on both the level of extraction, transformation, and loading (ETL) process and data warehouse operation. Future research needs to develop a prototype to validate the DQAC using empirical data.

Studies in the past decades have shown that Real-Time Location Systems (RTLS) have strong potential in improving safety and enhancing productivity on construction sites. Ultra-Wide Band (UWB) RTLS in particular, has demonstrated high suitability for construction site conditions due to its higher accuracy, lower dependency on the Line-Of-Sight (LOS), and less signal attenuation and multipath effect. However, higher tolerance to Non-Line-Of-Sight (NLOS) is not equivalent to not having impacts on the accuracy of location estimation. To reduce the impact of NLOS, typically standardized residual analysis of Least Square or Kalman Filter are used (to detect and remove the outliers). However, these methods can only be functional with a minimum of five anchors. The objective of this study is to determine and mitigate the impact of NLOS on location estimation accuracy of TOF-based UWB RTLS when only four anchors are utilized. The proposed mitigation methodology provides an accuracy improvement of more than 60%, which is significant especially for applications such as safety where accuracy is paramount on construction sites.

Many construction project owners group their projects into different work types to facilitate effective project management decisions. This categorization significantly helps owner agencies narrow down and analyze the historical projects of a similar type to extract meaningful patterns that can support various project management decisions. However, many owners, particularly state highway agencies, typically rely on project engineers’ subjective judgments to classify a new project or do not even have systematic project classification criteria. In addition, many projects are a mixture of work types with varying proportions. A systematic and objective process of classifying projects is desirable to generate a more accurate and less disputable categorization of projects, thereby improving project management decision-making. This study proposes a natural language processing-based model for grouping projects with similar work components and portions into the same group. For a specific project, work items’ descriptions and cost composition are the key input variables of the model. Bid tabulation data from a highway agency were collected and used for model development and evaluation.

The recent COVID-19 pandemic has impacted the way how everyday professional work is performed. Organizations have devised innovative ways to work remotely, staying home, or in secluded offices. One of the technologies that many industries are relying on is visualization for remote work. Professionals in the Architecture, Engineering, and Construction (AEC) industry are applying technology solutions to facilitate remote work, making the work site less crowded. Visualization technologies (VT)—video cameras, extended Reality, and Unmanned Aerial Vehicle’s ability to monitor construction activities—have the potential to help the AEC industry in remote work. However, it is unknown if the use of visualization technologies in the AEC industry has increased since the beginning of the COVID-19 pandemic and what tools and technologies are used the most. To understand this, the authors designed and implemented an industry-wide survey to analyze the state of VT use and the barriers to its implementation. The study found out that although remote work had increased significantly since the onset of the COVID-19 pandemic, the rate at which people utilized VT for remote work had not increased at the same pace, which indicates that the AEC industry is yet to utilize the use of visualization technologies to make remote work easier and more efficient.

Risk assessment is an important part of risk management in construction projects as it involves the identification, evaluation, and prioritization of potential risk events. This enables project teams to properly mitigate them. With the intent of getting risk input from multiple stakeholders, different systems have been developed to support collaborative risk assessment. However, these systems often face challenges such as no data confidentiality, poor data security, lack of transparency, and absence of traceability and auditability for the team members. Blockchain is an emergent decentralized digital technology that can provide solutions to overcome such deficiencies of centralized systems. In this study, the need for a novel blockchain-based methodology for collaborative risk assessment is analyzed. Multiple traits of blockchain technology have been found through the literature review. From these multiple traits, the key functional elements for the methodology have been identified and discussed in further detail. The research points out some of the limitations at the current state so that future work can be conducted to build a methodology that can be even more beneficial for the risk management of construction projects.

Electronic ticketing (E-ticketing) is the process of digital ticket delivery, and it can be combined with different technologies to produce a wide variety of benefits in the construction transportation industry. Many transportation departments and agencies are not willing to adopt this technology due to some limitations imposed by technology or organization policies. Therefore, the aim of this paper is to perform analysis of E-Ticketing adoption and its changes during the Covid-19 pandemic, evaluate the benefits and limitations of the technology, and determine the technology integration with other available automation platforms in construction industry. The outcome of study will help in validating that the primary uses of E-ticketing are not safety and documentation; it is also a tool for rendering automated inspections and reducing the cost and increasing the overall quality of transportation construction projects during and after the pandemic. This study incorporated an extensive literature review on policies, memos, guidelines, news articles, and research publications pertaining to the limitations, benefits, and key technologies that can be integrated into E-ticketing platforms. A qualitative analysis of recent technological trends indicated that intelligent compaction, drone surveying, infrared sensors, advanced imaging, and thermal profilers can be integrated into an E-ticketing platform to extract better results that are paramount to a project’s success. Four findings from this study will assist project stakeholders in developing guidelines and policies, reducing costs, and increasing the overall safety of the project. The main contribution of the paper is a framework for the platform that combines E-ticketing and other emerging technologies to facilitate efficient automated inspections, billing, and telematics in the construction transportation industry.

Traffic control devices in road construction zones play important roles, which (1) provide critical traffic-related information for the drivers, (2) prevent potential crashes near work zones, and (3) protect work crews’ safety. Due to the number of devices in each site, transportation agencies have faced challenges in timely and frequently inspecting traffic control devices, including temporary devices. Deep learning applications can support these inspection processes. The first step of the inspection using deep learning is recognizing traffic control devices in the work zone. This study collected road images using vehicle-mounted cameras from various illuminance and weather conditions. Then, the study (1) labeled eight classes of temporary traffic control devices (TTCDs), (2) modified and trained a machine-learning model using the YOLOv3 algorithm, and (3) tested the detection outcomes of various TTCDs. The key finding shows that the proposed model recognized more than 98% of the temporary traffic signs correctly and approximately 81% of temporary traffic control devices correctly. The construction barricade had the lowest mean Average Precision (50%) out of eight classes. The outcomes can be used as the first step of autonomous safety inspections for road construction projects.

With the advancement of the Internet of Things (IoT) technologies, smart homes have promoted human-building interaction and sustainability in occupants’ daily life. The rise of voice-based AI-powered virtual assistants has brought new potentials to provide occupants with a convenient and intuitive interface to interact with smart homes. Aiming at enhancing the human-building bi-directional communication, voice-based proactive virtual assistants integrated with smart home ecosystems—that is, Smart Home Assistants (SHAs)—were investigated in this study. A comprehensive data collection was conducted through an online experiment, in which 307 valid questionnaire responses with participants’ demographic background information and their feedback to proactive SHAs were collected. Occupants’ perception of the proactive SHAs was evaluated among different groups of users with various demographic backgrounds. Five occupants’ socio-demographic background features were identified to have a significant impact on their acceptance level to the energy-saving suggestions by proactive SHAs, including gender, age, education level, employment status, and the number of occupants in a residence. By utilizing these demographic features, ensemble learning models can predict occupants’ perception of proactive SHAs with good performance (accuracy in the range of 0.69–0.75). Findings in this study will provide a valuable reference for academic researchers and industry practitioners in the development of personalized proactive smart home systems for human-building interaction.

The construction industry is complex and dynamic by nature. Such dynamic nature of the industry requires construction companies to continuously monitor and track a variety of external factors, ranging from economic data to safety-related measures, in order to respond timely to the changing business climate and thus make effective strategic decisions. However, there is a lack of approach to streamline the abundant amount of information derived from multiple factors and to interpret the changing nature for informed policy- and decision-making in the construction industry. To address this gap, this study proposes a system-of-systems (SoS) framework to model the complex construction industry as a hierarchical system and to analyze the evolutionary and emergent phenomena, which significantly contribute to the changing and uncertain nature of the industry. The SoS framework employs a composite index as the means to broadly utilize the information derived from multiple factors and communicate the observed changes to decision-makers. To maximize the benefit of the proposed framework, this study also creates a mock dashboard to guide the use of the index for companies’ strategic decision-making.

The recent growth in the popularity of interconnected wearable devices with sensing, computing, and communication capability has been very rapid, paving the way for a new category of technology called Wearable Internet of Things (WIoT). Although more prevalent in other industries (such as healthcare, sports, and fitness), the use of IoT-based wearable devices (i.e., WIoT devices) has only found limited applications in the construction industry for safety management. Given the recent development, there is a pressing need to explore the decision-making process to evaluate and select the most appropriate WIoT devices for safety on construction sites. Thus, the purpose of this study is to present the results of an early pilot program to implement a decision-making framework to select WIoT for safety on construction sites. This pilot study followed a descriptive quantitative research methodology with a case study design. The results of this study provide the first attempt to help construction companies in their decision-making process regarding field implementation of WIoT devices.

Bridges and other infrastructure assets require periodic condition assessments. The recent Infrastructure Report Card prepared by the ASCE found the national grade for infrastructure remains at a D+. Most of the steel bridges built in the USA are old enough to require frequent inspection and maintenance. The bridge inspection is performed manually by completing a checklist to investigate the condition of the main elements of the bridge structure as well as the shape deformation and structural settlement. However, rarely bridge inspection is completed by constructing a 3D model of the existing structure. The increasing availability of terrestrial point cloud presents new opportunities for rapid, intuitive, and accurate bridge assessment. In this paper, we analyzed the condition of an existing steel bridge over eight months by scanning the bridge at different dates and constructing 3D point cloud models. Bridge main features were extracted from the 3D point cloud and analyzed for structural deformation, bending, and rust. We conducted a holistic unit analysis by comparing the three data sets using CloudCompare3D, an open source for point cloud, and mesh processing. The results showed the importance of constructing a 3D point cloud for a better inspection of the existing capital assets. However, the skills and computing requirement for completing 3D point cloud inspection requires a certain investment in staff training and technology by transportation agencies.

With commercial and residential buildings accounting for approximately 40% of the energy and 70% of the electricity consumption in the United States, there are substantial opportunities to improve energy efficiency in these buildings. Similarly, buildings also account for the large majority of electricity demand, particularly during peak use hours. As the electric grid becomes increasingly supported by renewable energy, buildings are ideal for supporting demand-side management, allowing for the electricity demand to meet the variable levels of electricity supply. Integrated controls of various building energy system components, including HVAC (Heating Ventilation and Air Conditioning), lighting, and shading devices, combined with advanced sensor and control technologies, can help to optimize system operations. This research aims to study the impact of integrated HVAC, lighting, and shading device controls, to estimate energy and demand saving in typical small office buildings in the US. This is achieved through a multi-step modeling process, including daylight simulation using Radiance to evaluate available daylight for each zone, then EnergyPlus to develop and implement various controls and estimate energy and demand savings using the Radiance results as input. The result of this work provides insights for a variety of stakeholders in the building, utility and grid operator industries, and quantifies the potential benefit of integrated systems.

E-Construction is a paperless system for tracking and managing contract documents and project documentation related to construction. E-Ticketing, a component of e-Construction, uses a combination of software and hardware to track and document material delivery and quantities on construction projects. E-Ticketing has been piloted in various state Departments of Transportation but has not been explored in the State of New Mexico, where there are large but sparsely populated areas with little internet access. As recognized by the New Mexico Department of Transportation (NMDOT), the implementation of an e-ticketing system in New Mexico will enhance worker safety, improve productivity, and increase project efficiency. Subsequently, a pilot study has been initiated by a research group from the University of New Mexico to test, identify, and implement an e-Ticketing system to replace the current paper-based ticket system used by NMDOT. This pilot study will also measure the performance impact and efficiency of implementation with respect to the currently used paper-based ticket system, which will lead to a better understanding of technology-based solutions to roadway construction. As part of the pilot study, this sub-study was focused on interviewing current NMDOT engineers and other stakeholders to document current practices for asphalt materials tracking and discuss the challenges and opportunities that will arise from implementing an e-Ticketing system in rural areas with poor internet access.

Construction is a high-risk industry characterized by many factors that are potentially hazardous to workers. The continuous monitoring of unsafe behaviors and conditions has been identified as a proactive and active means of eliminating potential safety and health hazards on construction sites. Digital technologies combined with deep learning and computer vision can be applied to create a robust learning environment and enhance the analysis of safety and health data for generating insights needed to improve safety and health performance. This study provides a framework that implements the use of Unmanned Aerial Vehicles (UAVs) and deep learning (DL) for worker safety and health activity monitoring to improve safety performance on construction sites. The findings of this study provide useful information with which UAVs and DL can be effectively deployed for the reduction and prevention of injuries, illnesses, and accidents on construction sites. It is expected that the application of UAVs and DL for worker safety and health activity monitoring can improve decision-making in safety management because rapid collection and analysis of safety and health data would enable safety personnel to take faster preventive actions to avoid future accidents.

To achieve a green transition in construction, all parts of the building lifecycle should be examined for potential emission reductions. Historically, much effort has been placed on operational energy of facilities. Recently, the focus has been supplemented with an embodied energy focus, looking at the production of materials and the circular economy aspect of reuse and recycling. However, the construction process itself has, until now, attained little focus in the green transition. At the same time, construction is inefficient and wasteful, as research shows that only 33% of working time is value-adding, resulting in longer construction duration. Longer duration requires more operation of machines, transportation, and daily running of construction sites. These non-value-adding operations require energy and emit unnecessary emissions. Consequently, today’s tacit acceptance of time and cost overruns is effectively the same as an industry’s acceptance of unnecessarily large emissions. To reduce construction emissions from equipment-related operations, a closer to real-time monitoring of construction equipment’s efficiency and their emissions is needed. This research’s vision aims to minimize emissions by optimizing the site layout and improving planning, hence reducing construction machine working hours. Our method incorporates real-time sensory data (mainly position, NOx, and particle emissions from internal combustion engines) and information modeling to achieve this novel agenda. This paper presents a small-scale sensor technology called SEMS (Simplified Emissions Measurement System) capable of rapidly measuring such emission data. Preliminary findings from experimental testing identify potentials and future methodological considerations how such data can be applied in information management and decision making.

This study proposes an efficient management direction for Korean construction disaster cases from KOSHA (Korea Occupational Safety & Health Agency) through a text data classification model based on CNN Algorithm. Five classes were defined to classify construction accidents: fall, electric shock, flying object, collapse, and narrowness. After the initial test, the classification accuracy of fall disasters was relatively high, while other types were mainly classified as fall disasters. These results showed that (1) specific accident-causing behavior, (2) similar sentence structure, and (3) complex accidents corresponding to complex types affect the model accuracy. Two improvement experiments were then conducted: (1) reclassification, and (2) elimination of complex accidents. With complex accidents eliminated, the classification performance improved by 185.7%. This result indicated that the multicollinearity of complex accidents was solved during elimination. In conclusion, this study suggests the necessity to manage complex accidents independently while preparing a system to describe the situation of future accidents in detail.

According to 2021 ASCE Infrastructure Report Card, 7.5% of the nation’s more than 617,000 bridges were found to be structurally deficient. About 42% of all bridges across the country need to be replaced, widened, or rehabilitated. Although bridges are regularly inspected every two years, the current bridge program views a bridge as an isolated entity for the condition measurements in a state. While the National Bridge Inventory has been operated for over 25 years, the database relies mostly on the inspection report data. Currently, no integrated database system is available to embrace the extensive information of a bridge from historical and transactional data in economic, physical, and social sources. This paper introduces new approaches of network and clustering analyses using text mining algorithms to gather multi-source heterogeneous data generated by a specific event (e.g., planning, design, inspection, and repair) and to interpret the complex data. Specifically, we performed a text mining tool to extract features (e.g., girder beam and bearing devices) from bridge inspection reports. The features were cleaned out by imputation and filtering out features with large missing values. Given the semi-structured features, further analyses, such as network analysis and clustering, were performed. This automatic data processing system shows the potential to extract features of interest and efficiently generate big data from historical and transactional data to establish the federated data analysis for bridge management. The federated data makes a framework from similar types of bridges by identifying interactions, interdependences, and interrelationships between them as pathognomonic signs or symptoms practiced in medical practice.

An efficient parking monitoring system can significantly improve parking facilities’ operation at high-traffic locations such as transport hubs, cities, and schools. Inefficient parking availability forces drivers to circle around congested areas searching for parking spots, thus adding to the existing congestion. Agencies managing parking facilities face challenges in controlling emissions and providing an acceptable service besides congestion. Recent developments indicate that Real-Time Parking Guidance (RTPG) systems could be a solution to these problems. We provide findings from a study conducted at California State University Fullerton’s (CSUF’s) parking lot to determine the effectiveness of an RTPG system. The results show that real-time RTPG systems reduce time to find parking spots and can be a useful tool in reducing overall congestion. The study also finds that the effectiveness of RTPG system is a function of market penetration. The RPTG system also reduces taxiing in parking lots, provides tool for congestion pricing, and better parking management.

With the trend of increasing automation and digitization, characteristic of the paradigm shift referred to broadly as Industry 4.0, the off-site construction sector is racing to keep pace with other industrial sectors in technological adoption and transformation. Although the rate of adoption is still relatively low, those off-site construction enterprises often proceed directly to investing in technologies before carefully identifying their needs and thoroughly analyzing and understanding their current state of operations. In this context, this study presents a six-step framework that will support decision makers in the following three areas: (1) understanding and evaluating the current state of their business operations; (2) identifying whether a new technology would be valuable to their business; and (3) properly implementing the technology, if needed. The framework was developed using a design science research approach and inductive reasoning. We briefly present preliminary observations based on an ongoing application of the proposed framework at an off-site construction company.

Conversational Artificial Intelligence (AI) systems have become more and more popular to provide information support for human daily life. However, the construction industry still lags other industries in developing a conversational AI system to support construction activities. The developed intelligent Building Information Spoken Dialogue System (iBISDS) is a conversational AI system that provides a speech-based virtual assistant for construction personnel with considerable building information to support construction activities. The iBISDS enables construction personnel to use flexible spoken natural language queries instead of detecting exact keywords. To build an iBISDS, it is necessary to understand the intents of natural language queries for building information. This research aims to develop a named entity recognition (NER) algorithm for iBISDS to recognize and classify keywords within natural language queries. A dataset with 2,008 building information-related natural language queries was developed and manually annotated for training and testing. A Neural Network (NN) deep learning method was trained to recognize named entities within natural language queries. After training, the developed NER algorithm was applied to the testing dataset which achieved a precision of 99.74, a recall of 99.87, and an F1-score of 99.81. The preliminary result indicated that the developed NER algorithm can recognize named entities within the natural language queries accurately. This research will facilitate the further development of conversational AI systems in the construction industry.

Construction of transportation infrastructure projects such as roadways generally involves a wide range of processes, participants, and machinery to successfully achieve the goal of a project. A systematic way of obtaining real-time data for effective and efficient management and communication is critically required for optimally managing logistics including ongoing construction activities and required resources. Recently, several studies have explored the concept of digital twin (DT) integrated with artificial intelligence (AI) technologies for real-time project organization and predictive analyses. This study proposes a new framework that integrates deep reinforcement learning (DRL) into an audio-based DT system for autonomous monitoring and management of paving activities in a roadway construction site. The objective of this study is to develop a framework that autonomously optimizes work cycles and resource allocation according to real-time site activities and data collected from diverse sensors. The framework obtained real-time data from the construction site through sensors and processed it on a cloud-based platform using DRL to autonomously optimize the work cycle and manage the resources involved in paving activity. This system is expected to be highly beneficial for reducing manual labor, idle time, human error, and wastage of resources in monitoring a gigantic-sized transportation construction site and managing complicated logistics.

Robotic automation of construction tasks is a growing area of research. For robots to successfully operate in a construction environment, sensing technology must be developed which allows for accurate detection of site geometry in a wide range of conditions. Much of the existing body of research on computer vision systems for construction automation focuses on pick-and-place operations such as stacking blocks or placing masonry elements. Very little research has focused on framing and related tasks. The research presented here aims to address this gap by designing and implementing computer vision algorithms for detection and measurement of building framing elements and testing those algorithms using realistic framing structures. These algorithms allow for a stationary RGB-D camera to accurately detect, identify, and measure the geometry of framing elements in a construction environment and match the detected geometry to provided building information modeling (BIM) data. The algorithms reduce identified framing elements to a simplified 3D geometric model, which allows for robust and accurate measurement and comparison with BIM data. This data can then be used to direct operations of construction robotic systems or other machines/equipment. The proposed algorithms were tested in a laboratory setting using an Intel RealSense D455 RGB-D camera, and initial results indicate that the system is capable of measuring the geometry of timber-frame structures with accuracy on the order of a few centimeters.

In construction applications, a robot is commonly seen a semi-automated tool or a piece of equipment that assists with specialized work tasks. However, as robots become more technically capable and widely available, they may be seen more as a teammate or co-worker that collaborates with human crews. Using a survey questionnaire, 63 project managers from two national construction management firms in the US were shown videos of three different applications of robotic systems, each exhibiting different characteristics, and were asked to share their perceptions of the robot. Through a between and across group comparison of their responses, we found that a robot was more likely to be seen as a teammate when its movement was less unpredictable, it was seen as more productive than human workers, it was considered durable, it remained constantly active, it took its surroundings into account before moving, it worked well alongside human workers, it was not unreliable, and it made the task more predictable. These findings identify clear challenges for human-robot teaming and the design of robotic systems for construction applications.

Exoskeleton as a human augmentation technology has shown a great potential for transforming the future civil engineering operations. However, the inappropriate use of exoskeleton could cause injuries and damages if the user is not well-trained. An effective procedural and operational training will make users more aware of the capabilities, restrictions, and risks associated with exoskeleton in civil engineering operations. At present, the low availability and high cost of exoskeleton systems make hands-on training less feasible. In addition, different designs of exoskeleton correspond with different activation procedures, muscular engagement, and motion boundaries, posing further challenges to exoskeleton training. We propose a “sensation transfer” approach that migrates the physical experience of wearing a real exoskeleton system to first-time users via a passive haptic system in an immersive virtual environment. The body motion and muscular engagement data of 15 experienced exoskeleton users were recorded and replayed in a virtual reality environment. Then, a set of haptic devices on key parts of the body (shoulders, elbows, hands, and waist) generate different patterns of haptic cues depending on the trainees’ accuracy of mimicking the actions. The sensation transfer method will enhance the haptic learning experience and therefore accelerate the training.

Amid the rapid development of robotic technologies and artificial intelligence, Human Robot Collaboration (HRC) has gained momentum in a variety of civil engineering applications. However, with robots only obeying redefined algorithms without human-like intelligence, there has not been a widely accepted method that enables a complete integration of human robot team in knowledge-and-experience-based tasks, such as building inspection. To enhance the efficiency of inspection tasks, a deeper insight must be gained into the advantages and limitations of human intelligence and robotic algorithms, respectively, and how these two can be seamlessly integrated. As the first step, in this paper we designed an experiment to compare human and robot performance in a building inspection task. A quadrupedal robot is simulated in ROS (Robot Operating System) Gazebo, which automatically navigate in and scan the buildings with SLAM (simultaneous localization and mapping) and RRT (rapidly exploring random tree) algorithms, while human experts finishing the same inspection task in Virtual Reality. The total identified structural defaults, inspection time, and routes are recorded and compared. The result shows that there is an apparent pattern difference of human route plans with considerably better accuracy and efficiency in building inspection.

Human-robot collaboration has gained popularity in various civil engineering applications. The key to a successful human-robot collaboration is the design of an intelligent robot system that is aware of human intents and can predict human motions. Despite the advances in human intent prediction in the context of human-robot collaboration, challenges still present. Most intelligent systems can only predict human motions based on the aggregated data from different human subjects. A method that is capable of capturing the changing characteristics of human motions and predict human motions in dynamic and open workplaces is needed. This paper proposes an innovative analytical method that predicts human motions using the Light Gradient Boosting Machine (LightGBM) with incremental learning. A virtual reality-based human subject experiment (n = 120) was performed to collect the gaze tracking data and the corresponding two-hands motion data in a pipe maintenance task. First, the relationship between the gaze focus and the hand motion is explored via the symbolic aggregate approximation (SAX) to identify the latency between a person’s gaze focus direction and the hand motions. Then, the continuous time-series data of gaze focus is used to predict the motions of the hand, with eye-hand delay adjustments incorporated. The proposed method can significantly improve the accuracy of human motion prediction in a complex pipe maintenance task, and thus benefit a better design of collaborative robotic systems.

With construction sites being dynamic and unstructured in nature, the safety of workers at a job site has always been a major concern for project managers, who often assign a significant portion of project resources and manpower for maintaining safety protocols. The current manual methods used for safety surveillance at a job site are not only laborious but also time-consuming and prone to human error, leading to numerous accidents at sites each year. This study proposes a new approach that employs a digital replica of a construction safety surveillance system capable of providing real-time safety analyses and predictions based on site conditions. A digital twin system developed in this study allows for real-time communication with a construction site facilitating safety managers in a decision-making process. The system encompasses three primary resources: (1) 4D BIM models, (2) cloud computing and database platforms, and (3) real-time field data captured by sensors and processed by artificial intelligence techniques. The proposed system is expected to improve real-time safety surveillance through identification of various site conditions and prediction of possible safety concerns.

Welding technique plays a pivotal role in many industries, such as construction, automobile manufacturing, and nuclear power plants (NPPs). However, the shortage of skilled welding workers is still controversial due to the severe working environment and conditions. Therefore, to conserve human labor and improve manufacturing efficiency, an automated welding process is necessary. Also, welding efficiency and quality are vital indicators requiring attention for automatic welding. Notably, in NPPs, minor welding defects can occur serious safety issues. Therefore, our research’s ultimate goal is to develop an automatic welding system to improve welding quality and manufacturing efficiency using visual sensors [e.g., a camera and light detection and ranging (LiDAR)], a robotic arm, and a welding machine. As the first step, this paper presents a method for automatically detecting different welding joints in real-time. Then, the different target joints are trained using a deep learning algorithm and detected by the camera. The results demonstrate the accuracy and effectiveness of the proposed method.

Construction progress monitoring is essential to construction processes. Automation of progress data capture during the construction process can address existing limitations and inefficiencies of the manual processes to enable systematic recording of construction progress. To achieve that, legged robots with the adaptability and mobility to traverse a variety of terrains on construction sites can be applied to automate on-site image capture. The goal of this research is to identify main criteria in effectiveness evaluation of technological performance of the quadruped robot in construction progress monitoring. This study investigates the implementation of legged robots in automating progress monitoring of construction through 360° image capture using Spot robot by Boston Dynamics. The study conducts a series of weekly experiments on two live construction sites to investigate robot-enabled reality capture in progress monitoring of construction projects. Core technological indicators and factors regarding economic justification and organizational performance of the new technology are explained.

The implementation of robots in the construction industry is poised to increase and improve. Human-robot collaboration (HRC) leverages dexterous workers and tireless robots to execute various complicated construction operations. However, physical aspects of HRC may create timing and collision hazards at construction sites. One of the main issues is that a construction robot cannot understand a worker’s moving intention. In this regard, the worker’s posture is an essential indicator for avoiding collisions. Therefore, to allow workers and robots to collaborate safely, robots need to be empowered to assess workers’ postures and movement intentions. To address this need, this study leverages computer vision techniques to enable collaborative robots to estimate worker positions and poses. The proposed approach employs a multi-stage Convolutional Neural Network to first identify workers’ joints. Subsequently, the network will assemble the results into full-body postures, using the Part Affinity Fields technique, to allow the robot to understand worker poses. To examine the feasibility of this approach, an experiment was designed in which four subjects were required to perform bricklaying tasks in collaboration with a masonry robot. The results reveal that the proposed approach enables robots to estimate subjects’ postures with a 63.3% precision using a metric of the percentage of correct key points. The findings pave the way to enable collaborative robots to understand workers’ intentions when moving, supporting safe HRC at construction sites.

The use of construction robots to facilitate labor-intensive tasks is increasing in many jobsites. However, human intelligence and expertise are yet crucial to support such robots in accomplishing their tasks. Therefore, it is essential that human-robot teams entail a human-in-the-loop approach for integrating humans into the system. To reliably design such systems, it is critical to understand workers’ confidence in such models, particularly in hazardous and complex construction jobsites. In this context, assessing workers’ trust can shed light on the efficient adoption of robots. To that end, this study investigates the impact of the human-in-the-loop approach on human-robot trust. Accordingly, a comparative experiment was conducted in an immersive virtual environment in which participants performed brick-laying tasks in collaboration with robots through two different approaches: a human-in-the-loop approach and a human-out-the-loop approach. The participants’ trust in the robot was measured using the Trust Perception Scale-HRI. The Wilcoxon signed-rank test revealed that the human-in-the-loop approach could lead to significantly higher levels of trust. The findings of this study provide insights into the effective adoption of robots in construction sites. Thus, this study can open new doors to facilitating the imminent paradigm shift toward the human-in-the-loop approach in construction robotics.

To allow full automation of building code compliance checking with different building design models and codes/regulations, input building design models need to be automatically validated. Automated architecture, engineering, and construction (AEC) object identification with high accuracy is essential for such validation. For example, in order to check egress requirements, exits of a building (and their presence or absence) need to be identified automatically through object identification. To address that, the authors propose a new AEC object identification algorithm that can identify needed code checking concepts from building design models based on the invariant signatures of AEC objects, which consisted of Cartesian points-based geometry, relative location and orientation, and material mechanical properties. Building design models in industry foundation classes (IFC) format are processed into invariant signatures, which can fully represent the model data and convert them into computable representations to support automated compliance reasoning. A systematic implementation of the above invariant signatures-based object identification algorithm can be used to automatically conduct building design model validation for code compliance checking preparation. An experimental testing on Chapters 4 and 8 of the International Building Code 2015 and a convenience store design model showed the model validation using the proposed identification algorithms successfully validated ceiling and interior door concepts. Comparing to the manual validation used in current practice, this new object identification algorithm is more efficient in supporting model validation for automated building code compliance checking.

Deviation between project baseline and actual performance is a common challenge in the construction industry. Traditional monitoring and tracking of construction project status is labour-intensive, error-prone, time-consuming, and costly. Despite research and technological developments, the optimal level of efficiency in automated construction project measurement has not been met. Not achieving the desired automation efficiency is caused by a significant gap between using automated progress measurements and traditional project measurement systems. This requires a structured framework for integrating automated data acquisition and progress measurement through baseline plans. This paper proposes a systematic, generic framework to integrate baseline development with 5D Building Information Modeling (BIM) and automated project monitoring and measurement. The integration process involves using work packages and S-curve to automatically measure progress values. Also, interviews with five industry experts were conducted to assure the applicability and feasibility of the proposed framework. Developing this framework will help the construction industry to maximize the benefits of implementing automated project measurements and minimize the problems of traditional measurement methods.

The construction industry is one of the largest economic sectors in the United States, contributing to 4.1% of the nation’s gross domestic product. Despite its economic importance, the construction industry is plagued with labor shortages, unsafe working conditions, stagnant productivity, and inefficiencies resulting from industry fragmentation. With the recent advancements in mechatronics, artificial intelligence, and sensing systems, the implementation of robotic technologies appears poised to revolutionize construction. However, the construction industry is still falling short of the widespread adoption of robotic technologies prevalent in other sectors. The present study investigates the contributory factors of this gap, delineates the potential barriers, and provides recommendations to lift these barriers and expedite the transition process. To determine the limiting factors to the widespread implementation of construction robots, a set of questionnaires were distributed to ten major construction companies. The questionnaires were developed by the experts in the related fields and validated by industry professionals. The statistical results of the questionnaires revealed the factors that impede the large-scale robotic adoptions, including high-capital investment and job security as primary factors. The findings of this research can provide the necessary insight to understanding and potentially lifting obstacles to the ubiquitous implementation of robotic technologies in the construction industry.

The architecture, engineering, and construction industry (AEC) is embracing digitization in the design, construction, and operation of built assets with the growing prominence of information technologies, such as building information modeling (BIM), internet of things (IoT), and artificial intelligence (AI). In this context, plenty of research efforts have been dedicated to digital twin (DT) applications. This research synthesizes state-of-art on DT in the AEC industry through bibliometric analysis, aiming to identify the research trends, challenges, and knowledge gaps in this growing area. A total of 75 publications regarding DT was identified and retrieved from Scopus. Then, VOSviewer was used for bibliometric analysis, including (1) keyword co-occurrence, and (2) citation analysis of selected publications. The identified research clusters and most-cited publications were discussed to clarify research trends and future needs. The findings revealed that future research should be directed to (1) data interoperability, (2) AIoT, and (3) AI. Moreover, extra research efforts should also be given to the DT applications during the design and construction phases of construction projects. This research contributes to the body of knowledge by quantitatively exposing research trends and needs for DT in the AEC industry.

During design coordination, identified relevant clashes are discussed in detail, and design changes and modifications are made to resolve the clashes prior to the construction. Currently, clash resolution is a slow manual process. Recent research focused on using supervised machine learning to automate the clash resolution process shows potential results to improve the efficiency and effectiveness of clash resolution. However, the model trained using supervised learning is limited in its effectiveness by the quality of training data provided. To overcome this limitation, the paper proposes a machine learning method that integrates supervised and reinforcement learning. In the proposed model, supervised learning will be used to establish the initial relationship between the clash information and the clash resolution decision. This relationship will act as pre-training for reinforcement learning, which will improve the relationship with subsequent iterations of the learning process, generating a more effective clash resolution policy than the initial relationship.

Utilization of technology has been proven as an efficient approach to construction inspection. Although a wide variety of technologies are available to support construction inspection, there is a lack of understanding of how mobile technologies can be effectively used for construction inspection. This study aimed at exploring the current practices of using mobile technologies to inspect highway construction projects. The main mobile technologies investigated in this study include tablet computers/smart phones, handheld data collectors, automated machine guidance, and virtual reality/augmented reality. A national survey was conducted to evaluate the current practices and identify benefits and challenges of using mobile technologies for highway inspections. Based on the responses from 42 state Departments of Transportation (DOTs), the result of this study showed that tablet computers/smart phones are the most common technologies used for highway construction inspection. DOTs typically use mobile technologies for the following activities: monitoring construction progress, earthwork inspection and quantities, and documentation of work completed for payment. The main barriers for adopting mobile technologies for construction inspection are the lack of reliable internet connection in remote locations; lack of training; knowledge and skills; and cost issues. This study contributes to the body-of-knowledge and the construction industry by exploring the current practices and identifying challenges in implementing mobile technologies for highway construction inspection.

The objective of this research is to propose a highway construction progress monitoring system using Unmanned Aerial System (UAS) technologies, computer vision techniques, and data analytic tools to automatically: (1) estimate the construction operation production, and (2) monitor the progress relative to an as-planned 3D model. The motivation of this research is to assist project managers in automatically and efficiently monitoring construction project progress during the construction phase. The proposed progress monitoring process consists of three steps of (1) collection of orthophotographs of ongoing construction operations using UAS, (2) use of image processing to support automated feature extraction, and (3) development of as-planned vs. as-built progress estimates based on a comparison of as-planned model and as-built reports. The method is implemented and tested for pavement removal and concrete pavement construction operations for the Michigan Department of Transportation (MDOT) I-496 corridor reconstruction project. The main contribution of this paper is to further automation approaches into highway construction inspection workflows.

The exponential growth of on-site visual data and the advent of computer vision techniques have created a unique opportunity to improve automated construction progress monitoring methods. To date, the state-of-the-art vision-based methods are capable of reporting the progress of a building element in terms of binary function. However, for better schedule control and micro-level monitoring, it is necessary to report the partial completion of tasks associated with an element. This research proposes a novel approach for computing and reporting the partial progress of tasks in terms of completion percentage using the on-site visual data, 4D BIM, and deep-learning-based computer vision algorithms. The approach leverages geometry modeling and appearance detection to automatically calculate the percentage completion of tasks associated with each element. The proposed approach is applied to a building construction project, and the preliminary results demonstrate its applicability to generate completion percentage per task in the lookahead schedule for accurate daily progress report generation.

Construction, the last major analog and craft manufacturing industry, is showing early signs of industrialization through the emergence of new robotic and automated systems that can perform construction tasks in situ. While much is understood about the technical and economic challenges to be overcome for widespread adoption of robotics, less is known about the human barriers to adoption, and much less is summarized. Considering the amount of human cooperation required by existing robotic applications, a comprehensive review of barriers that are cognitive or perceptual in nature using a systematic literature assessment methodology is warranted. However, such a review is not straightforward to design. While matters of cognition and perception as pertinent to construction and automation may be queried directly from the literature, there is no certainty that a review based on directly querying abstract phenomena (i.e., perception) could be comprehensive. Thus, systematically reviewing this topic calls for a robust methodology for the design of database queries. In this paper, we perform text analysis with the quanteda package for R in order to (1) understand the language composition of an initial review corpus, and (2) with that understanding design further queries to capture additional articles otherwise not possible through standard query design. Findings indicate that performing text analysis on a systematic review design can produce valuable insight into a review corpus and inform queries that capture additional unique literature relevant to the review.

Modular construction offers an effective solution to schedule and quality problems in construction. However, if a defect on a module is identified on-site, it can easily cause problems and rework, leading to additional delays and cost overrun due to long lead time. Therefore, quality assessment of modular components is very important. There were many efforts to leverage 3D laser scanners (i.e., terrestrial) for inspection of building structures. These studies aimed at optimizing scanning locations while ensuring detailed scanning of the whole buildings. However, there was little to no research on leveraging 3D laser scanners for modular components that require scanning from the surrounding views. Therefore, this paper proposes a scan planning method that generates and optimizes scan locations for prefabricated components. The presented method creates the geometric relationship between the 3D BIM/CAD and the scanner’s properties (i.e., angular step, range, and field of view) to assess visibility and quality while evaluating best and optimal scan positions. The experimental results show that the proposed method can generate scan positions that meet the user requirement to the extent scanning is geometrically possible.

Building energy retrofit has a significant role in achieving clean energy goals in the United States, creating a vast construction market. The initial step to improving energy efficiency is a detailed building inspection. However, building inspections are often time-consuming and not scalable. Therefore, the application of drone thermography, which can be used to create digital geometry, is explored for an automated building inspection. This application has the advantages of reducing time, labor, and safety concerns for inspection activities while solving problems of inaccessible building envelope components. Consequently, we used Infrared Thermal cameras installed on an Unmanned Aerial System to collect thermal images from building sites. The proposed workflow was started with designing a drone flight path for data collection, followed by implementing computer vision algorithms to analyze the data to identify thermal anomalies of building envelope. The paper examined a campus building in Syracuse, New York. Thermal anomalies related to cases of beam to wall thermal bridge, material degradation, and air leakage were successfully detected for the case study. Additionally, the proposed approach may be integrated with building simulation toward enriching building energy assessment. This study contributes to solve problems associated with productivity in the construction sector and to improve commercial inspection efficiency.

Graphical networks have been used for generative design to create maximally planar graphs to create initial layouts. In this work, we propose the use of graphical networks to (1) implement sparsity network constraints, (2) analyze the relationships between nodes at multiple scales, and (3) create a graphical network of nodes based on the flow of resources (e.g., nurses). In this approach, we present a solution procedure for macro and micro scales in a hospital layout problem optimized for the flow of nurses (i.e., between the department and within-department scales). The theoretical formulation and solution procedures are based on user requirement data. Sparsity constraints are introduced to reduce the solution complexity of a layout that includes many nodes while also addressing a common limitation of graphical networks, feasibility. The resulting generated design conceptually aligns to hierarchical user input, bridging a gap between prior work on facility layout optimization. The approach creates a framework for future algorithms that can scale between macro and micro (in this case, Hospital-level and Department-level) to determine a generated and optimized form for conceptual planning. The implications of this work will enhance quantitative analysis in healthcare facility planning and general layout planning procedures for civil and construction engineering layout planning problems at multiple scales.

With the advancement of automation and robotic technologies, the teleoperation has been leveraged as a promising solution for human workers in a hazardous construction work environment. Since human operators and construction sites are separated in a distance, teleoperation requires a seamless human-machine interface, an intermediate medium, to communicate between humans and machines in construction sites. Several types of teleoperation interfaces including conventional joysticks, haptic devices, graphic user interfaces, auditory interfaces, and tactile interfaces have been developed to control and command construction robotics remotely. The ultimate goal of human-machine interfaces for remote operations is to make intuitive sensory channels that can provide and receive enough information while reducing the associated cognitive and physical load on human operators. Previously developed interfaces have tried to achieve such goals, but each interface still has challenges that should be assessed for enhancing the future teleoperation application in construction workplaces. This paper examines different human-machine interfaces for excavator teleoperation in terms of its on-site usability and intuitiveness. The capabilities of the interfaces for excavator teleoperation are evaluated based on their limitations and requirements. The outcome is expected to provide better understanding of teleoperation interfaces for excavators and guiding future directions for addressing underlying challenges.

“Industrialized Construction” refers to a construction method where the majority of work is performed in a climate-controlled factory. Volumetric modular construction (VMC) is a subset of industrialized construction where 2D components are prefabricated in a factory then assembled into 3D modules. 3D modules are constructed from materials identical to conventional construction using techniques that are common throughout the conventional construction industry. The 3D modules are transported to the construction site where they are stacked by a crane and connected to each other to form a building. Standardization and automation are used to reduce the need for skilled workers. Automation tools and procedures plus the ability to work 365 days per year allow VMC companies to produce buildings faster and at competitive costs when compared to conventional construction. While all of these benefits are well documented, a less documented fact is that VMC uses more material than conventional construction due to the redundant assemblies necessary to create transportable 3D volumes. This paper focuses on the dematerialization of the VMC process by comparing programmatically identical conventional and VMC multi-family buildings. Each building is five stories tall and constructed from wood. Comparison data are generated to perform framed area takeoffs for vertical and horizontal assemblies within each construction type. The summarized quantities of horizontal and vertical building elements are shown to indicate the areas of excess materials and assemblies. Further studies on design for manufacturing and assembly would reveal new construction techniques and materials that lead to the dematerialization of the VMC process.

Precise extraction of the required activities from contract documents is crucial to develop a complete and accurate construction schedule. Currently, the scheduler is required to read the complete lengthy contract package to extract activities and convert them into a structured format to prepare a schedule. The current manual methods of activities extraction for schedule development are inefficient, laborious, and time-consuming. Although a few researchers have developed models for automated information extraction from legal documents, those models are applicable to quantitative requirements only. To address this, the current study proposes an automated information retrieval system to extract the activities provided in contracts using natural language processing (NLP) methods. The study implemented the dependency parsing technique that employed the syntactic features of requirement text to extract the actors, actions, objects, and conditions specified in the requirement. The model achieved an average recall and precision of 94% and 95%, respectively, on the requirements of a real design-build project. The proposed automated activities retrieval system is expected to help the schedulers develop the construction schedules.

Drones’ recent advancements, distinctive flying capabilities, and ability to efficiently accomplish tasks have allowed these platforms to shift from passively following humans’ control commands to actively collaborating and interacting with other robots and construction personnel on jobsites. This shift has highlighted the need for advanced research on deployment challenges associated with drone usage on construction jobsites with the aim of better integrating this technology in the domain. A narrative review was conducted to propose a research roadmap and discuss the challenges and requirements of drone-related deployment limitations in construction. The paper categorizes and discusses four elements of this proposed roadmap: (1) drone technology improvement; (2) operator training and psychological effect assessment; (3) flight space and weather condition adaptation; and (4) legislative and ethical rule standardization on jobsites. The contribution of this study is to better understand the human-drone interaction requirements that would ultimately ensure safe and efficient drone adoption in construction.

Building Information Modeling (BIM) has revolutionized the way we plan, design, execute, and monitor the built environment. Various stakeholders in different compartments are brought under one umbrella to function as an integrated team in BIM concept. The construction industry is gradually evolving to a manufacturing and product delivery process by collaborating in a multi-disciplinary environment where design for manufacturability and assembly takes precedence. Off-site construction methods like pre-engineered, pre-cast, and modular technology are being adopted in many large span and high-rise building projects due to its cost effectiveness in large volume. Optimization of targeted values viz manufacturability, assembly, schedule, lifecycle cost, and sustainability becomes critical to ensure a smooth execution and maximum value generation of the project. Design for a specific target value according to the utility has been an evolutionary process. The objective of the study is to investigate the impact of BIM in managing and minimizing change orders by design optimization through visualization during the planning process in off-site construction. The study also examines the various factors influencing the smooth interface from design to construction and critically examine the bottlenecks in achieving a near perfect transformation to manufactured construction. Design for Assembly (DfA) approach gives the stakeholders more rationale inputs for monitoring and decision making in a more controlled environment of off-site construction. Design optimization in off-site construction is achieved as a byproduct of the DfA process when real time cost visualization is integrated into BIM framework. Collaboration of disciplines and optimization of resources can make the design-construction interface hitch free to ensure a near perfect BIM execution.

Modular construction has been an alternative to traditional construction processes to reduce environmental impact and construction waste as well as to deal with space constraints in highly dense urban construction sites. Furthermore, since modules are pre-fabricated in a controlled environment, modular construction has the advantage to achieve automation and optimization as compared to traditional construction. However, due to the one-of-a-type nature of construction projects, automation in construction is still in its infancy as compared to other manufacturing industries. Meanwhile, recently, advancements in technologies such as computer vision and deep learning provide opportunities to train machine intelligence to solve problems that were not possible before. In this study, we propose an approach to automatically generate high-resolution synthetic training data for scene understanding in the modular construction context. Evaluation of the approach in testbed factory settings shows that we can systematically capture and label AEC components such as walls and doors on RGB-D images as synthetic datasets for applications of supervised learning in relation to modular construction. The proposed method can provide a mechanism to feed the necessary but missing large-scale datasets to train scene understanding models in modular construction factories as modular projects and corresponding workpieces change.

The number of outpatient care facilities (e.g., primary care clinics, pediatric care centers) in the US has increased 51% in the last decade, with significant technological implementations (e.g., electronic health records, telemedicine). These trends are compelling outpatient practices to improve their flexibility in providing care to meet this rising demand, including tailoring building spaces to changing needs of users (e.g., physicians, patients). Responsive environments can perceive their users’ needs and interactions in spaces, the technologies they use, and provide mechanisms for prompt transformations of spatial features (e.g., size of space, furniture layout, and temperature). Previous research studies have developed innovative responsive environment applications such as context aware lighting systems and architectural robotics but have yet to address situations unique to outpatient clinic settings. This paper presents research findings from investigations with healthcare professionals that map out the needs for and challenges in transforming existing spaces for increased demand and technology assisted practice. This paper will also discuss the challenges and opportunities for implementing responsive environments, specifically highlighting major events of spatial transformations relevant to outpatient settings, such as detected user activities, relevant sensor and software systems, and transformed architectural elements and spatial features (e.g., walls, furniture, and lighting). Findings from this paper serve as a basis for frameworks for designing new outpatient clinic spaces and capturing requirements for responsive environments.

The construction industry is increasingly adopting off-site and prefabricated methods due to advantages offered in safety, quality, and lead time. Applying industrialized methods for plant management in offsite construction factories requires the collection of large volumes of production process data, which is a tedious task when performed manually. Recent attempts to automate this process have relied on sensor-based data collection methods which are susceptible to noise, expensive, and difficult to validate. Computer vision methods, however, enable process data collection from videos without the limitations of the other sensor-based methods. This technology has not been applied for offsite construction except in very few instances and therefore, this study proposes a novel method to reliably collect the production process data using computer vision method in near real-time from widely used surveillance cameras in offsite construction. The proposed method allows the user to annotate the workstations of interest on the video as ground truths and process these areas throughout the entire video to track the units entering and leaving stations, while continuously updating a near real-time schedule of the production line. This framework was validated by implementing on the surveillance videos of the production process of modular home manufacturing in a factory. The results consistently provided 100% accuracy, after denoising, for all the videos processed including 60 h of work for a station. The developed method enables real-time tracking of station performance, which can enable continuous improvement methods for factory management and resource allocation.

Physical 3D objects can be expressed as an abstracted format in digital space through parametrized profile. The idea of using object-based parametric modeling has been adopted in the Building Information Modeling (BIM) domain by reducing human labor efforts via digitizing conventional manual drawing-based 2D modeling into a computing-based 3D modeling. However, research efforts on the adoption of parametric modeling has been primarily placed on early design, while applications have been rare for generating an as-built BIM of existing structures. Thus, the main objective of this paper is to investigate viability of leveraging parametric modeling paradigm for generating as-built BIM from semantically segmented objects. Practical implementations are shown with a commercial parametric modeling software, Dynamo visual Application Programming Interface (API), which is adopted for constructing 3D parametric models within the Autodesk Revit BIM environment. Using the Stanford Large-Scale 3D Indoor Spaces Dataset (S3DIS) as baseline point cloud segments, a detailed set of modeling procedures is illustrated for three object categories which are grouped based on their parametric properties. Our study provides a stepwise advancement toward automating Scan-to-BIM, avoiding dependence on modeler experience for reconstructing an as-built 3D model from raw point clouds.

In the past, the construction industry has been slow to adopt new technology. There has been a rapid expansion of technologies, often referred to as Industry 4.0, to aid in the use of automation. One challenge paralleling these new technologies is implementing how a robot interprets design information, specifically information from a Building Information Model (BIM). This paper presents a method for identifying and transforming information from BIM to support robotic material placement on the construction site. This research will include a review of what information can be directly extracted from the model and what must be supplemented to the model for the robot to perform defined tasks within a construction site. The construction sites’ dynamic nature poses multiple challenges that must be addressed for the information extracted from a model to be used by a robot in daily construction operations. This research also identifies barriers and limitations based upon current practice, such as different levels of development or model content as well as needed precision within the information provided for a mobile robot to complete a defined task.

The integration of robotic systems into the manufacturing sector has greatly helped in the improvement of manufacturing quality, safety, and productivity. This prompted construction professionals to incorporate robotics into various construction activities ranging from onsite activities to offsite construction. However, a lack of understanding of the financial and technical benefits of robotics, as well as identical features of each construction project, have been barriers to utilizing the full potential of robotics in the construction industry. This study develops a decision support system for the selection of robotics in offsite construction to address the existing technical issues and cost uncertainties. The proposed study employs simulation techniques to assess the performance of multi robots in an offsite manufacturing plant. Five robotic characteristics (i.e., reach, payload, axis speed, degrees of freedom, and position repeatability) are chosen as criteria, and a scoring system is developed to rank the best-performing robots. The results show a 52% increase in annual production rate and a 21.56% overall cost reduction when multi robots were used. The proposed decision support system is intended to support construction automation and enable construction professionals to make accurate decisions when implementing robotic systems.

Automated code processing has been appearing research area for the last four decades. Hence, there are functioning prototypes that have been developed worldwide; there remains no consensus on one unified methodology for automated rule checking, and one of the significant gaps in these systems provided a black box interface for the user who can edit the existing checking rule without creating his one. This paper provides the functional prototype to check the code of the IBC emergency egress. The proposed model depended on a new automated code checker approach that enables designers to check building design during the design process and on the same design platform, “BIM Platform.” This model developed as a Dynamo model works inside Autodesk Revit. This model breaks down into three modules for checking and calculating the egress paths. The outcomes of this model are a graphical floor-plan illustrated all egress paths and simulated all exit marks for each room. Then, classify all rooms by the egress path length and export the results as a report. As a result, this tool saves a significant amount of time in processing and developing life safety drawings.

A construction 3D printing system could result in automated infrastructure development at reduced cost and time, leading to a significant boost in overall productivity. Although there has been a growing interest in using construction 3D printing for projects such as house construction, implementing this innovative technology for infrastructure development, particularly large-size bridges, has not been investigated as extensively. Bridges are complex structural systems subject to various static and dynamic loads, making automated bridge construction using construction 3D printing challenging. This study investigates the available construction 3D printing technology for bridge construction in the United States and worldwide. First, a comprehensive literature review is performed to investigate the applicability of construction 3D printing technology for infrastructure development. Then, the implemented technologies and construction methods for the 3D printed small-size bridges (i.e., pedestrian and biking bridges) in different countries are discussed. Finally, the limitations and challenges and the future directions of construction 3D printing in bridge construction are highlighted.

Previous studies have identified motivation as one of the most important factors affecting the efficiency of labor utilization in construction processes. However, there is a lack of research on simulating the impact of motivation on labor productivity to track and devise productivity improvement strategies. Fuzzy system dynamics (FSD) has been used to model labor productivity, because it captures subjective uncertainties and the dynamism of construction systems. However, FSD fails to capture complexity arising from individual components (e.g., crew members) that lead to emerging behaviors in crew motivation modeling. The main contributions of this paper are: (1) proposing a framework for combining FSD and fuzzy agent-based modeling, leading to a more comprehensive method for studying the impact of crew motivation on productivity; and (2) facilitating identification of more effective productivity improvement strategies by allowing construction stakeholders to track the dynamic relationships between motivation and labor productivity.

Image-based 3D reconstruction is a prerequisite technology, prior to establishing the as-built spatial data of the built environment for many purposes, including progress monitoring of construction sites, quality control, energy performance assessment, and structural integrity evaluation. Unmanned aerial vehicles (UAVs) are an emerging technology that allows for effective remote image acquisition through the integration of digital cameras. It has been widely applied to data surveying in various fields due to its lower cost, ease of use, and flexible access to hard-to-reach locations. While UAVs provide great potential in efficiently capturing high-quality images for 3D reconstruction, the spatial data collection process is prone to image degradation issues. The external environmental factors like wind that may lead to instability of the UAV platform, and inappropriate camera settings can cause degradation in the quality of captured images such as motion blur, overexposure, and underexposure in light. In this paper, the influence of image degradation issues on image-based 3D reconstruction is explored. Four types of commonly occurring degradation sources including motion blur, out-of-focus blur, as well as underexposure and overexposure to light are studied. The image degradation effects were artificially generated on a real scenario and the quality of 3D reconstruction was evaluated. In the experiments conducted, the intensity of degradation and percentage of degraded images were varied, for each source of degradation, to understand the effect on reconstruction quality. The results indicated motion blurred images led to significant loss in semantic information which posed the greatest impact on 3D reconstruction quality.

The emergence of Virtual Reality (VR) in recent years has incepted a new paradigm in education. In the past decade, VR has become a reliable tool in the design, development, and simulation of various scenarios in different industries. The use trend of VR along with other technological advances indicates an ascending growth, which entices more researchers and educators to explore its capabilities, strengths, and challenges and take advantage of opportunities created. VR encompasses a unique set of characteristics and features that are effectively aligned with educational purposes. VR applications provide rich content in a customizable pace and delivery method. VR graphical features intrigue users, especially the new generation of students. VR training applications also minimize the physical risks of real situation-based training tools and provide students with easy access to anytime-anywhere educational sources. Like many other advancements in the construction area, VR has been explored and utilized with a delay. However, the broad scope of construction in industry and education furnishes an appealing path for VR development. Despite this opportunity, the lack of a VR design standard or framework has led construction developers to move forward with trial and error in their design or development processes. This paper provides a background of VR in construction education followed by a framework for VR construction education applications. Also, a retrospective analysis of a design and development of a VR training application for construction safety, derived from a case study, provides lessons learned for further similar applications.

Facility owners and operators lose over $14 billion per year due to imprecise facility information. Visualization technologies such as virtual, augmented, and mixed reality (collectively XR) can make this information clearer. XR is an emerging research topic that can potentially improve facility maintenance. This study text mined keywords in literature from 2010 to 2020 to create word maps to determine the prevailing research direction and linkage between facility maintenance visualization technologies and information. The study collected 332 scholarly papers from Web of Science and visualized co-occurrence word clusters and word maps with VOSviewer. The word clusters illuminated that the research has converged on Building Information Modeling (BIM) for information and augmented reality (AR) for visualization. This underlines an opportunity to incorporate BIM as the foundation for virtualization of building information for facility maintenance. This study contributes by highlighting research efforts in facility maintenance visualization and suggests technologies to study further.

The development of robotic machines has shown the potential to promote automation and address issues in construction. One of the critical steps to make these machines work with onsite construction workers as teams is to provide a user-friendly interface to support their interactions. So far, little related work could be found in the construction domain. This paper tried to fill the gap and proposed a vision-based framework to recognize the construction workers’ hand signals to interact with construction robotic machines. The framework includes three main components: (1) detection and tracking, (2) region cropping, and (3) hand signal recognition. Specifically, it started with the visual detection and tracking of a construction signalman from a video sequence. Based on the detection and tracking results, the regions of the signalman were cropped to form a hand signal recognition queue. Then, if a hand signal was detected, a pre-trained classifier would be used to identify the meaning of the signal. The framework was tested with two construction site videos. The test results indicated the effectiveness of the proposed framework on common construction hand signal recognition.

The expected benefit of smart contract (SC) relies heavily on its completeness and adaptability. Previous studies have discussed the application of SC on construction at the conceptual level, while a method to reflect human preferences as action items to enforce schedule executability is not available for construction managers. The objectives of this research are (1) to develop a method to generalize human preference as scheduling practices, and (2) to utilize SC technology to facilitate the execution of the practice. Performance data of different labor investment and production rate arrangements were obtained through a simulation of a high-rise construction project operation. This research used a Choquet Integral-based approach to aggregate owners’ preferences on workspace overlap, manpower utilization, and project duration to customize the optimum scenarios for resource utilization. Such decision rules were bundled with payment transactions and then embedded as agreements in an SC-based scheduling platform. The SC platform developed can provide the optimized resource utilization solution under different preferences, meanwhile timely disseminate schedule information and facilitate payment transactions. The research findings will assist project managers to quickly react to unexpected changes in management preferences in manpower, space, or schedule by developing adaptable resource utilization solutions. It will also help them to enforce the execution of the plan timely using the SC scheduling platform.

Building Information Modeling (BIM) offers high-fidelity representations of the building contextual data. Internet of Things (IoT) data enhance this information set by providing real-time and recordable status from the actual operations in construction and operations. Building Automation System (BAS) groups IoT devices with logical control and management. With digital technologies like BIM, IoT, and BAS, increasing data is being created. Data silos in Architecture Engineering and Construction (AEC) industry emerged. These data silos lack common data representations, thus prevent information exchange between buildings and innovative applications. This problem has been noticed for a while. Efforts to address the interoperability issue have limitations. The isolation between BIM-based building contextual information, IoT devices’ time-series data, and BAS metadata exist. This research aims to develop a framework to facilitate information exchange between BIM-based building contextual data, IoT devices’ time-series data, and BAS metadata using Semantic Web technology. The research methodology demonstrates the framework creation. This framework lays a foundation for future studies utilizing BIM, BAS, and IoT data such as facility management, energy benchmarking, and fault detection. The framework brings higher-level connectivity, sets a foundation for data exchange between isolated information islands in the AEC industry, and brings insights for cross-domain data interlinking.

Computer vision-based over-height vehicle detection has received increasing attention for preventing vehicle-to-structure collisions. However, the prior studies usually rely on blob detection of vehicles for vehicle height measurement, which is oftentimes error-prone because the traditional blob detection methods are sensitive to shadow interference and illumination change. Also, it is practically difficult to extract the vehicle’s pixel height from the detected blobs automatically and accurately due to the lack of depth information on 2D images, especially for vehicles with irregular top shapes. Furthermore, the applications of the existing methods are still plagued by the occlusion issue (e.g., part of a vehicle is blocked by another vehicle) that commonly occurred in surveilled views. Therefore, this research proposed a new method for vehicle height estimation. In the proposed method, the vehicle instances are first segmented from images using Mask R-CNN, which can significantly eliminate the adverse effect of shadow interference and illumination change. Then, 3D bounding box is established for every surveilled vehicle with or without being partially blocked by leveraging the identified vehicle instance and the three orthogonal vanishing points in the surveilled scene. By doing so, the vertical edges of established 3D bounding box are directly associated with the vehicle’s height, thus overcoming the difficulty of identifying the vehicle’s pixel height in 2D images. Finally, the vehicle’s physical height is estimated using a known object length in the same surveilled scene according to view geometry. The evaluation results signify the potential of the proposed method for use in over-height collision prevention.

Vehicles are classified as 13 classes in the Federal Highway Administration vehicle classification scheme, including motorcycles, passenger cars, buses, and various types of trucks. Truck traffic exerts a significant impact on pavement structure and is the main force causing pavement structure damages and distresses. Detailed truck traffic data are continuously recorded using Weight-in-Motion devices on the highway system. There usually exist a noticeable amount of unclassified and misclassified vehicles in the recorded dataset, which are those that the devices failed to identify their vehicle types based on the integrated criteria. As truck traffic is the main input of mechanistic-empirical pavement design, inaccurate vehicle classification would cause overdesign or underdesign of pavement structures and would consequently result in unnecessary extra construction costs as well as highway user costs. This paper presents a method to identify and reclassify the misclassified and unclassified vehicles utilizing the convolutional neural network in combination with spatial pyramid pooling. Through fine-tuning a series of hyperparameters, the proposed model successfully identified and reclassified a major portion of the unclassified and misclassified vehicles.

Weigh-in-Motion (WIM) techniques are emerged as a prominent solution to regulate weight-related violations, but high installation and recurring maintenance costs are the major obstacles of employing them. Also, the installation process of the inductive loop for vehicle presence detection can cause damage to pavement structure and lead to early deterioration of the pavement surface. Recently, Virtual Weigh-in-Motion (V-WIM) techniques consisting of WIM scales, traffic surveillance videos, and other sensors become a new technological trend that attracted interests from state DOTs to deal with size and weight enforcement that reaches beyond the WIM’s conventional role in data collection. Accurate detection of the specific vehicle-type information to activate the WIM sensors is the first major step for the V-WIM from roadside captured images/videos containing vehicles. However, issues of existing vehicle detection such as the lack of vehicle-type recognition, low-detection accuracy, night and dusk/dawn conditions, and slow speed are the major challenges for developing a powerful V-WIM. This paper applied machine learning based You Only Look Once (YOLO) algorithm trained on a custom dataset as an improved automatic vehicle detection process to work on existing traffic surveillance videos/images as the vehicle sensor data source. The results showed that it could detect vehicles close to the camera with high confidence. Such an improved automatic vehicle detection process may enhance the performance of V-WIM support to develop a dynamic traffic monitoring system and increase the applications of traffic surveillance videos/images.

Construction fires are a big threat to human life and property safety. Buildings under construction are largely unprotected as they are not yet equipped with fire protection systems. Consequently, any construction operation that fails to adhere to fire safety procedures will increase the fire risk to the site itself as well as to the adjacent buildings and put occupants at risk of injury and death. To enhance safety awareness and understanding of fire risks and consequences, it is necessary to provide construction participants a tool for fire risk assessment during early phases of a construction project at an activity level and on a daily basis. Therefore, this study proposes a modeling solution that applies a four-dimensional (4D) (3D + time) Building Information Model (BIM) along with the Fire Dynamics Simulator (FDS) tool to simulate and assess fires in buildings under construction. A numerical implementation is conducted as a case study to validate the practicality of the proposed work. This study is expected to promote awareness of construction fire safety as well as provide a tool for construction participants in better understanding and identifying construction fire risks. The study results can in turn help to mitigate fire hazards and prevent incidents from occurring during the early phases of a construction project, and ultimately reduce labor fatalities and injuries as well as property loss as a result of a construction fire hazard.

The Architecture, Engineering, and Construction (AEC) industry has explored, implemented, and reaped benefits from various Extended Reality (XR) applications. However, a large portion of these applications was developed in an ad-hoc approach for different technologies and different stages of a project. Sharing digital content among these applications is challenging, although many of them have the same development platform and original 3D model. This paper reviewed XR applications in the lifecycle of a construction project and summarized key information requirements for 3D models, development software systems, and processes for information transfer. Results show that a generic and adaptable model that satisfies these requirements can streamline the development process of different XR applications in the same project. Creating a generic model does not require additional activities. Instead, it relocates model pre-processing activities to the beginning of the XR development and precludes repetitive activities when possible. This paper contributes to the body of knowledge by specifying the requirements of a generic model that enhanced the interoperability of digital assets among XR technologies and applications.

Offsite construction (e.g., wood modular houses) has many advantages over traditional stick-built construction, ranging from schedule/cost reduction to improvement in safety and quality of the built product. Unlike stick-built, offsite construction demands higher levels of design and planning coordination at the early stages of the construction project to avoid cost overruns and/or delays. However, most companies still rely on 2D drawings in the development of shop drawings, which are required for the fabrication of the building components such as walls and roofs. In practice, the process of developing shop drawings is usually based on manually interpreting the 2D drawings and specifications, which is time-consuming, costly, and prone to human errors. A 3D information model can improve the accuracy of this process. To help achieve this, the authors developed a semi-automated method that can process 2D orthographic views of building components and convert them to 3D models, which can be useful for fabrication. The developed 3D information model can be further transformed to building information models (BIMs) to support collaboration amongst users and data exchanges across platforms. The developed method was evaluated in the development of wall components of a student apartment project in Kalamazoo, MI. Experimental results showed that the developed method successfully generated the 3D information model of the wall components. A time comparison with the state-of-the-art practices in developing the wall components was performed. Results showed that the developed method utilized approximately 22% of the time it took the state-of-the-art manual method to generate the 3D models.

To facilitate a better understanding of building codes, the visualization of the embedded structures of the provisions and requirements of the codes is needed. Existing research efforts in building code compliance checking mostly do not purposefully represent building codes in formats that facilitate human understanding and interaction with the codes, such as XML and hypertext (text with links to other text). Visual programming commonly represents building codes more visually as flowcharts. However, flowcharts are static, and the generation of flowcharts is still manual. To address this lack of interactive visual representation of building code requirement structures, this paper proposes an automated building code structure extraction and visualization method for visualizing building code contents in a way that clearly shows the inter-connections between requirements and allows intuitive user interaction. In this method, to extract the chapter-section-subsection hierarchical structure and cross-reference structure, a new extraction method named Building Code Network Generator (BCNG) is proposed to automatically generate an interactive visualization using a directed network. The performance of the proposed BCNG was empirically tested on Chapters 5 and 10 of the International Building Code 2015, with a resulting precision, recall, and F1-score of 99.4%, 96.3%, and 97.8%, respectively. In addition, the extracted hierarchical and cross-reference structures were displayed using an open-source network visualization tool to facilitate human understanding and interactions with the building code requirements in automated compliance checking systems.

The construction industry has been exposed to many risks emerging from the COVID-19 pandemic. This can be seen from project schedule delays, cost overrun, legal claims to supply chain distribution, and workforce restrictions. To this end, the purpose of this study is to identify and analyze the risks emerging from the COVID-19 pandemic for the construction industry in developing countries by considering the case of the Iraqi construction industry. To achieve this, a mixed-methods approach was adopted as follows: (1) focus group session with eight construction experts to identify the risks emerging from the COVID-19 pandemic; (2) development of Analytic Hierarchy Process (AHP) method based on probability and impact risk matrix by considering two additional dimensions, namely Estimated Recovery Time Level (ERTL), and Expected Cost Level (ECL); and (3) survey data collection from 41 Iraqi construction experts to quantify the identified risks. Ultimately, a total of 29 risks were identified and categorized into four construction themes, namely contractual/legal implications, construction financial market, supply chain operations, and health and safety of the construction workforce. The findings show that the most significant COVID-19 pandemic risks are claims arising under a construction contract, virus spreading due to poor hygiene and lack of sanitizing stations, and contractor bankruptcy. This research offers a better understanding of the critical risks emerging from COVID-19, which can be used as a reference for researchers and industry stakeholders in planning to address the impact of the COVID-19 pandemic risks during the life cycle of a construction project.

Cybersecurity risks, together with the associated physical security risks, are hindering the advancement of smart building innovations and applications. The cyber-physical (CP) security for smart built environments is an emerging, critical challenge that cannot be addressed solely by existing cybersecurity means nor traditional physical security means. Currently, because of rapid developments in smart building-related technologies, such as the Internet of Things (IoT), Cyber-physical Systems, and Artificial Intelligence (AI), researchers and industry professionals need a state-of-the-art overview of the solutions for CP security threats in smart built environments. Through a systematic literature review, this paper summarizes and discusses existing management means and methods to achieve more secured built environments. The content analysis results show that most research studies are in the area of smart grids and power systems’ CP security, with focuses such as intrusion detection, cyberattack impact mitigation, system status monitoring, and system resilience improvement. The existing gaps of built-environment CP security research include human factors, coordinated attacks, preventive solutions, and the interactions between cybersecurity and physical security.

Understanding the attentional patterns of construction professionals in design, construction, and facilities management processes is critical to identifying the performance bottlenecks and improving the current workflow designs in the Architecture, Engineering, Construction, and Facility Management (AEC/FM) industry. Recent developments in eye-tracking technologies have attracted an increasing attention of construction scholars as it is an effective tool for attentional analysis. However, the existing monitor-mounted eye trackers are designed for tracking gaze focus on 2D screens (such as attentions during a 2D drawing review) and thus are limited for understanding attentional patterns in 3D model reviews. To track gaze focus in 3D environments, scholars have to use expensive eye trackers integrated with glasses or headset displays such as Virtual Reality headsets. As an alternative, this paper proposes an effective solution for tracking gaze focus in 3D model reviews on 2D screens using the regular monitor-mounted eye trackers. The proposed 3D Desktop-Based Eye-Tracking System called “3D-DesktopEye” leverages the eye-tracking function of the Unity Game engine and a set of geometric conversations to project the 2D eye-tracking points into virtual 3D models. As a result, it can track the focus points of a user looking at a 3D model on a 2D screen. We tested the 3D-DesktopEye system in a facility management scenario—a pipe maintenance task. We also compared the effectiveness of the proposed 3D-DesktopEye system with the traditional 2D eye-tracking system. The results confirmed the usability and feasibility of the 3D-DesktopEye, and the proposed 3D-DesktopEye can help the user achieve better task performance.

Fire has been one of the main disasters of human society for centuries. Although professional practitioners and scholars have conducted a lot of research and experimental work on building fire simulation and related development, the trends for fire-related deaths and economic loss in the United States have increased in the past decade. On the other hand, the emergence of new building forms and the needs of special building functions expose building fire simulation to new challenges and problems. This paper provides a survey of the state-of-the-art works on building fire simulations by investigating the research that has been conducted in this field from 2015 to 2020. This review reveals that the fire simulation of high-rise and public buildings (e.g., subway stations and museums) has received more attention from researchers than others. This paper also summarizes the latest technological advances in this field. The results of this paper will help professional practitioners and scholars to identify challenges and problems to inform future studies.

Recent studies have been focusing on developing unmanned aircraft system (UAS) based construction safety management applications. However, it has failed to investigate the potential performance of the UAS-based safety management because it is hard to define the relationships between factors as well as quantify the performance. A causal loop diagram (CLD) can represent the process in which various factors and impacts are intertwined. Therefore, this research aims to define the variables affecting the UAS performance for safety management as well as to develop a CLD containing feedback structures through the extensive literature review. The CLD will be used to develop a system dynamics (SD) model that can assess the safety management performance using UAS over time. One of the potential challenges is limited validation of the model in the real environment. However, this study makes a significant contribution to the body of knowledge on this topic to assess the performance of how the system is effective for construction safety management.

As the cities keep expanding and the population continues growing, an immense need to improve the existing buildings’ energy performance has emerged in the energy-intensive built environment. One of the critical challenges to improving buildings’ energy efficiency is to accurately model and portray the buildings’ energy performance. Current research mainly focuses on simulation- and data-driven-based methods to help predict the buildings’ energy consumption. However, without information from the buildings’ actual use, both methods rely on assumptions and simulated parameters to predict the results. This research developed a computer vision-driven building energy modeling framework that models the power consumption of equipment and home appliances by semantic segmentation of the reality model. The framework is a simulation-based method with interior model based on actual use condition using computer vision-driven method. The framework can improve the simulation result and analyze the energy-consuming source in an existing building.

Building Information Modelling (BIM), integrated with innovative visualization technologies such as Augmented Reality (AR) and Virtual Reality (VR), can provide a holistic and system-driven approach at every stage of the building life cycle. The Architectural, Engineering, and Construction (AEC) industry, relative to other industries, has been slow in adopting AR/VR technologies, partly due to a lack of feasibility studies examining the actual cost of implementation versus an increase in profit. AR and VR have the potential to improve projects’ overall performance, simplify construction processes, and boost stakeholder collaboration. However, there is still a lack of proper understanding of how well such new technologies have been adopted, how much time and cost can be saved, and what limitations do users have. This paper aims to explore the current level of understanding of AR/VR technologies adoption in the AEC industry. To investigate the adoption of AR/VR at the company level, implementation in projects, benefits, and obstacles of using AR/VR, a survey is distributed to the AEC industry professionals across the United States. Responses from 158 industry experts as well as academic researchers are evaluated to assess the current state, growth, and saving opportunities of AR/VR technologies in the AEC industry.

Masonry structures have been widely used around the world because of their low cost and simplicity of construction. In particular, unreinforced masonry (URM) structures represent a high percentage of the residential building stock in the central and eastern United States. Unfortunately, many URM buildings are brittle and have a severe failure mode under seismic loads. Although it is urgent to retrofit the existing housing stock, resources are limited, and it is not practical to retrofit all URM buildings in one area. This study aims to examine the proof of concept of optimal resource allocation for retrofitting URM structures based on variables that directly influence damage costs, such as the priority of the building, size of the building, and cost constraints. To this end, a novel model based on the Conditional Value at Risk (CVaR) was developed through a genetic algorithm and validated under three CVAR scenarios. The findings revealed that this model could be used to prioritize buildings such as hospitals and schools based on importance and vulnerabilities. This research should help the decision-makers optimally allocate money to retrofit buildings in a disaster-prone area before the next earthquake occurs.

As the use of Building Information Modeling (BIM) for retrofits and Operations and Maintenance (O&M) activities in existing buildings becomes increasingly widespread, modeling the as-is building conditions becomes a requirement. Pipes are among the elements of most interest in these endeavors because they represent a significant portion of the building systems and have major impacts on O&M efforts and expenditures. Given the usually complex configurations of piping systems in existing commercial and industrial facilities and the tedious, time-consuming, and error-prone process associated with manual BIM modeling, automated methods for pipe modeling from point clouds have been proposed, which drastically reduced modeling times. A significant limitation that persists, however, refers to the fact that the classification of the piping systems is still a manual process in all current applications. While other researchers have reconstructed pipes, none have provided an automated method to capture and include the critical semantic information pertaining to the system type and usage. This paper, which is the first part of a major project that aims to reconstruct and classify piping systems in existing buildings using data from laser scanners and thermal cameras, shows the initial results of the proposed method for piping system reconstruction. The proposed reconstruction method is based on the estimation of the skeletons of straight pipes, followed by the identification of tees and elbows based on the relationships among straight pipe segments. The main results include the computation times for the process and the comparison of the as-is model to the reconstructed BIM model.

This paper presents a convolutional encoder-decoder network model for pixelwise segmentation of drone-based ortho-images of construction sites. The input ortho-image disassembling and output label-image prediction assembling algorithms are supplemented with the encoder-decoder to process high-resolution inputs. Parameter analyses were conducted to evaluate the performances between differently sized small-patches and different model training epochs. Testing results showed using the 64 × 64-pixel patch with 100-epoch can produce a well-trained encoder-decoder for pixelwise segmentation with a pixel accuracy of 0.98 in validation and 0.93 in testing, and had weighted average results of the precision, recall and f1-score larger than 0.98 in validation, and larger than 0.93 in testing. The developed method was also applied to object detection in drone photogrammetric orthophoto for objects’ contour extraction, then the extracted contours were used for as-built modeling in AutoCAD. The results of this work can benefit construction professionals in automatic as-built modeling and earthwork estimation if they have the corresponding elevations.

Research shows, even with the continuous development of coating technology, coating systems are still replaced before the expected lifespan. Recent advancement on data analytics (e.g., machine learning) offers an opportunity to leverage the wealthy historical data processed by State Departments of Transportation (DOTs) for improved understanding of coating premature failures. However, the coating-related data (e.g., coating condition data, coating application data) are stored in heterogeneous and unstructured formats (e.g., PDF) with limited data analytics capability. Therefore, better managing the data and knowledge related to coating systems is needed. To address the need, this paper presents a study that focuses on developing a semantic knowledge management system on steel bridge coating systems. As part of it, this paper focuses on presenting an ontology-based semantic model that formally represents and reasons about the knowledge of steel bridge coating systems. The model together with the knowledge management system will eventually support more systematic data-based analysis of coating premature failures, which will lead to new knowledge on factors that impact coating performance and potentially better practices that improve coating lifespans in the field.

Severe damage caused by past hurricanes calls for more effective evacuation in high-risk areas. Existing warning systems often define coarse-grained warning zones without consideration of the built environment. This research designed a targeted risk communication process based on the building-level hurricane risks considering hurricane wind speed, land use, and flooding risks. To evaluate the communication process in motivating evacuation, we utilized agent-based modeling to simulate the local households’ evacuation decision-making and compare the simulation outcomes with the real-world mobility data collected from over 758 smart devices during Hurricane Irma in North Miami Beach. The comparison outcome showed that household-targeted risk communication can better motivate residents at high risk to make timely decisions to evacuate. This research integrates human, environmental, and hazard factors in disaster risk communication. The research improves the understanding of information and human behavior interactions under emergency response situations. It contributes to fine-scale risk assessment and nuanced warnings for more effective risk mitigation and disaster response.

Knowing dynamic and fine-grained interactions between people and the built environments (e.g., facilities and infrastructure) during extreme events could assist in multiple emergency management operations. This study achieves the objective with agent-based modeling (ABM) simulation, which captures traveling patterns of affected population and reconstructs high-resolution intra-urban population movements under the perturbation of extreme events. Specifically, we extracted traveling patterns and activities referring to massive high-resolution, anonymized GPS traces and visited Point of Interests (POIs) (e.g., restaurants and grocery stores) collected from the city of Tallahassee in response to Hurricane Irma in 2017. We identified disparities in affected people’s traveling patterns regarding visitation frequencies, activity durations, activity start time, and traveling distances under normal and extreme event circumstances. We then developed a data-driven ABM to simulate activity-based travels for normal and adverse conditions. The research outcomes could show the dynamic spatio-temporal patterns of human mobility and facility utilization under the perturbations of extreme events, which pinpoint locations with large throngs and critical facilities/infrastructure with more visitations and, therefore, informs future city and infrastructure planning projects for enhancing disaster resilience.

It has become more common to use Building Information Modeling (BIM) to effectively integrate information among different phases of a construction project. However, due to the employment of different BIM software tools or platforms among different stakeholders, inconsistent or missing data are usually observed, which leads to difficulty in data communication and therefore low-working efficiency. To address that, an information checking, provision, and application process (ICPAP) framework was proposed in this paper with three main functions: model information analysis, guided information provision, and information consumption application (e.g., cost estimation). Algorithms developed under this framework can automatically integrate and check information from Industry Foundation Classes (IFC) models to guide information provision to further feed other information consumption services. An experiment was conducted to test the framework at the material information level, in which a 12-storey model was selected to conduct information analysis and information provision based on customized input requirements of concrete material, to serve further BIM applications (e.g., structural analysis and cost estimation). The evaluation was conducted from time efficiency and accuracy perspectives in comparison with pure manual operation, which demonstrated that the proposed framework led to higher performance on time efficiency and accuracy compared to a manual procedure. It enables the information communication from decentralized platforms to a centralized BIM, and considers the application needs during the checking to facilitate automated information provision suggestion for targeted applications. The proposed framework provides a new approach to facilitate BIM enrichment to better serve Architecture, Engineering, and Construction (AEC) applications.

Healthcare projects are the most sophisticated facilities to design, construct, and operate in the architectural/engineering/construction (AEC) industry. A systematic and transparent architecture verified with practical use cases for building a digital twin (DT) would be the foremost step for effective operation and maintenance of healthcare facilities. This manuscript provides a system architecture for DT that is specifically designed at the healthcare operation level. Five layers of the digital twin healthcare facility management (DTHFM) are demonstrated in detail, including data acquisition layer, transmission layer, integrated middle platform layer, service layer, and target layer. Based on this architecture, a DT-based smart ward in Shanghai Tongji hospital was developed. It integrates heterogeneous data sources to support real-time data querying and analysis. And it promotes effective decision-making with better healthcare service, further bridging the gap between human and healthcare assets and spaces. Finally, it provides a clear roadmap of future directions and timely opportunities in DT-based healthcare facilities management.

“Digital twins” as models for information-driven management and control of physical systems have emerged over the past years in multiple industrial sectors and recently also in construction. However, in the domain of construction safety, a digital twin remains undefined, with little or no consensus among researchers and practitioners of two essential aspects: (1) the connection between the physical reality of a construction site (the “physical” twin) and the corresponding computer model (the “digital” twin), and (2) the most effective selection and exploitation of real-life data for supporting safe design, planning, and execution of construction. This paper outlines the concept for a Digital Twin for Construction Safety (DTCS), defining three essential steps in the digital twin workflow: (1) safe design and planning for hazard prevention, (2) risk monitoring and control for proactive prediction and warning, and (3) continuous performance improvement for personalized- or project-based learning. DTCS should be viewed as a system-based approach enhancing the overall safety performance rather than exclusively integrating sensing information or safety knowledge in Building Information Modeling (BIM) for safety purposes. The result is an outline of our vision of the DTCS and a description of its components. Additionally, we point toward future research on the topic.

Given the fatal risks that highway workers encounter, securing the safety of highway work zones is one of the most pressing issues in the highway maintenance and operation community. Recent advances in Artificial Intelligence, Augmented Reality, and the Internet of Things (IoT) suggest that departing from current reactive safety systems to predictive and sophisticated solutions is feasible. To this end, real-time communication among the utilized agents plays a crucial role in the success of such systems. Therefore, in this paper, we investigate the challenges and requirements of establishing a local IoT system in highway work zones. To this end, we devised a case study to design a network among multiple endpoints and measured the latency of real-time communication among the utilized devices based on different network architectures. Our results indicate that in the most crowded network, 4 endpoints were able to wirelessly communicate with an average latency of 10 milliseconds at farthest distance. We also identified distance, the number of endpoints, and heavy moving objects as the potential barriers impacting network performance. Overall, this paper contributes to the body of knowledge by providing preliminary insights toward utilizing IoT networks in the context of highway work zones.

Neural Networks (NNs) are becoming more widely used in construction research. They have shown potential when working with prediction and detection algorithms. As found in several papers, NNs outperformed other algorithms in the area of machine learning and have recently improved their applicability, that is, using GANs. The types of NNs are used in construction to solve problems such as recognition of heavy machinery, workers’ pose assessment, and unit price estimations. They range from NNs with one fully connected layer to CNNs, RNNs, and LSTMs with many hidden layers. To understand the future potential of NNs, we completed a bibliometric analysis of the existing literature including publications from journals since 2010. The areas in which NNs have been used in construction applications are defined and mapped to understand the connections between them. The analysis also included a review of the NN model architecture that was used to solve the problems. This work found a wider field of applications for NNs in the construction industry than originally known. New architecture such as transformer networks has not yet been used in construction research but could lead to higher performance networks. A path forward is presented and discussed to realize this potential.

Asset information management plays a critical role in the successful operation of facilities. However, current documentation and distribution of facility asset information are fragmented. As an example, usually, maintenance staff do not have access to visual asset information. Even in cases where sufficient data are captured, it is often not readily accessible in a useful form for maintenance workers. Therefore, it is crucial to assess the alignment of information collected and received by maintenance crews against their needs to identify gaps and develop a useful information management system. The main goal of this research is to develop a process for structuring, documenting, and distributing data starting from the construction stage and extending into later use in facility management (FM) and maintenance to support maintenance workers’ needs. This research aims to review the current process of data collection, storage, and distribution, as well as identify any gaps based on the needs of maintenance crews. First, a literature review was completed on FM data requirements. Afterward, a content analysis of common information management systems was conducted at a large institutional owner (The Pennsylvania State University). The information needs of maintenance staff were gathered through semi-structured interviews, assessing the alignment of the current asset management system with their needs. Finally, suggestions were made to enhance the current data collection and distribution system based on the existing gaps. Results and a structured approach can benefit both facility owners and maintenance crews by enhancing the information requirements and management process.

Virtual reality (VR) is the most advanced form of visualization and simulation. Although its benefits on critical thinking and cognitive abilities are proven, VR is not strongly utilized in construction. The research is driven by the fact that “learning by doing” is more effective than traditional methods. Therefore, the goal of this research is to develop and test a prototype for the subsequent large-scale utilization of VR in enhancing safety in construction sites and in education. The objectives are to (1) verify the applicability of VR in construction education, (2) evaluate the prototype developed in terms of efficiency in enhancing safety performance, (3) evaluate the model in terms of sickness and usability, and (4) provide recommendations to be implemented in the model to be used in the large-scale testing of the VR technology within education in the construction field. After the development and testing of the prototype, results indicate that the immersive VR model significantly enhanced hazard identification skills compared to traditional safety training. In addition, the assessment of its efficiency and feasibility have shown the potential of VR in improving construction education; thus, paving the way for having VR as the main complementary tool to traditional education on a larger scale in the future.

The construction industry requires effective communication between project stakeholders in various locations, which has traditionally been facilitated through in-person site visits or phone calls. These methods can lead to wasted time driving to remote sites for in-person visits or misunderstandings from omitting critical site details on phone calls. Augmented Reality (AR) has the potential to enable a consistent understanding of site conditions provided through in-person meetings with the efficiency of phone calls. However, the majority of studies exploring AR in construction present results from controlled implementations or laboratory environments. Conversely, this investigates AR calls between practitioners on active construction sites to identify factors and behaviors relevant to successful use of AR in real workspace. Furthermore, interviews with participants enabled the authors to understand the perceptions of the practitioner participants that may have guided their behavior. The results helped explore about six drawbacks of AR application in real workspaces and their associated solutions. Manipulation, color of annotation feature, and the time needed to upload large files are examples of the experimented drawbacks. Practising, training sessions, and the use of other available colors for annotation are explored solutions to overcome some of the revealed drawbacks. By altering their behavior and with more practice and repetition, users overcame some of the explored challenges. These findings will inform researchers and practitioners interested in using AR to effectively support communication on construction projects.

Change orders, conflicts, and miscoordination between different disciplines are challenges that still faced the construction sector. These challenges occur continuously during the project lifetime, particularly in the construction phase leads to time and cost overrun. Day by day, the construction sector saturating with Building Information Modelling (BIM) strategies. BIM introduces uses like 3D coordination that play an important role in the construction sector, for its ability to solve conflicts in the early stages of design. Consequently, firms have tried to develop their systems but not always decide to adopt BIM, partially due to the limited knowledge of the BIM’s effective processes, protocols, and workflows. This paper presents statistics that were collected before and after using BIM in a real case, to clarify the value of adopting the BIM. The main objective of this paper is to adopt BIM in the design phase to avoid those problems that arise in the construction stage. A case study comparing the quality and time requirements of the same project was measured by three scenarios: at the first, using a 2D CAD tool; second, using the current traditional use of BIM and finally, using full BIM enhanced by (BIM360, Dynamo, Revit, and Naviswork applications). The data for analysis were obtained via a consultant firm database, which focuses on the detection of clashes during the design development. As a result, the implementation of BIM has been verified using the proposed technologies to reduce the clashes and eliminate field conflicts.

In the construction sector, contractors deal with many uncertainties during their projects such as cost overruns, schedule delays, and safety issues. To overcome such complications, contractors pursue diverse risk management approaches which reflect their risk behavior and attitude toward expected risks as well as bidding decisions. Risk attitude is influenced by various factors that include individual characteristics, demographics, social and cultural influences. This paper conducted a thorough literature review to identify the different internal and external factors that influence contractors’ risk behavior adaptation to varying bidding situations. It also ranked the said factors by order of importance based on a survey administered to construction professionals. Survey findings showed that the most influential factors are the completeness of project information and the contractor’s need for work. Then, the paper presented a bidding framework that aims to mimic the competitive construction bidding process and to study the long-term impact of a contractor’s adaptive risk attitude on his financial performance and growth in the market.

Labor productivity and its influencing factors including ergonomics play a vital role in affecting the performance of construction projects. In fact, studying ergonomics and understanding the interactions among workers and their assigned tasks have shown a decrease in workers’ discomfort, a positive impact on labor productivity, a reduction in project costs, and an increase in value creation. As such, several studies have been conducted in an attempt to properly assign construction tasks and optimize the performance of crews. However, no study has yet been carried out to estimate the physiological workloads of construction tasks and match them with the corresponding workers’ capabilities. Therefore, this research study takes the initial steps and aims at using Digital Human Modeling (DHM) to model different construction activities and generate physiological task demands. Several construction activities that require various body postures and affect different body parts are selected and modeled using DHM. The ergonomic and physiological results are then recorded for each activity. The resulting physiological task demands will, in future work, become the foundation of a simulation framework targeted at enhancing the worker-task assignment process and properly mapping the modeled tasks to construction workers based on their physiological capabilities.

The scope of earthworks in the heavy civil projects represents a substantial risk that might cause a schedule delay and cost overrun. The high uncertainty in the outcomes of the risk arises from the several dynamic variables that impact earthwork operations. Past experiences showed such complex and dynamic interactions of variables are hard to control once occurred. A proactive approach would contribute to anticipating these interactions and support a timely decision-making process. This paper presents an approach that utilizes unmanned aerial vehicles (UAVs) to estimate earthwork volume compared to prior known surveying methods when tracking earthwork volume. The process involves collecting georeferenced images using UAVs, then processing them through a photogrammetry cloud computing platform to generate a 3D point cloud model and used to create the surface for the respective site. The model will be utilized to conduct a cut/fill analysis using a computational algorithm to create a volumetric map. The proposed method is expected to provide a better understanding of the earthmoving fleet planning and the distribution of the work scope on the volumetric map. Using this method can provide operations a more efficient means to creating a real-time surface models as opposed to having traditional time-consuming methods involved.

Based on an ethnographic action research study for a Digital Twin (DT) deployment on an automated highway maintenance project, this paper reports some of the stumbling blocks that practitioners face while deploying a DT in practice. At the outset, the scope of the case study was broadly defined in terms of digitalization and software development and deployment, which later pivoted toward the concept of Digital Twin during the collective reflection sessions between the project participants. Through an iterative learning cycle via discussions among the various project stakeholders, the case study led to uncovering the roadblocks in practice faced by the Architecture, Engineering, and Construction (AEC) practitioners. This research finds that the practitioners are facing difficulty in (1) creating a shared understanding among the participants due to the lack of consensus on the Digital Twin concept, (2) adapting and investing in Digital Twin due to inability to exhaustively evaluate and select the appropriate technological capabilities in a Digital Twin, and (3) allocation of resources for Digital Twin development due to the inability to assess the impact of DT on the organizational conditions and processes.

This study considers projects that employ multi-skilled labor resources in performing different tasks aiming at improving labor utilization efficiency. Based on field observation, the journeymen employed in a steel girder fabrication shop for bridge construction exemplify multi-skilled labor resources in a practical setting. In particular, the need for crew transferring and waiting between various workstations on the shop floor gives rise to the bulk of semi-productive labor time. Unpredictable and unnecessary semi-productive worker hours are considered as a kind of waste as per lean principles. Increasing labor flow efficiency by properly allocating limited labor resources to project activities would reduce the semi-productive labor hours while enhancing the labor flow reliability, leading to better productivity and leaner processes. Labor Flow Waste Index (LFWI) is defined based on the determination of the semi-productive worker hours using resource-constrained project scheduling analysis. Further, the optimization problem of minimizing LFWI is formulated. A case study was conducted Utilizing Microsoft Excel Solver, resulting in significant decrease on the waste in labor resource flow.

Construction inspection is a crucial stage that verifies if a construction project has met all contractual requirements. However, the lack of resources due to the budget reduction for maintaining the large aging transportation network has greatly affected the construction inspection capabilities. There is an emerging need for risk-based construction inspection practices that can help highway agencies optimize the use of limited resources without compromising inspection quality. Automated prioritization of requirements according to their criticality would be extremely helpful since contractual requirements are typically presented in an unstructured natural language in voluminous text documents. The current study introduces a novel model for predicting the risk level of contractual requirements using the Long Short-Term Memory (LSTM) neural network. The train data include sequences of requirement texts which were manually rated with severity level by industry professionals. For feature extraction, the study used word embeddings. The developed requirement risk prediction model was evaluated using root mean squared error. The proposed model is expected to provide construction inspectors with a means for the automated classification of voluminous requirements by their importance, thus help to maximize the effectiveness of inspection activities under resource constraints.

Augmented Reality (AR) systems can be used for different purposes in the AEC industry, including augmentation of physical environments with virtual objects that enhances information understanding and delivery. Previous research developed single-purpose solutions with limited context identification. Broad context identification and holistic AR application across industry stages and stakeholders is lacking. This article presents a prototype framework for an intelligent Context-aware AR System (CaARS) to enable broader applications in the AEC industry. The framework considers novel enablers, including multi-object scanning, a model slicer for customizing a single 3D model, using anamorphic images for object recognition, and an interactive user interface. The framework was analyzed for three case studies that represent various applications of CaARS in the industry. Unlike previous single-purpose AR applications in the industry, the approach described in this article seeks to address the multiple scenarios where AR augmentation serves to increase design and cognitive understanding, decision-making, and task support.

Speed feedback signs have been implemented in transportation systems for decades. They are a critical part of civil infrastructure and transportation systems. However, a lack of evidence is available to show whether they perform as expected to prevent drivers from potentially harmful behaviors. One reason is the risk for researchers to observe and record driver behaviors on road. Therefore, this study develops a simulator to enable researchers to record driver reactions in an experimental setting without jeopardizing other drivers on road. The simulator combines virtual reality (VR), data recording, steering wheel control, and virtual civil environment. The simulator can simulate virtual rural–urban environment with various road conditions and allow researchers to track participants’ reactions to dynamic signs. The simulator tracks and records driver behaviors when they approximate to signs; and exports the records for data analysis.

The construction industry has suffered from stagnant productivity rate compared to manufacturing over the last 50 years due, in part, to its heavy reliance on manual processes and the highly complex and uncertain environment within which work must proceed. The manual nature of construction environment creates safety and productivity issues that are prone to human judgments and errors. AI techniques have potentials offer solutions to these issues, removing personnel from hazardous environments and enabling the implementation of industrialized concepts within a poorly structured environment. In recent years, AI techniques are starting to be adopted within construction, but its potential remains largely untapped. The application of industrialized concepts to construction, whether within a factory or on-site, comes with significant challenges such as the need to cater for one-off designs with limited repetition of construction components, sporadic demand for work, and logistical challenges resulting from the fact that each project is constructed and/or assembled at a unique location. This research reviews the state of art in AI techniques in manufactured construction and identifies the current and future potential for the application of AI techniques. The objective of this research is to identify the potential of AI applications in manufactured construction and provide a direction for the future research.

Typical construction project monitoring and control primarily depends upon paper-based documentation, which is prone to mistakes and cannot organize and retrieve relevant data efficiently. Building information modeling (BIM) is a modern concept that can be used for enhancing the efficacy of information management in the project monitoring and control process. This paper proposes a computerized system that connects the BIM models of a construction project and a dashboard for reporting the overall project progress, the progress payments for contractors, and the work progress of each project section through 3D models. The conventional project monitoring and control process was examined and analyzed. The findings were transformed to the structure of a BIM-based construction project monitoring and control process, which was then used to establish the system architecture of the proposed system. The system adopts Autodesk Revit for authoring BIM models, Microsoft EXCEL for storing data, Dynamo for linking Autodesk Revit and EXCEL, and Unity for creating a BIM-enabled dashboard system. The system was applied to an actual building project to illustrate its efficiency and practicality.

From the total electricity consumed in the US, a large share accounts for energy used in the buildings. Human energy-related actions in buildings significantly affect the building operation systems and the corresponding energy consumption. Simulation of these actions (e.g., occupants’ presence, movement, and interactions with building energy systems) plays a vital role in predicting and evaluating the actual energy consumption in buildings. Therefore, it is essential to comprehensively review the existing literature on human-building energy-related actions from the perspective of simulation methods. The present paper carries out a systematic literature review of the existing studies focusing on inputs, outputs, and simulation techniques used to model each human-building energy-related action to define the current literature, existing gaps, and future research directions in this field. A total of 57 journal articles, published since 2005, are collected and reviewed to analyze the simulation methods along with simulation input and output parameters regarding human-building energy-related actions in the built environment. It was concluded that statistical and data-driven methods were the most popular methods of simulating human-building energy-related actions. Also, the results show that the Markov Chain technique is the most popular technique for simulating various human-building energy-related actions. This study summarized the most practical simulation methods and presented the future research potentials and direction in human-building energy-related behavior modeling and simulation.

This research proposed a simulation method using Anylogic software to model passenger in and out flowlines and determine the operational capacity of a metro station in line 3. The logic was divided into three parts, namely passengers flow in, passengers out, and train arrival. The train arrival logic is linked with injection function to the passengers out and passengers in are generated using hourly flow distributions. Different ranges for passengers’ flow distributions were tested on the model to determine the optimum number of operational ticket windows. The criteria of selection were that the ticket level waiting time is within the train headway and a high crew utilization in the ticket offices. The following logic was simulated in a 2D and 3D view to visualize the pedestrians’ behavior and define congested areas. Based on the analysis of flowlines, some modification to the exit and entrance gates were suggested. The model shows demonstrative capabilities in defining the overall operational level and suggested plausible improvements to the facility layout organization.

Simulating a dam construction is an effective method for planning and resource management at the pre-construction stage especially in an environment with adverse rainfall. A dam body can either be concrete gravity dam or earth fill dam or it may include both. This work focuses on the construction of the embankment that can either be part of a dam or the whole dam. A simulation model is established to predict the construction duration of an embankment dam in a rainy environment at different confidence levels of precipitation given certain available resources. Initially, a rain forecast model was built to predict the expected precipitation for the upcoming years using Easyfit software and rainfall dataset collected from NASA for a specific location for years from 1995 to 2019. The simulation software used is Anylogic where the forecast rain model is inputted to the simulation model to study the impact of the precipitation given different confidence levels. The model result shows a direct relationship between the confidence level and the estimated construction duration.

Despite the recent trend of increasing adoption of building information modeling (BIM) in the infrastructure (i.e., horizontal construction) domain, there is still a long way to go before BIM-based platforms for bridges actually take full advantage of the BIM-associated benefits at a level that is analogous to that achieved in the building (i.e., vertical construction) domain. An immediate challenge is the need of processing traditional 2D bridge drawings to 3D information models for BIM-based computational tasks, as bridges in records are mostly documented in 2D PDF drawings. With the recent adoption of the industry foundation classes (IFC) standard as the national standard for roadway and bridges by the American Association of State Highway and Transportation Officials (AASHTO), most state Department of Transportations (DOTs) are faced with the challenge of converting their traditional 2D bridge drawings to 3D models that comply with IFC standard, as manually performing this task could be cost prohibitive considering the number of bridges managed by an agency could be at the level of hundreds of thousands. To help achieve this task with reduced need of manual efforts, the authors developed a semi-automated method and corresponding algorithms for automatically processing existing 2D bridge drawings in PDF and converting these processed 2D bridge drawings to 3D bridge models. The proposed method and developed algorithms were experimentally tested on two bridge drawings in 2D PDF formats. The generated 3D bridge models using the developed algorithms were compared against developed 3D models using the traditional manual method. Experimental results showed that the proposed method successfully generated 3D bridge models from 2D PDF drawings with a similar quality but significant (>90%) time saving comparing to the traditional manual method.

The state-of-the-art interoperability between Building Information Modeling (BIM) and Building Energy Modeling (BEM) is mainly one-way. BEM software like EnergyPlus takes design input from BIM and simulates the energy consumption of the building. BIM, however, is unable to leverage energy simulation results from BEM smoothly. This lack of bi-directional interoperability between BIM and BEM prevents buildings from seeking more energy-efficient design options and limits BIM applications in optimal generative design. In this research, the authors have conducted the following activities related to two-way interoperability between BIM and BEM: (1) surveyed its current status through literature review, (2) investigated barriers to the bi-directional interoperability, and (3) suggested possible approaches to enable bi-directional interoperability and research outlook in the field. As a result, the authors have identified three main barriers to BIM BEM interoperability, namely, inefficient collaboration, limited scope in information transformation, and lack of BEM to BIM transformation. Future directions are suggested to address these barriers.

Departments of Transportations (DOTs) are facing challenges in managing their infrastructure. Aging infrastructure, limited maintenance funds, and high user demand increases the pressure on the decision-makers to find efficient approaches to ensure a safe and reliable infrastructure system. DOTs heavily depend on the availability of data and budget to keep the infrastructure system in operation. The availability of reliable data coupled with an efficient approach for processing and use of the data can drastically improve the way we manage the transportation infrastructure. Toward that end, we are exploring the potential of developing a cloud-based framework that can provide decision-makers with a better approach for maintaining and managing transportation capital assets. The proposed framework involves integrating the assets Geographic Information System (GIS) and 3D point cloud data on cloud computing platforms. The framework consists of (1) selecting the candidate capital assets, (2) developing a 3D point cloud model of the assets using Light Detection and Ranging (LiDAR), (3) data processing and registration, and (4) a high-level digital. Utilizing cloud computing to visualize 3D point cloud data enables asset management engineers to make informed decisions and enhance asset management.

Errata identify incorrectly published items and technical errors. This file contains errata that shall be used from their effective dates; the most recent effective date is 05 19 2022.