Advances in Information Technology in Civil and Building Engineering

Ancient bridges lack adequate maintenance strategies and public attention compared to modern bridges. The current bridge maintenance standards are tailored for modern bridges and cannot be directly applied to ancient bridge maintenance because of differences in structure designs and construction materials. Besides, due to the urban development and the evolution of traffic, the frequency of using the ancient bridges has tapered off; people gradually elided the maintenance of ancient bridges. Nevertheless, some ancient bridges still serve as integral hubs in the transportation network and require more inspection due to their common features of aging structures and complex damage history. Previous studies have mainly applied sensor-based analysis for structural deformation problems in ancient bridge health monitoring. The mainstream inspection technologies include sonic transmission, radiography, infrared thermography, and ground-penetrating radar (GPR). However, these methods can only partially depict the interior condition of the bridge, and are time-consuming and complicated to implement in practice; their feasibility on ancient bridge maintenance is debatable. This paper proposes an image-based detection method to provide an effective solution for the maintenance of ancient bridges using Deep Neural Networks (DNNs). A masonry arch bridge in Hong Kong, built in the 1880s, was investigated. Unmanned Aerial Vehicles (UAVs) were deployed to collect the bridge surface information, and a 3D model generated with Structure from Motion (SfM) was preserved for further bridge health monitoring. In addition, an assessment criterion was purposed to evaluate the ancient bridge health condition, which is beneficial for the decision-making on ancient bridge maintenance.

As countries around the world gradually return to life before the Covid-19 pandemic, it is important for facility management divisions across education sectors to use innovative technology and unique solutions to provide healthy and safe learning environments. This research aimed to improve the management of educational buildings by leveraging innovative technologies. The first objective was to develop a system for real-time occupancy levels within lecture venues. To achieve this, a branch of artificial intelligence known as computer vision was combined with existing CCTV cameras to count occupants in real-time. This was achieved by training a convoluted neural network on a dataset of 15 000 images of ‘human bodies’ extracted from Googles Open Images v6. The second objective was to measure indoor air quality. A medical grade air quality device was placed within the assessed lecture venues and real-time occupant count was correlated against real-time indoor air quality data. The results from this study demonstrate the successful use of computer vision combined with existing CCTV cameras to accurately count occupants in real-time. The study utilised open-source AI resources and provides a method for further computer vision research. Regarding indoor air quality within the assessed educational buildings, the results of this study indicate that even under significantly reduced occupancy levels, carbon dioxide accumulated within assessed venues, indicating inadequate ventilation. The Covid-19 pandemic will not be the last pandemic we encounter. However, facility management can utilise innovative technologies to ensure educational buildings are managed using data-driven strategies that ensure learning environments remain safe and accessible spaces for students.

Government buildings, similar to any other structure, require regular maintenance to retain the original and ensure that they perform their functions. Building maintenance is separated into two areas building maintenance management and building maintenance technology. Building maintenance management must be looked at precisely as unfolding how a system of maintenance initiative might be planned to deal with a building maintenance problem. There are a few factors that influence the decision to accomplish the maintenance work. The study focused on assessing a few selected factors affecting public building maintenance based on user perspective in order to develop an effective public building maintenance strategy.A survey will be undertaken based on previous literature among building maintenance professionals. The research will enhance the body of knowledge about factors affecting maintenance management of public buildings.

Low operational efficiency and sustainability characterise South African government buildings, which partly emanates from lack of innovation in the South African public sector. As a result, inefficiency and lack of sustainability in the use of energy and water are major challenges, consequently affecting sustainability in buildings and causing detrimental effects to the environment. It is therefore imminent that the South African government adopts innovative technologies that ameliorate the public built environment. The goal of this paper is therefore to understand, from a critical review of literature, how harnessing technological innovation can improve operational efficiency and sustainability in energy use and water consumption in government buildings in South Africa. Careful selection of the most appropriate scholarly sources was done, which were then appraised to understand how the different latest technologies can be utilised and how they can be helpful in improving efficiency and sustainability in energy use and water consumption in buildings. The internet of things, digital twin, big data analytics and smart meters, were identified to be useful in improving efficiency and sustainability in energy and water consumption in buildings, whilst also improving indoor environmental quality. The result would be reducing the cost of energy and water management in South African Government buildings, and elimination of the energy and water crisis in South Africa, as well as minimisation of harm to the environment.

This study aims to improve the performance of optical character recognition (OCR), particularly in identifying printed Korean text marked by hand-drawn circles from images of construction defect tags. Despite advancements in mobile technologies, marking text on paper remains a prevalent practice. The typical approach for recognition in this context is to first detect the circles from the images and then identify the text entity within the region using OCR. Numerous OCR models have been developed to automatically identify various text types, but even a competition-winning multilingual model by Baek et al. does not perform well in recognizing circled Korean text, yielding a weighted F1 score of just 69%. The core idea of the lumped approach proposed in this study is to recognize circles and named entities as one instance. For this purpose, the YOLOv5 is fine tuned to detect 65 types of named entity overlapped with hand-drawn circles and yields a weighted F1 score of 94%, 25% higher than a typical approach using YOLOv5 for circle detection and a model by Baek et al. for subsequent OCR. This work thereby introduces a novel approach for developing advanced text information extraction methods and processing paper-based marked text in the construction industry.

Several regulations, standards, and requirements govern the lifecycle of the built environment. Such instances include legislations, government development control rules, environmental compliances, project and contractual requirements, and performance standards to be followed during construction. Compliance checking is a complex task that is often conducted manually, making it a resource-intensive, error-prone, and time-consuming affair. Past researchers have worked on methods and processes of automated compliance checking systems (ACCS); however, there has been a negligible adaptation in the industry. To this end, this study tries to recognize the reasons for the gap in the implementation of the pre-construction permit compliance systems in the Indian construction industry. The study understands the reasons from the end-user’s perspective through the means of a focus group study. Key findings indicate manual pre-processing of data is a significant hurdle in Building Information Models (BIM) for application in ACCS. ACCS applications developed are restricted in their area of usage due to the limitation in applicability beyond explicit building code clauses. Rule-based validation of regulation requires enriched structured data, which is generally absent from the models developed by architects in the design phase. The study indicates the necessity of automated data pre-processing step that includes intelligent model filling suggestions and semantic enrichment to increase the adoption of ACCS. This study points out that automatic semantic enrichment (SE) can be achieved by applying machine learning (ML). Areas of application of SE in ACCS are identified in the study, which can enhance the industry’s user experience and adaptation of ACCS.

Automated Code Checking (ACC) can be defined as a classification task aiming to classify building objects as compliant or not compliant to a code provision at hand. While Machine Learning (ML) is a useful tool to perform such classification tasks, it presents several drawbacks and limitations. Buildings are complex compositions of instances that are related to each other by functional and topological relationships. This type of data can be easily supported by property graphs that provide a flexible representation of attributes for every instance as well as the relationships between the instances. This, together with the recent developments in the field of graph-based learning led the authors to explore a novel approach for ACC supported by Graph Neural Networks (GNN). This paper presents a new workflow that implements GNNs for ACC to leverage the advantages of ML but alleviate the limitations. We illustrate the suggested workflow by training a GNN model on a synthetic data set and using the trained classifier to check compliance of a real BIM model to accessibility requirements. The accuracy of the classifier on a test set is 86% and the accuracy of obtained results during the accessibility check is 82%. This suggests that GNNs are applicable to ACC and that classifiers trained on synthetic data can be used to classify building design provided by the industry. While the results are encouraging, they also point to the need for further research to establish the scope and boundary conditions of applying GNNs to ACC.

Safety-critical infrastructures, such as bridges, are periodically inspected to check for existing damage, such as fatigue cracks and corrosion, and to guarantee the safe use of the infrastructure. Visual inspection is the most frequent type of general inspection, despite the fact that its detection capability is rather limited, especially for fatigue cracks. Machine learning algorithms can be used for augmenting the capability of classical visual inspection of bridge structures, however, the implementation of such an algorithm requires a massive annotated training dataset, which is time-consuming to produce. This paper proposes a semi-automatic crack segmentation tool that eases the manual segmentation of cracks on images needed to create a training dataset for machine learning algorithm. Also it can be used to measure the geometry of the crack. This tool makes use of an image processing algorithm, which was initially developed for the analysis of vascular systems on retinal images. The algorithm relies on a multi-orientation wavelet transform, which is applied to the image to construct the so-called ‘orientation scores’, i.e. a modified version of the image. Afterwards, the filtered orientation scores are used to formulate an optimal path problem that identifies the crack. The globally optimal path between manually selected crack endpoints is computed, using a state-of-the-art geometric tracking method. The pixel-wise segmentation is done afterwards using the obtained crack path. The proposed method outperforms fully automatic methods and shows potential to be an adequate alternative to the manual data annotation.

The performance of deep learning models can be significantly degraded on unseen data that has different visual characteristics compared to a domain where training data was collected. A simple and obvious way to maintain the performance of deep learning models is to prepare training data again in a new domain where target objects and backgrounds have different appearances compared to the original. However, it is not a trivial task considering time and efforts required in data preparation. To address this issue, this study proposes a pseudo label generation method from images that can automatically collect video clips for objects of interest and assign labels. The proposed method consists of a moving object detector to extract target objects in images and a classifier to assign labels on the extracted regions. The findings of this study provide important knowledge for construction site monitoring in securing the performance of computer vision models in various environments.

In the construction field, Building Information Modeling (BIM) offers various application scenarios in the design, construction, and maintenance phases of a building’s lifecycle. The utilization of BIM in the Operation and Maintenance (O&M) phase of existing buildings is a particular challenge, as suitable BIM models are usually not available as a basis. In this context, many researchers have investigated various methods to automatically create BIM models from existing information and data. Focusing on images, numerous methods have been investigated for detecting and classifying certain objects in buildings. However, despite the importance of fire protection in buildings, the research field has not focused on the benefits of the automatic recognition of fire safety equipment (FSE, e.g., fire blankets) in much detail. Particularly in existing buildings, the recurring inspection and maintenance of fire safety equipment is a responsible task and required by law. It is the responsibility of the owners and facility managers to ensure the availability and proper functioning of fire safety equipment in the building.Consequently, this work aims to contribute to this research area by investigating the state-of-the-art Mask Region-Based Convolutional Neural Network (Mask R-CNN) for instance segmentation and object detection of FSE in RGB images. The results show that this approach automatically extracts valuable semantic information that provides the presence of fire safety equipment in images. In addition, this study investigates the influence of hyperparameter adjustment on the detection of FSE objects in indoor scenes. It is also examined how the additional use of augmented data improves the performance of the neural network.

In the scope of this paper the profitable use of blockchain technology or so-called distributed ledger technologies in the context of Structural Health Monitoring (SHM) will be examined.SHM is the continuous or periodic and automated method for determining and monitoring the condition of an object. This is done by measurements with permanently installed or integrated sensors and by analyzing the hereby generated data. During those monitoring periods it often happens that some of the installed sensors are not working correctly. This may be caused by interference, by a corruption or even a manipulation. Therefore, the sensors can deliver abnormal data. Consequently, the delivered data is the starting point to overwatch the function of the sensors. Since on the monitored object multiple sensors are in use the measured data of those different sensors can be compared. If one sensor delivers deviating data it is likely that the sensor is not working the way it should. This verification process is being automated with blockchain technology. The correctness of the sensor data is then stored with the measured data itself on the chain. Thus, the entire generated information is securely and reliably tracked within the blockchain. Another advantage is the high scalability and decentralization of the blockchain. This is especially important when dealing with monitoring systems that can also have the need for scaling. For this described use case the implementation and application of the blockchain Hyperleder Fabric will be shown in the scope of this work.

Accurate and automated classification of workers’ activities is critical for safety and performance monitoring of workers, especially in highly hazardous working conditions. Previous studies have explored automated worker activity classification using wearable sensors with a sole type of data (e.g., acceleration) in controlled lab environments. To further improve the accuracy of worker activity classification with wearable sensors, we collected multimodal data from workers that conduct highway maintenance activities such as crack sealing, and pothole patching, in an Indiana Department of Transportation (INDOT) facility. Several activities were identified through field videos, including crack sealing, transferring material and walking. Two datasets were developed based on the collected data with one containing acceleration data only and the other one fusing acceleration data with multimodal data including heart rate, electrodermal activity (EDA), and skin temperature. The K-nearest neighbors (KNN) models were built to classify workers’ activities for the two datasets respectively. Results showed that the accuracies for detecting crack sealing, transferring material, and walking without the data fusion were 1.0, 1.0 and 0.71. With the data fusion, the accuracies for detecting crack sealing, transferring material, and walking became 1.0, 0.93, and 0.93. The overall accuracy for classifying the three activities increased from 0.9069 to 0.9535 with the data fusion.

If water trash exceeds the allowable load of a trash barrier, water trash barriers could be destroyed and the spilled waste negatively impacts the environment. Therefore, it is essential to measure the trash load in the water infrastructure to process collected water trash in a timely manner. However, there has been little investigation about how to monitor water trash in an automated way. To fill the knowledge gap, this study presents detailed investigation of water trash monitoring methods based on object detection models. To verify effective detection models and their performances, a new dataset is established, called the Foresys marine debris dataset. The dataset consists of a total of 6 water trash categories (Plastic, Vinyl, Styrofoam, Paper, Bottle, and Wood). State-of-the-art detection models were employed to test their performance, such as YOLOv3, YOLOv5, and YOLOv7 pretrained on the COCO dataset. The experiments showed that the detection models could achieve decent performance with proper amount of training image data; a number of training data required to secure decent performance varies by target class. The findings of this study will give a fresh insight for developing an automated water trash management system.

As a 3D model is considered the core element of building information modeling (BIM) applications, current research focuses on optimizing and automating the creation of BIM models, especially for existing buildings. In this regard, engineering drawings are the most common data source for extracting geometrical information. With the recent emersion of artificial intelligence (AI) applications and the availability of advanced algorithms, researchers have utilized computer vision approaches to process engineering drawings. However, the data used in previous studies is rarely derived from actual projects and is, in many cases, noise-free. Such datasets overlook the need to create BIM models for existing structures, which account for most buildings. To address this issue, the current paper describes an approach for extracting information from engineering drawings that originate from an existing infrastructure project in Germany. More precisely, the drawings depict the buildings located on both sides of a subway line. The scanned drawings are old and, in some cases, damaged. Moreover, such data collections are usually submitted or archived without a clear structure or logical naming convention, making data organization time-consuming and labor-intensive. The presented approach divides the extraction of information into two levels. The first level classifies the drawings according to the main content, such as different views of the building. The second level collects information from the first level’s identified views concerning structural and architectural aspects such as staircases, rooms, and openings. As the approach follows the idea of a data-centric AI, the pipeline includes an intensive exploration of the data, as well as its preprocessing, augmentation, and handling of damages in the drawings. The described approach is tested across multiple datasets and shows promising results. This study may represent an important step toward the automatic creation of digital twins for existing buildings.

Context-aware safety monitoring based on computer vision has received relatively little attention, although it is critical for recognizing the working context of workers and performing precise safety assessment with respect to Personal Protective Equipment (PPE) compliance checks. To address this knowledge gap, this study proposes vision-based monitoring approaches for context-aware PPE compliance checks using a modularized analysis pipeline composed of object detection, semantic segmentation, and depth estimation. The efficacy of two different approaches under this methodology was examined using YUD-COSAv2 data collected from actual construction sites. In experiments, the proposed method was able to distinguish between workers at heights and on the ground, applying different PPE compliance rules for evaluating workers’ safety. The depth estimation model achieved an Average Precision of 78.50%, while the segmentation model achieved an Average Precision of 86.22%.

To facilitate road image data collection, participatory sensing has been proposed in the literature utilizing a dashboard camera of a normal vehicle. It is not trivial to identify road cracks in such crowdsourced images due to the dynamic natures of photographing conditions which results in inconsistent the image quality. Although previous studies presented promising ways to identify road damages using deep convolutional neural networks (CNN), the performance is insufficient to be implemented in practical monitoring purposes. This study investigates core problems in improving the road crack segmentation performance by applying state-of-the-art segmentation models based on CNN and transformer architectures. Using a benchmark dataset, it was found that coarse annotation on crowdsourced images is detrimental to the performance evaluation and further development of participatory sensing-based monitoring technology. Interestingly, segmentation models could be trained by training data with coarse annotation. This study will give a fresh insight of advancing the knowledge in participatory sensing-based infrastructure monitoring.

A major concern in urban mechanised tunnelling projects is avoiding damage to the existing buildings and the tunnel boring machine (TBM), which may be adjusted by an advanced precise excavation simulation. Because a realistic simulation must account for multiple interactions between the boring machine and the subsurface, an exact representation of the ground’s geological profile must be created beforehand. Due to the limited monitoring and sampling, several geologic anomalies may have been overlooked when sketching the geologic profile. As a result, the geological profile should be updated alongside the construction phases when new information becomes available. To accomplish this, one can use the boring machine’s recorded data to detect any irregularities in the drilling process caused by changes in geological conditions. This research compares various cutting-edge anomaly detection approaches on time series. Due to a large amount of sensor data, the visualization of multiple sensors/features over time was first performed, and the critical features with the highest impact on the detection process for identifying anomalies were selected. Anomaly detection techniques include isolation forest, k-Means, k-Means Sequential Time Series Cluster, Auto-Regression Integrated Moving Average (ARIMA), and Convolutional Neural Network (CNN) Auto-encoders are among the main aspects. The methods presented here were applied to a given data set from an actual tunnelling operation in Germany to locate the location of some concrete slabs in a relatively homogeneous ground. The obtained results agree well with the exact location of anomalies. The performance of various methods is evaluated through error quantification measures.

As-is geometry of existing structures in the built environment can be captured with high accuracy using laser scanning. Frequent measurements and automation of data processing steps allow digital representations of physical assets to be kept up to date at a justifiable cost, even if they are subject to frequent changes. Before operators can execute stationary laser scanning, scan planning has to be performed to estimate the required effort and choose equipment, settings, and locations. In contrast to the conventional, expert-based method usually conducted as an assessment in the field, automated offline approaches aim to solve this task exclusively with pre-existing data describing the scene. These methods are more efficient, add transparency to the existing process landscape, and are a prerequisite for sensible implementation of robotic automation, enabling actual repeatability. The novel method proposed in this paper works in complex 3D environments while considering multiple criteria relevant to the feasibility of acquisition and the quality of acquired datasets. The proposed method introduces the scene as a triangulated mesh within which viewpoint candidates are automatically generated. This mesh is further used in a deterministic approach for visibility and coverage analysis between scene and viewpoint candidates. Based on this analysis, viewpoints are selected to form a solution set that fulfils all pre-defined requirements regarding surface coverage in the scene using a greedy algorithm. Connectivity in the solution is enforced to ensure the captured data will allow targetless registration. The objective function used for evaluating potential solutions allows for consideration of all necessary objectives and constraints in the greedy algorithm while retaining flexibility for applying other solution heuristics and optimization methods.

The building inspection process is an important step in the maintenance phase of a building. However, human resources are limited, and the inspection process is still predominantly a manual process: Damage is documented on paper, recorded with cameras, and manually entered into databases. Digital tools could improve this process, making it more time and work efficient. Continuous advancements in Machine Learning (ML) and Augmented Reality (AR) can support inspectors during damage documentation tasks, allowing for combined capture and documentation of damage.This paper presents an approach to damage documentation with the Microsoft HoloLens 2 (HL2), a head-mounted optical see-through augmented reality device. To this end, we train and review image segmentation models based on over 5,500 images and deploy the best-performing model on the HL2. The segmentation model is trained to distinguish four concrete damage types. Model inference time is compared between the deployment of the ML model on the HL2 and a compute server. The application is tested on-site for a bridge inspection task, investigating the feasibility of the developed AR-ML-based sub-concept for damage documentation.

Public transportation has been reeling under the coronavirus pandemic. To curb the spread of Covid-19 national governments-imposed lockdown regulations at various scales. The transport industry in developing countries bore the initial brunt of lockdowns leading to the grounding of fleets. Ostensibly, very little has been documented on the mechanisms adopted and implemented to develop sustainable mobility solutions in developing countries during the pandemic. Consequently, using the city of Johannesburg as a case study this paper adopted a quantitative research approach to investigate commuters’ perceptions and expectations of the quality of service during the Covid-19 pandemic. Using Supervised Machine Learning techniques, a quality-of-service model was developed to assess the quality of service and inform approaches for sustainable increasing public transport ridership. The results show that there was an increase in retail and recreation-based trips and a decline in work-based trips. This was due to an increase in telework (working from home) during the Covid-19 pandemic. The finding also reveals machine learning techniques can be used to understand commuters’ cognitive decisions or their final outcomes. The trip duration was the most influential feature of the city of Johannesburg also experiments using information gain reveal that increased investment to improve other public transportation features such as reliability and accessibility leads to an increase in public transport ridership. In conclusion, the paper calls for intensified investment in innovative approaches to plan for sustainable public transportation post the Covid-19 pandemic. This can be achieved through upscaling existing uses of technology such as using machine learning in scenario planning.

Industries like manufacturing use Machine Learning (ML) algorithms to conceive and produce excellent consumer goods. This achievement has persuaded other economic sectors, including the construction sector, to attempt and incorporate intelligent algorithms. The most recent developments in ML algorithms have made it possible to automate those non-trivial jobs that were thought unsolvable years back. Early involvement of Construction researchers in the ML process is necessary to ensure that they have sufficient awareness of the advantages and disadvantages. It is worthy of note that construction organisations have concerns due to the peculiarity of the sector. As such, adopting machine learning (ML) for profitability predictions or cost-saving results can be challenging. Construction industry stakeholders are eager to discover how ML may help improve operations, and the benefits of ML algorithms, among others, before adopting these algorithms for decision-making. To assist construction industry stakeholders in the adoption of ML algorithms, the study adopted a systematic literature review. The study helps in the proper identification of the uses of ML algorithms to improve the construction industry processes and product.

This study presents an algorithm for identifying potential points of failure (leakage) in water distribution systems by evaluating the risk of pipe burst at each node. The algorithm calculates a burst index for each node using burst factors derived from the average District Metered Area (DMA) burst pressure and relevant parameters such as pipe length, age, and thickness. A sensitivity analysis based on pressure observations at the nodes under different leakage scenarios is conducted to identify deviations from the no-leak scenario. DMAs are then ranked according to their total burst risk, and nodes are partitioned based on their average DMA burst pressure. Although this algorithm may generate a larger number of candidate nodes than necessary due to the lack of optimization, its output can be used as input for sensor location optimization algorithms, reducing the search space for these algorithms.

Houses, the most constructed structures all over the world, can be exposed to extreme winds that may damage their roofs. Generally, roofs are the most vulnerable parts of houses against wind pressure rather than the other elements. Most housing roofs in South Africa are constructed with metal sheeting or tiles as covers, while these systems can exhibit different performance upon their exposure to extreme wind events. The present study is aimed to provide an assessment on the structural behaviour of housing roofs comprising metal and tiled systems under strong winds. Fluid dynamics, wind codes and computational fluid dynamics (CFD) have been employed to analyse and quantify wind loads. Combined wind and dead loads at ultimate limit state served as quasistatic forces acting on roof samples during laboratory experiments. Findings show that cover-to-truss connections are being pulled-out as the sheet deflects under uplift wind forces, even to the removal of batten-to-truss nails when there are no clips. Tiles are cracked gradually till they break as the whole system deflects under wind forces. It can be concluded that the performance of metal sheeting roofs depends on the resistance of cover-to-truss connections against wind pressures, whereas that of tiled roofs depends on the individual resistance of each tile, their interlocking forces, their ability to act as a system, and their attachments’ resistance at the edges.

The designing of structures is an intricate process during which various aspects should be considered, one being fire resistance. Conventional materials such as concrete help structures withstand the effects of a fire. A composite sandwich flooring system, referred to in this paper, is completely void of concrete and has limited inherent fire resistance that is provided. For a flooring system like this, heat transfer will occur more rapidly. Fire-resistant boards are used to protect the system, but the fire rating of these boards is compromised by service holes—which is the focus of this paper. Experiments were conducted to examine the rate of heat flow through a sandwich ceiling system consisting of a Calcium Silicate (CaSi) board on the exposed face, a Voidcon steel sheet in the middle, and a Fibre Cement Board (FCB) on the unexposed face. Various holes, to simulate light and service penetrations, were made in the ceiling system. The experimental results made provision for a 30-minute rating for all the tested samples, with the shortest test taking place over approximately 40 minutes, but this is significantly less than the 60-minute original resistance. The experimental results were compared with simulated results generated using ABAQUS. This paper presents a summary of the comparison between the experimental and simulated results, highlighting important considerations and behaviour.

The design of detached houses involves multiple choices to meet many criteria, such as energy performance, and zoning regulations. Many design factors, including the form of the house, often come into play making the design process more complicated and time-consuming. The number of designs and simulations performed is usually limited due to time and cost constraints. This study aims at proposing a Generative Design (GD) framework to automate the design process of detached houses, and simultaneously optimize many design aspects. Numerous design parameters mainly relating to the house geometry and its energy efficiency were included in the GD study. The GD framework was developed in Dynamo, an Autodesk Revit internal generative design tool that uses the Non-Dominated Sorting Genetic Algorithm (NSGA-II). The floorplan boundary lines having variable dimensions were first created in Dynamo to serve as a reference for walls, floors, roof, and other geometric components. Several Dynamo nodes such as “Walls.ByCurveAndLevel”, “Floor.ByOutlineTypeAndLevel, “Roof.ByOutlineTypeAndLevel”, and “FamilyInstance.ByFace” were utilized to generate houses with variable walls, floors, roof, windows, and doors, respectively. Afterward, the Dynamo graph was ready to be exported to create and run the GD study and generate different feasible design solutions. Preliminary results included a fully automated design of a single-family house envelope. The designer can run the GD study to generate, compare, and explore different design options, examining the geometry and analysis results to select a final design solution. The findings of this study will maximize the productivity of designers/developers and will tremendously reduce the financial strain and time consumed designing energy-efficient single-family houses using traditional techniques.

Stakeholders in the construction industry have over the years made frantic efforts in seeking solutions to the problems facing the industry. The advent of technological innovations such as machine learning applications seek to abate some of these challenges by modernizing construction processes and activities, and ultimately improving on construction projects delivery. This study seeks to examine the propelling factors for the adoption of machine learning applications in the construction industry. Data gathered was subjected to appropriate data analysis techniques. Findings from the study revealed that the most significant drivers for the adoption of machine learning applications are the fast changing, field based and project nature of the construction industry, and the need for accurate results. Also, it was revealed that there is no difference among the different professionals’ view of the drivers of machine learning applications in the construction industry. The study made recommendations that would aid the integration of machine learning applications in construction activities for better and more efficient processes in construction project delivery.

Buildings account for a large proportion of energy consumption, and improving building energy efficiency during the operation phase has attracted increasing research attention to achieve the carbon neutrality goal. Building energy simulation is a powerful tool to predict and manage building energy performance during the operation phase. Decision making can be informed by timely and reliable simulation results. However, building energy simulation requires various data and information including building geometries, thermal properties of constructions, Heating, Ventilation, and Air-conditioning (HVAC) systems, etc. Collecting these data and information from different sources (e.g., Building Information Modeling (BIM), Building Management Systems (BMS)) can be a tedious and time-consuming job, which limits a timely prediction and decision-making. This study proposes an ontology-based framework which can integrate data for building energy simulation from different data sources. Firstly, we collect and integrate four types of data (i.e., weather, building, internal heat gain, and HVAC system) from different sources. Ontology models are designed by integrating existing ontology models Brick Schema and Building Topology Ontology (BOT). An inference rule for thermal zoning is proposed. Secondly, a group of rooms in a campus building are selected as a case study to demonstrate the implementation of the models and the inference rule. The proposed models reduce the efforts needed to collect and integrate data for building energy simulation during the operation phase. Using the proposed model, data needed for building energy simulation can be obtained promptly and accurately, which strongly supports building energy management towards energy efficiency.

In the AEC sector, energy performance targets of buildings continuously increase for contributing to reduce carbon dioxide. This is usually done on building level, but the focus continuously shifts to larger scales such as neighborhoods, e.g. for identifying buildings with the most retrofitting potential. For this, low detailed GIS models can serve as a basis for energy simulations and are broadly available. However, neighborhood energy simulations hold many challenges, such as the lack of accurate and sufficient data to perform reliable simulations. Information such as window positions or thermal parameters of the building elements can thereby help to increase the quality of the energy simulation results. Therefore, in this paper, challenges of data collection are presented and discussed. To enable users to find a trade-off between accuracy and reliability of a neighborhood simulation and the effort to provide this data, the authors developed the concept of the Neighborhood Model States (NMS). Furthermore, occurring challenges in enriching the GIS model for each NMS are discussed on the example of buildings from the UBC campus.

Housing inspection is an essential task to ensure quality delivery during construction operations, in accordance with relevant building manuals and standards. It is critical to ensure that customers are satisfied with housing delivery by monitoring construction activities per stage inspections. The evidence gathered from housing construction inspectors as well as from previous studies suggests that the effectiveness and efficiency of inspection processes currently in the South African construction industry are unsatisfactory. This paper reports on South African research that explores the feasibility of adopting an automated inspection approach to incorporate digital technologies such as 3D laser scanners, drone technologies, and Building Information Modeling (BIM) into the housing construction inspection process to facilitate more effective and efficient data collection, processing, and reporting for improvement of the inspection process. The paper also discusses the interview results of housing construction inspectors about the possibility of adopting an automated inspection approach. The results show that the automated approach with the new digital technologies has the potential to improve housing construction inspection for the delivery of quality housing. The research efforts will also enhance the current knowledge of the housing construction inspection process.

Healthcare facilities play a key role in responding United Nations goals, such as sustainability, health and welling. The outbreak of the COVID-19 epidemic has driven much attention to expanding healthcare capacity through advanced digital technologies, such as Building Information Modelling (BIM). Nevertheless, a systematic review of research achievements is lacking. This research uses bibliometric and systemic literature review methods to investigate BIM applications in Healthcare Design and Construction (HDC). The bibliometric investigation focuses on country, journal co-citation, and keyword clustering analyses. The systematic review classifies application domains, BIM actions, and other digital technologies accompanying BIM. Finally, 17 major BIM actions are summarized for six major domains, including operability, resilience, collaboration, sustainability and constructability. This study reveals that the outbreak of COVID-19 has greatly stimulated the academic interest in digital technologies for HDC, and there is geographical uniqueness highly relevant to local government policies and national healthcare services. However, related research is still in a relatively preliminary stage.

The 21st-century industry world is constantly seeking diverse ways to shrink costs and time while boosting productivity and efficiency. Utilising digital twin technology is a veritable means of achieving that in the building sector. This will help better predict the future, enhancing decision-making and, in turn, reducing operational cost and downtime while simultaneously enhancing building efficiency and productivity. This study, therefore, investigates ways that digital twin technology can be used to make facility management systems proactive in nature. The study is designed to follow a methodical review of literature. It draws relevant data and information from extant studies conducted on digital twin technology within the built environment field as well as the entire field of science, technology, and engineering. A framework for management was conceptualised through this research and named ‘Digital Twin Based Smart Management Plan’. A strategic process for the Smart Management Plan was developed and classified into initiation, modelling, utilisation and reuse phases. The Digital Twin-Based Smart Management Plan framework ensures complete interaction among the process, people, place, and device.

This paper presents a new approach for modeling construction state inferencing rules using Semantic Web ontologies. This approach focuses on facilitating the shareability of the rules with ontology formalized meta information and SHACL-based rule body. Meanwhile, using the DiCon ontology as a unique terminology box, the rule could be reused directly for different construction data sources. The modeled rules would thus be collected as a shared library so that different users could search and reuse the rule in the library. The proposed ontology and CPL framework modeled were tested in an example case to demonstrate the usage of the ontology and framework.

This study aims to identify indicators to measure the productivity of building information modeling (BIM) through a critical literature review of previous studies on BIM productivity and the factors used in the studies. Measuring BIM productivity, to which end quite a few efforts have been made, is important for efficient workforce management. The authors reviewed 14 papers collated from Scopus and Google Scholar. Examples of topics included productivity factors related to 2D/3D design modeling. Previous studies on BIM productivity have followed the definition of productivity as output/input. The input and output of BIM productivity indicators were classified according to the research purpose of the 14 previous studies. Among the BIM productivity indicators suggested in 14 previous studies, the most frequently used BIM productivity indicators were output as time and input as the number of elements in the 3D model. Among the factors affecting BIM productivity, modeling time was the most commonly considered. Other factors included model quality, modeling behavior, project size, model size and model review.

As part of the digitization of the AEC industry, the Digital Twin concept is becoming increasingly important. Originating in the manufacturing industry, the concept at its core involves a bidirectional coupling of the physical product and its digital counterpart with the aim of keeping the two in sync. Without appropriate capabilities to realize such synchronization, the concept always remained as an unattainable vision for the AEC industry. Adapting additive manufacturing (AM) for construction, however, creates unique opportunities to realize this vision by enabling automation in both directions, from digital to physical product and vice versa. As a fully automatable manufacturing method where robotic processes are typically controlled by the digital representation of the product, AM realizes the digital-to-physical link for this purpose. Conversely, based on the same digital representation of the product, the acquisition of the physical implementation of the manufacturing process can be automated, enabling the physical-to-digital connection. This paper uses three AM application scenarios to illustrate, on the one hand, the need for automating quality control and, on the other hand, to describe approaches for its realization. In particular, the benefits of synergy between automated quality control (QC) and fabrication information modeling (FIM) to form a digital-physical-digital loop are explored.

This paper shows a design structure schema for highway route design based on the conceptual pyramid information model and the connector model in previous studies. In the design structure schema, the design process is classified into four stages: Conceptual Design, Outline Design, Basic Design, and Recursive Design, and it shows that the highway design process can be viewed as a function of the designer’s thought process. Moreover, a method for analyzing the fluctuation in highway distance was proposed. A distance-conversion table for each measurement that lengthens the highway can be linked to the dispositional information. Using this method, we can start compiling knowledge from the beginning about changes in distance and will be able to reproduce the route-selection process in a system.

The paper reports on the buildingSMART International project IFC-Tunnel that is developing an extension of the vendor-neutral data exchange standard Industry Foundation Classes (IFC). The paper highlights the importance of a well-defined development process and the involvement of international domain experts. It discusses in detail the requirements analysis conducted as well as its outcome in terms of the tunnel types and the use cases covered. It subsequently reports on the conceptual model that includes all proposed extensions and modifications, as well as the alpha version of the EXPRESS encoding which will be integrated into future versions of the IFC standard.

Information-rich BIM models are rarely usable off-the-shelf for operations tasks. Change decisions made on the construction site can lead to significant differences between the as-designed and as-built state of buildings. The responsibility for keeping the digital representation in sync with its physical twin is not defined and will likely only fully be assigned when automatic methods facilitate the geometric update process. To this end, previous research succeeded in (1) identifying if an element was erected at the time and position it was initially designed, and (2) updating the parametric design geometry to fit its LiDAR-measured as-built state under a set of assumptions and threshold values.The research presented in this paper aims at updating the as-designed model in case of significant pose differences between the as-designed and as-built state. The method leverages graphs to encode the topological connectivity between geometric elements, once for the as-designed BIM model and once for the as-built point cloud. A similarity metric, namely the cosine distance, allows for a quantitative comparison of the topologically enriched point cloud clusters and their corresponding BIM element. The results show that a convincing type-wise similarity can be found in the feature space between the as-built point cloud clusters and the BIM elements. This similarity score becomes meaningful once the element’s topological arrangements are included. An instance-wise similarity score of above 90% is achieved for matching-pairs of free-standing columns and allows for a large-scale pose update in the as-designed BIM model.

In recent years, digital twins have become a more significant strategic trend in the construction industry. Stakeholders in the industry view it as a technology-driven innovation that has the potential to support the design, building, and operation of constructed assets, alongside advancements in other new-generation information technologies such as the Internet of Things (IoT), artificial intelligence (AI), big data, cloud computing, and edge computing. However, the construction project context generates various organizational and functional information through model-based domain-specific information models that require integration and analysis. Furthermore, commercial technologies enable the integration of real-time data sources with building information models (BIM), but these tools are often proprietary and incompatible with other applications. This lack of interoperability among heterogeneous data formats is a major obstacle to the reliable application of digital twins in the construction industry. To address this challenge, this study presents a multimodel framework developed using Information Container for Linked Document Delivery (ICDD) that can integrate multiple data models from autonomous and heterogeneous sources, including real-time data sources, in their original format at the system level. This framework enables stakeholders to analyze, exchange, and share linked information among the built asset stakeholders, relying on linked data and Semantic Web technologies.

Building information modelling (BIM) adoption in the construction industry has become a fundamental ingredient for best practices globally. The benefits have been established In many BIM leading countries. However, due to many challenges, most developing countries are still struggling to achieve industry-wide diffusion. On this premise, this study assessed the profound constraints to the adoption of BIM in the South African construction industry. The study identified the key constraints to BIM adoption in the South African construction industry through a quantitative research approach. It also identified the critical stakeholders that must drive widespread BIM adoption in the South African construction industry. The data was collected from construction industry professionals using a well-structured questionnaire. The key constraints to BIM implementation in the South African construction industry were identified based on the data collected and analysed. The study also identified the key stakeholders for BIM implementation in the South African context. Among other recommendations, the study recommends that the government provide a conducive business environment to enable the implementation by stakeholders.

High-quality BIM model is the premise to ensure the effective application of BIM technology. Thus, many countries and organizations have developed standards to clarify the requirements of BIM model quality on data completeness, consistency, and correctness. However, due to the complexity of BIM model and the professionalism of standards, checking BIM model quality manually is challenging and time-consuming, so it is necessary to develop automatic check systems for BIM model quality. To develop the system, computerized representation of BIM model quality standards is the key and basic step. Towards the objective, this paper proposes an ontology-based computerized representation method for BIM model quality standards. In this method, OWL is applied to represent the concepts and relationships, and SWRL is used for representing BIM model quality rules in the standards. After representing the Chinese BIM model quality standard, a rule base composed of 41 SWRL rules is constructed and the effectiveness of the method is verified. This research contributes to the development of BIM model quality automatic checking system and the promotion of BIM technology.

Fire is a major hazard for any building. Therefore, fire protection engineer can take various measures to prevent the spread of fire and smoke. One of the most effective fire safety measures is dividing a construction facility into separate fire and smoke compartmentation, which represents an essential part of integral fire protection planning. Fire compartmentation consists of a single room or a series of related rooms and have to be integrated directly into the construction structure. In this context Building Information Modeling (BIM) or more specifically the open-BIM standard Industry Foundation Classes (IFC), which can represent a digital building model with its structure by describing geometry, materials, and relationships between objects, can be used for the digital management of fire compartmentation. In this paper, we present an approach for representing fire compartmentation using the IFC data structure, which has been triggered by investigating current scope of standardization work pertaining to fire safety. By using the IFC data structure for the exchange of information, a reliable basis for the consideration of fire compartments is created, thus preventing expensive compensation measures as well as facilitating cooperation between different designers. Throughout the paper, we explore the possibilities, advantages and challenges of this approach.

The traditional quality control on construction site relies mostly on visual inspections and daily or weekly reports generated based on those inspections. This traditional practice relies heavily on the inspectors’ personal judgement, detailed knowledge of the project, observation, and experience, resulting in a high probability of incomplete and inaccurate reports.Over the past few years, the architecture, engineering, and construction industry (AEC) recognized the urgent need for fast and accurate control of construction quality to increase productivity and optimize construction costs by avoiding the rework of deficiencies. Digital tools, including the BIM model and the digital twin of the construction site, will improve productivity and should help to bring down the costs of non-quality. But they require detailed knowledge of digital practices and existing processes in construction, which is the mandatory condition for the construction industry to embrace new technologies through rationalizing its production processes. An “end user” approach needs to be led with on-site workers to understand the issues they face and how they currently work to understand better their needs regarding quality monitoring and how the digitalization of associated processes could help. This article focuses on the methodology used to conceive a digital twin for the construction site where construction companies (end users) were put at the heart of the approach. It also drives novel insights on quality control and discuss future directions to improve and automate quality control at construction sites.

Accurate green building assessment (GBA) represents one of the best opportunities to understand the holistic sustainability strengths and weaknesses of existing buildings to inform their retrofitting decisions. However, the current process for GBA of existing buildings is very challenging, tedious, complex, time-consuming and costly, and suffers from lack of important data and information. Moreover, most GBA results are not leveraged to retrofit and improve the sustainability performance of existing buildings – they are mostly for just recognition and market edge. To address these limitations, this study aims to develop a framework for using Digital Twin (DT) technology to automate and improve GBA. Although unavailable static building data can be obtained from scan-to-building information modelling (BIM) process, real-time dynamic data cannot. Hence, real-time dynamic data from the internet of things (IoT) sensors and other data should be integrated into the BIM model to create the DT model of the building. A plug-in software can then be deployed to assess the sustainability performance level of the building within the DT environment automatically. The framework is based on the Building Environmental Assessment Method (BEAM) Plus, which is Hong Kong’s leading GBA system. A real DT should feedback into the physical twin after receiving and processing data from it. Therefore, the automated GBA results should inform retrofitting decisions of the physical building. This study contributes to the understanding of how DT can be used to automate and improve GBA, and how the results can be used to improve retrofitting decision-making.

With the increasing availability of equipment to automatically monitor on-site construction, construction management should change considerably in the future. A data-driven construction management should be a system which receives periodically data from the field, reasons over construction information present on BIM models, and has a representation of the possible outcomes based on input data / information. Such scenario is one of the uses of Digital Twins for Construction. Instead of instrumenting the construction site, one option is to create “smart building components”, by attaching to each one of them an IoT device which will report their own state over time. Analyzing patterns on data from IMU and RSSI, as well as GPS when on open field, it is possible to identify and quantify movement and waiting times, as well as the precise moment in which a component installation happened – a micro-management more akin to Lean Construction. Although simplistically, this representation could be a prototype for a Discrete Event Simulation (DES). Construction progress monitoring thus become an automatic and remote process that could send signals and drive a representation of the process itself. Employing Location-Based Management System and Building Information Modeling, we create an ecosystem that takes advantage of the IoT solution to register and inform the decision makers over construction progress, allowing management of resources to remain on time and on budget with the advance of construction. In this paper experiments in real environments are presented with the automatic progress report of framing installation in the façade of a building.

In construction, simulation can provide production planners with forward-looking or predictive situational awareness of the potential impact of proposed changes before implementation. Planners can experiment extensively with various alternative production plans and systems without suffering real-world consequences of failure. Addressing the need to have proper control of the jobsite, DTC is a model for managing production in construction that leverages data streaming from different monitoring technologies and artificially intelligent functions. Overall, DTC offers accurate project status information (PSI) and proactive analysis and optimization of ongoing design, planning, and production processes. The integration of automated monitoring and information integration algorithms contemplated within the DTC framework may be able to provide the kind of information needed for practical simulation at short intervals, thus offering construction planners a powerful tool to optimize the decision-making process regarding any necessary changes to designs or plans, by automatically generating accurate and reliable simulation models based on the current jobsite progress, resource information, and safety conditions. This paper describes an automated system for parametric generation of simulation models for this purpose from project intent and status information stored in a DTC database. This is one aspect of broader research that involves design, development and testing of a DTC simulation and optimization system. A construction case study is provided to demonstrate the technical feasibility of automatically and parametrically producing simulation models based on data from a digital twin.

With the increasing adoption of the Digital Twin concept in the construction industry in the operations and maintenance phase, researchers and practitioners are increasingly seeking suitable technological solutions for the design and construction phases. While it is widely accepted that the required platforms hosting the digital twin must be cloud-based to fulfill the requirements of ubiquitous accessibility and centralized consistency, questions regarding the need for data schema remain. Some academics argue that a structure-free organization of data is suitable for realizing digital twins and the data streams from and to the respective platform. Hands-on experience in the BIM2TWIN project supports a counter argument, i.e., that structure-free data is insufficient for most use cases around AEC Digital Twins. The sheer information complexity of construction projects requires well-defined data structures enabling unambiguous and error-less interpretation. This becomes apparent when reflecting on the well-established concept of the data-information-knowledge pyramid describing that raw data must be processed into understandable and meaningful high-level information for human decision makers, subsequently providing the basis for cross-project domain knowledge. Based on this observation, we highlight that object-oriented modeling is a widely recognized information modeling technique that facilitates the structuring of complex domain information. We compare it with ontology-based model concepts that provide a similar, yet more abstract means for information modeling.

Construction sites are dynamic and complex systems with significant potential for time and cost efficiency improvement through digitization and interconnection. Construction 4.0 is the use of modern information and communication technologies known from Industry 4.0 (I4.0) to interconnect construction sites with cyber-physical systems (CPS). In these decentralized systems, construction workers, machines, and processes become smart I4.0 components that can exchange data and information with each other in a decentralized and self-controlled manner. The basis for informed decision-making to control and optimize relevant on-site processes is real-time detection of the current construction progress and the machines used on the construction site. Computer vision (CV)-based tracking systems offer a technical solution that can reliably detect construction workers and machines and track construction progress and processes. These tracking systems generate large amounts of data that must be processed and analysed automatically. The goal is to integrate the tracking system into the CPS as an I4.0 component. Essential for this is the digital twin as a virtual representation of an I4.0 component that centrally collects, processes, and provides data for the respective component. This paper presents an I4.0-based digital twin approach for digitizing and interconnection of the construction site into a CPS. The approach integrates a CV-based tracking system as an I4.0 component to locate construction equipment on the construction site. The tracking system is a multi-camera multi-object tracking system that uses stereo vision cameras and a real-time capable detector. The asset administration shell (AAS) is used as the platform for the digital twin.

For efficient and sustainable use of precast concrete modules, all relevant information must be collected digitally and in real-time and made available in digital twins. Digitization should be carried out in such a way that continuous quality management is possible at all times. This also includes whether the produced concrete modules also meet all the requirements from the initial design. For example, the precast concrete parts must be able to absorb certain forces or have precise connections and joining options. The Requirements Interchange Format (ReqIF) can be used to describe requirements digitally and exchange them between different IT systems and stakeholders. The creation of automated quality control (QC) protocol for the flow production process can be implemented based on this already structured and formalized requirements format. In this paper, the Asset Administration Shell (AAS) from the context of Industry 4.0 is enhanced to enable the formal description and automated verification of requirements for precast concrete based on the ReqIF interchange format. For this purpose, a smart service integrates the ReqIF-compliant requirements into an AAS submodel. Via this smart service, a mapping assistance tool lets stakeholders assign measurable properties of the precast concrete modules to the requirements, thus enabling an automated quality check. The presented approach is validated based on a virtual precast concrete wall for which a chain of linked requirements is described and automatically checked within the scope of a case study.

Building Information Modeling (BIM), Lean Construction Management (LCM) and many other methods are already used to optimise the processes and increase the efficiency in the construction industry. Construction site management can be digitalised using common Industry 4.0 technologies, like smart devices and assets (tablets, smartphones, sensors, beacons, etc.). Mobile devices are nowadays an essential tool in our everyday life, and applications that can be used on construction sites support staff in organisation and communication. Furthermore, they can aid with project control and monitoring, as well as with supply chain management processes.This paper presents a concept for connecting precast structural material with integrated RFID tags with BIM models in real time by using electronic readers and mobile device applications. In order to connect the physical components with their virtual or digital representations, the attributes of BIM models were used, into which the unique ID of the RFID transponder were implemented.As a novel approach, BIM models were connected wirelessly to physical building materials by using RFID and wireless IoT technologies in a cross-platform application, thereby enabling the BIM models to be actively used throughout the life cycle of a building (planning, production, construction, maintenance and operation). Application areas include material tracking, warehouse inventory management, transportation planning, real-time model-based scheduling and model based navigation.