Augmented reality can improve construction and facility management by visualizing an as-planned model on its corresponding surface for fast, easy, and correct information retrieval. This requires the localization registration of an as-built model in an as-planned model. However, the localization and registration of indoor environments fail, owing to self-similarity in an indoor environment, relatively large as-planned models, and the presence of additional unplanned objects. Therefore, this paper proposes a computer vision-based method to (1) homogenize indoor as-planned and as-built models, (2) reduce the search space of model matching, and (3) localize the structure (e.g., room) for registration of the scanned area in its as-planned model. This method extracts a representative horizontal cross section from the as-built and as-planned point clouds to make these models similar, restricts unnecessary transformation to reduce the search space, and corresponds the line features for the estimation of the registration transformation matrix. The performance of this method, in terms of registration accuracy, is evaluated on as-built point clouds of rooms and a hallway on a building floor. A rotational error of 0.005 rad and a translational error of 0.088 m are observed in the experiments. Hence, the geometric feature described on a representative cross section with transformation restrictions can be a computationally cost-effective solution for indoor localization and registration.

Evaluating the site-selection process for photovoltaic (PV) plants is essential for securing available areas for solar power plant installation in limited spaces. Although the vicinities of highway networks can be suitable for installing PV plants, in terms of economic feasibility, they have rarely been investigated because the impacts of various factors, including geographic or weather patterns, have not been analyzed. In this respect, this study conducts a case study on selecting the site for PV-panel installation in the vicinity of a highway (e.g., slopes) by integrating geographic information system (GIS) and building information model (BIM) techniques. Using location (e.g., highways, lakes, rivers), monthly solar power output, and orographic (e.g., slope) data, suitable regions are identified with the geo-spatial analysis; then, the amount of power that can be generated is evaluated in the regions. For the proposed sites, the surface conditions and potential PV systems are transformed and visualized in a BIM environment. In the results, the power output at optimal sites selected from the case area was computed at a total of 8227 MWh and was transformed into solar-panel families in three-dimensional environments. Thus, this study may help to identify optimal sites for PV plants in the unused areas near highways.

The performance of a safety management system (SMS) is influenced by a large number of internal and external factors. Loss of control of these factors can create unsafe conditions, which can eventually lead to worksite accidents. Identification and prioritization of these factors is crucial for optimal SMS performance. However, because of the complex and dynamic nature of accident occurrence, identification and control of critical SMS factors and accident precursors remains difficult in practice. This paper proposes the use of a risk-rating approach for understanding the complexity between SMS factors and accident precursors from a holistic perspective. This approach allows for not only the identification of high-priority SMS factors and accident precursors, but also the evaluation of associations between these groups. To demonstrate the effectiveness of the proposed approach, risk ratings for a comprehensive set of SMS factors and accident precursors were collected from 96 construction professionals in Alberta, Canada. The results demonstrated that contract schedule, skilled worker availability, and change orders were the highest-priority SMS factors in the region. Results also found the highest-priority accident precursors to be workers’ failure to identify hazards, congestion, and workers’ neglect of hazards. Associations between factors and precursors were also examined and were found to be present between several SMS factors and accident precursors. Overall, this method allows practitioners to identify high-priority accident precursors and to determine which SMS factors are associated with these precursors. These findings can assist practitioners in determining which SMS factors should be controlled if certain accident precursors are observed and, conversely, which accident precursors are most likely to occur if critical SMS factors are present on site.

The application of deep learning (DL) for solving construction safety issues has achieved remarkable results in recent years that are superior to traditional methods. However, there is limited literature examining the links between DL and safety management and highlighting the contributions of DL studies in practice. Thus, this study aims to synthesize the current status of DL studies on construction safety and outline practical challenges and future opportunities. A total of 66 influential construction safety articles were analyzed from a technical aspect, such as convolutional neural networks, recurrent neural networks, and general neural networks. In the context of safety management, three main research directions were identified: utilizing DL for behaviors, physical conditions, and management issues. Overall, applying DL can resolve important safety challenges with high reliability; therein the CNN-based method and behaviors were the most applied directions with percentages of 75% and 67%, respectively. Based on the review findings, three future opportunities aiming to address the corresponding limitations were proposed: expanding a comprehensive dataset, improving technical restrictions due to occlusions, and identifying individuals who performed unsafe behaviors. This review thus may allow the identification of key areas and future directions where further research efforts need to be made with priority.

Workers in the construction industry are frequently exposed to physically demanding manual tasks with a high level of ergonomic risk. To prevent ergonomic injuries and disorders, posture-based ergonomic evaluation methods, which require inputs describing the worker’s posture (e.g., body joint angles), have been developed and are used widely in practice. However, the reliability of these ergonomic methods has not been investigated fully from the input measurement perspective, as when collected by a human observer or motion capture sensors, and which may inevitably contain measurement errors (e.g., human perception errors and sensing errors in estimating human postures). Thus, this study examines the imprecision associated with acquiring the required inputs for ergonomic assessment and investigates its impact on the final result of the analysis. The two primary methods of obtaining the inputs of posture-based evaluation tools, i.e., human observation and recordings of motion sensing devices, were examined, and a stochastic approach was proposed to evaluate the impact of the input errors on the final result of the ergonomic assessment. The findings indicate that the imprecision associated with inputs from both visual observers and sensors have significant impact on the results of ergonomic analysis and should be incorporated into the analysis accordingly. The contribution of this paper allows practitioners and researchers to understand possible ranges of outputs that can be caused by observation and measurement errors, and to determine allowable tolerance of sensing errors required for ergonomic evaluation through stochastic analysis.

Although the use of safety-related measures (SRM) has been advocated in the literature, construction companies often rely on reactive indicators instead of SRM to evaluate safety performance. A challenge with using SRM in practice arises when faced with the uniqueness of construction organizations. Distinctive practices and cultures affect the availability and quantity of data, which in turn limits the reliability of conventional SRM-based evaluation methods. This study proposes a case-based reasoning (CBR) approach that makes efficient use of existing SRM available within an organization to reliably evaluate safety performance. The model was validated using statistical tests to compare model output with actual performance. To test model applicability, we conducted a case study: based on information available in an industrial construction organization, including schedule performance and payroll data, 27 measures were identified. The results demonstrated that the proposed approach is able to evaluate safety performance trends in consideration of unique organizational characteristics using SRM. We expect that the model approach and results will support managers in tailoring safety goals and programs to particular contexts.