Construction site layout planning (CSLP) is a typical problem in construction site management that urgently needs to be solved. A reasonable construction site layout can improve efficiency and reduce costs and unnecessary loss of time. To solve the CSLP problem, some emerging optimization algorithms and enhanced mathematical models have been developed. The building information model (BIM) provides data support for the implementation of previous research. However, manually extracting and modifying parameters from the BIM for use in optimization algorithms is cumbersome in practical applications, wasting considerable time and human resources. In this paper, we propose a framework of automated prefabricated building CSLP to realize the integration of the BIM and an optimization algorithm, which utilizes the Dynamo visual programming platform to create an information channel between the BIM and the optimization algorithm. A typical and practical case, which is optimized by the genetic algorithm (GA) and particle swarm optimization (PSO), is demonstrated to validate the proposed approach and to show an effective reduction in material transportation cost and unnecessary loss of time.

Limited to the current building information model (BIM) development stage, many engineers are still using CAD 2D drawings for top-down design, and then using drawings to generate BIM models for BIM-related applications. As the manual method of creating a semantic-rich BIM model is time consuming and could cause model defects, engineers are reluctant to apply the semantic-rich BIM model. To solve this problem, a semiautomatic methodology is proposed to generate a semantic-rich structural BIM model from two-dimensional (2D) computer-aided design (CAD) drawings. First, it uses the information classified by CAD layers to generate geometric models of beams, slabs, and columns. Then, the semantic information is appropriately structured and stored in the corresponding element models. Finally, the feasibility of this methodology is verified with two case studies of frame structures. The case studies show that this method can quickly and accurately create a structural BIM model with many elements that reflect the structural details described in the construction drawings. Nearly all text information in the construction drawings is accurately imported into the structural BIM model, and additional parameter spaces are reserved to facilitate the expansion of building information. The engineers in charge of creating a BIM model could benefit from this proposed methodology.

The location information of entities in construction sites, such as workers and construction machines, is valuable in project management and safety. Therefore, as nonintrusive and accurate solutions, various vision-based methods have been proposed to track entities in construction sites and obtain their three-dimensional (3D) coordinates. However, most existing vision-based methods realize 3D localizations by basing entity matching on the epipolar line, which brings instability in entity matching due to the calculation error of the epipolar line or failure to match entities when multiple entities are located on the same epipolar. To solve this problem, a novel framework based on reidentification is proposed to automatically match workers across two camera views, thereby obtaining their 3D coordinates in construction sites. In this framework, deep-learning-based computer vision algorithms are firstly used to detect and track workers in two camera views. Then, the reidentification (ReID) algorithm is applied to utilize tracked workers’ visual features to match the workers across both two camera views and different frames. As a result, for every matched pair, the worker’s pixel locations in two camera views can be obtained to calculate the 3D coordinates through triangulation. The implementation of videos recorded from a real construction project proves the feasibility and accuracy of this framework. Specifically, through utilizing the ReID algorithm to match workers, the framework achieves competitive results on workers matching with precision, recall, and accuracy of more than 99%, 93%, and 93%. Furthermore, it also effectively addresses the practical problems of ID repetition and ID switching. Meanwhile, this paper extends the application scenarios of reidentification algorithms in construction sites, thereby contributing to the future application of multiple-camera vision-based methods in construction sites.

Cyber–physical systems (CPSs) are generally considered to be the next generation of engineered systems. However, the actual application of CPSs in the Architecture, Engineering and Construction (AEC) industry is still at a low level. The sensing method in the construction process plays a very important role in the establishment of CPSs. Therefore, the purpose of this paper is to discuss the application potential of computer vision-based sensing methods and provide practical suggestions through a literature review. This paper provides a review of the current application of CPSs in the AEC industry, summarizes the current knowledge gaps, and discusses the problems with the current construction site sensing approach. Considering the unique advantages of the computer vision (CV) method at the construction site, the application of CV for different construction entities was reviewed and summarized to achieve a CV-based construction site sensing approach for construction process CPSs. The potential of CPS can be further stimulated by providing rich information from on-site sensing using CV methods. According to the review, this approach has unique advantages in the specific environment of the construction site. Based on the current knowledge gap identified in the literature review, this paper proposes a novel concept of visual-based construction site sensing method for CPS application, and an architecture for CV-based CPS is proposed as an implementation of this concept. The main contribution of this paper is to propose a CPS architecture using computer vision as the main information acquisition method based on the literature review. This architecture innovatively introduces computer vision as a sensing method of construction sites, and realizes low-cost and non-invasive information acquisition in complex construction scenarios. This method can be used as an important supplement to on-site sensing to further promote the automation and intelligence of the construction process.