Optimally leveraging depth features to enhance segmentation of recyclables from cluttered construction and demolition waste streams.

This paper addresses the critical environmental issue of effectively managing construction and demolition waste (CDW), which has seen a global surge due to rapid urbanization. With the advent of deep learning-based computer vision, this study focuses on improving intelligent identification of valuable recyclables from cluttered and heterogeneous CDW streams in material recovery facilities (MRFs) by optimally leveraging both visual and spatial features (depth). A high-quality CDW RGB-D dataset was curated to capture MRF stream complexities often overlooked in prior studies, and comprises over 3500 images for each modality and more than 160,000 dense object instances of diverse CDW materials with high resource value. In contrast to former studies which directly concatenate RGB and depth features, this study introduces a new depth fusion strategy that utilizes computationally efficient convolutional operations at the end of the conventional waste segmentation architecture to effectively fuse colour and depth information. This avoids cross-modal interference and maximizes the use of distinct information present in the two different modalities. Despite the high clutter and diversity of waste objects, the proposed RGB-DL architecture achieves a 13% increase in segmentation accuracy and a 36% reduction in inference time when compared to the direct concatenation of features. The findings of this study emphasize the benefit of effectively incorporating geometrical features to complement visual cues. This approach helps to deal with the cluttered and varied nature of CDW streams, enhancing automated waste recognition accuracy to improve resource recovery in MRFs. This, in turn, promotes intelligent solid waste management for efficiently managing environmental concerns.

Deep learning-based models for environmental management: Recognizing construction, renovation, and demolition waste in-the-wild.

The construction industry generates a substantial volume of solid waste, often destinated for landfills, causing significant environmental pollution. Waste recycling is decisive in managing waste yet challenging due to labor-intensive sorting processes and the diverse forms of waste. Deep learning (DL) models have made remarkable strides in automating domestic waste recognition and sorting. However, the application of DL models to recognize the waste derived from construction, renovation, and demolition (CRD) activities remains limited due to the context-specific studies conducted in previous research. This paper aims to realistically capture the complexity of waste streams in the CRD context. The study encompasses collecting and annotating CRD waste images in real-world, uncontrolled environments. It then evaluates the performance of state-of-the-art DL models for automatically recognizing CRD waste in-the-wild. Several pre-trained networks are utilized to perform effectual feature extraction and transfer learning during DL model training. The results demonstrated that DL models, whether integrated with larger or lightweight backbone networks can recognize the composition of CRD waste streams in-the-wild which is useful for automated waste sorting. The outcome of the study emphasized the applicability of DL models in recognizing and sorting solid waste across various industrial domains, thereby contributing to resource recovery and encouraging environmental management efforts.

AI explainability framework for environmental management research.

Deep learning networks powered by AI are essential predictive tools relying on image data availability and processing hardware advancements. However, little attention has been paid to explainable AI (XAI) in application fields, including environmental management. This study develops an explainability framework with a triadic structure to focus on input, AI model and output. The framework provides three main contributions. (1) A context-based augmentation of input data to maximize generalizability and minimize overfitting. (2) A direct monitoring of AI model layers and parameters to use leaner (lighter) networks suitable for edge device deployment, (3) An output explanation procedure focusing on interpretability and robustness of predictive decisions by AI networks. These contributions significantly advance state of the art in XAI for environmental management research, offering implications for improved understanding and utilization of AI networks in this field.

Application of recycled crushed glass in road pavements and pipeline bedding: An integrated environmental evaluation using LCA.

The study aims to conduct a comprehensive life cycle assessment (LCA) of mixed glass waste (MGW) recycling processes to quantify the environmental impacts of crushed glass as a partial substitute for virgin aggregate. Upstream washing, crushing, and sorting conducted at material recycling facilities (MRF) are the prime activities to assess whether reprocessed MGW in pavement construction is an alternate feasible solution. None of the previous studies explicitly account for the relative uncertainties and optimization of waste glass upstream processes from an environmental perspective. The study calculates environmental impacts using the LCA tool SimaPro considering design factors attributed to transportation, electricity consumption, use of chemicals, and water for reprocessing glass waste. Relative uncertainties of design variables and the national transition policy (2021-2030) from non-renewable to renewable energy sources have been validated by performing detailed Monte Carlo simulations. The correlation coefficients (r = 0.64, 0.58, and 0.49) of successive variables explain how the higher environmental gains of the glass recycling process are outweighed by diesel, energy consumption, and transportation distances. Compared to natural quarry sand, the recycled glass aggregate produced through crushing and recycling of its by-products reduces CO2eq emissions by 16.2 % and 46.7 %, respectively. The need for a washing line at the plant, in addition to crushing, results in a higher environmental impact over natural sand by 90.1 % and emphasizes the benefits of collecting waste glass through a separate bin, hence avoiding contamination. The result indicates that the benefit of lowering emissions varies significantly when considering waste glass landfilling. Moreover, this study evaluates the potential impacts on asphalt and reinforced concrete pavements (RCP) with 5 %, 10 %, 15 %, and 20 % replacement of natural sand with recycled glass aggregate. The LCA emphasizes the limitations of energy-intensive waste glass reprocessing. The obtained results and uncertainty analysis based on primary MRF data and recycled product applications provide meaningful suggestions for a more fit-for-purpose waste management and natural resource conservation.

Automatic generation of structural geometric digital twins from point clouds.

A geometric digital twin (gDT) model capable of leveraging acquired 3D geometric data plays a vital role in digitizing the process of structural health monitoring. This study presents a framework for generating and updating digital twins of existing buildings by inferring semantic information from as-is point clouds (gDT's data) acquired regularly from laser scanners (gDT's connection). The information is stored in updatable Building Information Models (BIMs) as gDT's virtual model, and dimensional outputs are extracted for structural health monitoring (gDT's service) of different structural members and shapes (gDT's physical part). First, geometric information, including position and section shape, is obtained from the acquired point cloud using domain-specific contextual knowledge and supervised classification. Then, structural members' function and section family type is inferred from geometric information. Finally, a BIM is automatically generated or updated as the virtual model of an existing facility and incorporated within the gDT for structural health monitoring. Experiments on real-world construction data are performed to illustrate the efficiency and precision of the proposed model for creating as-is gDT of building structural members.

Investigating the Barriers to Applying the Internet-of-Things-Based Technologies to Construction Site Safety Management.

The utilization of Internet-of-Things (IoT)-based technologies in the construction industry has recently grabbed the attention of numerous researchers and practitioners. Despite the improvements made to automate this industry using IoT-based technologies, there are several barriers to the further utilization of these leading-edge technologies. A review of the literature revealed that it lacks research focusing on the obstacles to the application of these technologies in Construction Site Safety Management (CSSM). Accordingly, the aim of this research was to identify and analyze the barriers impeding the use of such technologies in the CSSM context. To this end, initially, the extant literature was reviewed extensively and nine experts were interviewed, which led to the identification of 18 barriers. Then, the fuzzy Delphi method (FDM) was used to calculate the importance weights of the identified barriers and prioritize them through the lenses of competent experts in Hong Kong. Following this, the findings were validated using semi-structured interviews. The findings showed that the barriers related to "productivity reduction due to wearable sensors", "the need for technical training", and "the need for continuous monitoring" were the most significant, while "limitations on hardware and software and lack of standardization in efforts," "the need for proper light for smooth functionality", and "safety hazards" were the least important barriers. The obtained findings not only give new insight to academics, but also provide practical guidelines for the stakeholders at the forefront by enabling them to focus on the key barriers to the implementation of IoT-based technologies in CSSM.

Quantitative Analysis of Safety Risks and Relationship with Delayed Project Completion Times.

Dynamic work environments in construction and civil infrastructure sectors remain susceptible to safety risks. Although previous research has resulted in improvements, there is currently a gap in measuring temporal impacts of safety risks quantitatively. Precise modeling of potential delays caused by safety incidents is vital for efficient management of risks and making informed decisions on project contingency. Toward this aim, the current research adopts a nondeterministic modeling method to simulate and quantify safety incidents and find correlations with project delays. Using a deductive approach, three research questions were formulated, and investigations conducted on Australian data collected from 2016 onwards. Quantitative impacts of safety risks on project completion times were numerically measured. Furthermore, safety risks were ranked based on their significance of temporal impacts on project performance. This paper contributes to the theory of safety management by developing a nondeterministic method to model impacts of safety risks at both industry and project levels. Practical contributions and outcomes can facilitate using machine learning methods to plan proportionate time buffers to address safety risks.

The Cost Burden of Safety Risk Incidents on Construction: A Probabilistic Quantification Method.

The construction sector is vulnerable to safety risk incidents due to its dynamic nature. Although numerous research efforts and technological advancements have focused on addressing workplace injuries, most of the studies perform empirical and deterministic postimpact evaluations on construction project performance. The effective modeling of the safety risk impacts on project performance provides decisionmakers with a valuable tool toward incidents prevention and proper safety risk management. Therefore, this study collected Australian incident records from the construction industry from 2016 onwards and conducted discrete event simulation to quantitatively measure the impact of safety risk incidents on project cost performance. Moreover, this study investigated the correlation between safety risk incidents and the age of injured workers. The findings show a strong correlation between the middle-aged workforce and the severity of incidents on project cost overruns. The ex-ante, nondeterministic analysis of safety risk impacts on project performance provides insightful results that will advance safety management theory in the direction of achieving zero harm workplace environments.