Search

• A digital twin-driven (DTD) framework and model are developed for predicting degraded lifting capacity (LC) of aging tower cranes. • A DTD model predicted the degraded LC of a scaled-down prototype, achieving a mean-square error (MSE) of 0.2253 and a coefficient of determination (R2) of 0.9973. • A DTD model is validated using k-5 cross-validation with a prediction accuracy of 0.97 (R2). • Degraded load charts assist operators in placing safe loads and preventing unexpected failures. Aging tower cranes face an elevated risk of failure, primarily due to structural fatigue and deterioration. Surprisingly, the degradation of aging-induced lifting capacity (LC) remains an unexplored domain. In response to this research gap, this paper introduces a digital twin-driven (DTD) framework and model to predict the degraded LC of aging tower cranes. This framework combines theoretical and numerical analysis of fatigue and degradation behavior in tower cranes with real-time vibration data obtained during cyclic load scenarios on the actual cranes. Machine learning (ML) techniques are employed to develop a model that accurately predicts the degraded LC caused by aging. A scaled-down tower crane prototype is adopted as a demonstrative case to illustrate the feasibility and effectiveness of the DTD framework. The DTD model predicts the degraded LC of the prototype with high accuracy, achieving a mean-square error (MSE) of 0.2253 and a coefficient of determination (R2) of 0.9973. The predicted degraded load charts of the tested tower crane for each decade of usage from 0 to 70 years are also presented to assist crane operators in applying safe loads, preventing unexpected failures and damages, and enhancing workplace monitoring and safety. This study helps monitor the safety conditions of tower cranes that are aging and susceptible to structural fatigue and deterioration, facilitates the prediction of the deterioration of complex machines and systems in the construction industry with real-time data, and highlights the potential of DTD approaches in improving efficiency, safety, and decision-making. [ABSTRACT FROM AUTHOR]

• This paper proposes an online fault diagnosis method for rotating machinery. • A novel multi-level information fusion method is proposed for multi-sensor signals. • Sample and class incremental learning capability are developed for rapid updates. • Two datasets of key components are employed to validate the superiority. Intelligent Fault Diagnosis (IFD) models are all trained in one-time learning way, lacking the incremental learning capability for continually incoming samples and newly occurring faults. Although Online Fault Diagnosis (OFD) models with incremental learning capability have begun to attract sustaining attention and extensive research, they are just suitable for a single-sensor signal, limiting their diagnostic performance on machinery operating under complex environment. To solve these problems, this paper proposes a multi-sensor fused incremental broad learning with D-S theory for online fault diagnosis of rotating machinery. Firstly, a feature fusion method based on mutual attention mechanism is designed to sufficiently explore the similarity relationship of the multi-sensor data after fast Fourier transform (FFT). After that, the fused feature matrices are inputted to broad learning system for training and the outputs are fused based on D-S evidence theory, achieving multi-level information fusion to ensure adequate fusion of the multi-sensor signals. Additionally, the sample and class incremental learning are developed to rapidly update subsequent models without retraining. Finally, two experiment datasets concerning the key components of rotating machinery are employed to verify the effectiveness and superiority of the proposed method. Benefiting from the advantages of sample incremental learning, the model can be updated in approximately 6 s and 0.15 s, respectively, which is less than the initial model. Meanwhile, the testing accuracy of subsequent models is still maintained at 100% as more and more samples are employed for model updates. Besides, it can be updated in 13 s and 1 s respectively to accommodate new faults, while still has excellent accuracy after fusion. Consequently, the proposed method is an effective online fault diagnosis method because it represents significant advantages in reducing time consumption and improving the accuracy and reliability of diagnostic results. [ABSTRACT FROM AUTHOR]

Image data collected after natural disasters play an important role in the forensics of structure failures. However, curating and managing large amounts of post-disaster imagery data is challenging. In most cases, data users still have to spend much effort to find and sort images from the massive amounts of images archived for past decades in order to study specific types of disasters. This paper proposes a new machine learning based approach for automating the labeling and classification of large volumes of post-natural disaster image data to address this issue. More specifically, the proposed method couples pre-trained computer vision models and a natural language processing model with an ontology tailed to natural disasters to facilitate the search and query of specific types of image data. The resulting process returns each image with five primary labels and similarity scores, representing its content based on the developed word-embedding model. Validation and accuracy assessment of the proposed methodology was conducted with ground-level residential building panoramic images from Hurricane Harvey. The computed primary labels showed a minimum average difference of 13.32% when compared to manually assigned labels. This versatile and adaptable solution offers a practical and valuable solution for automating image labeling and classification tasks, with the potential to be applied to various image classifications and used in different fields and industries. The flexibility of the method means that it can be updated and improved to meet the evolving needs of various domains, making it a valuable asset for future research and development. [ABSTRACT FROM AUTHOR]

In this paper a method to learn a similarity metric from expert assessments via questionnaires is presented. The approach employed provides a solution to the modelling of a DPSIR sustainability approach where budgetary resources are limited and thus there is a need to select the most informative variables from the identified possibilities. This paper also shows the proposed approach already implemented by the local council of Las Palmas of Gran Canaria as part of the work to create a sustainability system to better control the impact of human pressure in the local region. The metric is learned using a weakly supervised approach and the expert assessments are modelled through variable triplets. The employment of machine learning approaches in the creation of sustainability models is fairly recent and rare but presents a great opportunity to contribute to one of the main challenges that human societies have to face nowadays. [ABSTRACT FROM AUTHOR]

Product customization is a complex task that relies heavily on consumer preferences. Eliciting these preferences can be challenging for firms looking to develop novel products and require significant investments of both time and effort. Prediction models can serve to replace traditional methods of understanding consumer preferences such as elicitation, focus groups or the designer's intuition, while speeding up the production cycle and saving cost. Current prediction models generally focus on one specific product type and require large amounts of data or historical product data. The idea of predicting consumer preferences for products based on the product type and its features using a clustering approach has not been explored in literature. This paper presents a proof-of-concept consumer preference prediction and decision support model based on a data-driven approach to design for product customization. First, consumer preference information is crowdsourced using surveys with 307 individual responses that are collected for a data set of thirty-seven training products and three validation products. Second, clustering techniques are assessed for user-generated clustering variables along with features that are extracted with artificial intelligence (ChatGPT). A threshold metric is proposed to evaluate the accuracy of different clustering algorithms. Third, a recommendation model is developed for customization decisions, and it is validated with three different products with an average accuracy of 70%. Areas for future work to improve the accuracy and expand the scope of the model are discussed including the use of a larger training data set, different machine learning approaches, and the improved use of ChatGPT. [ABSTRACT FROM AUTHOR]

Rehabilitation of contaminated soils in urban areas is in high demand because of the appreciation of land value associated with the increased urbanization. Moreover, there are financial incentives to minimize soil characterization uncertainties. Minimizing uncertainty is achieved by providing models that are better representation of the true site characteristics. In this paper, we propose two new probabilistic formulations compatible with Gaussian Process Regression (GPR) and enabling (1) to model the experimental conditions where contaminant concentration is quantified from aggregated soil samples and (2) to model the effect of physical site discontinuities. The performance of approaches proposed in this paper are compared using a Leave One Out Cross-Validation procedure (LOO-CV). Results indicate that the two new probabilistic formulations proposed outperform the standard Gaussian Process Regression. [ABSTRACT FROM AUTHOR]

Track geometry is highly important for ensuring railroad safety. Predicting track geometry degradation can support preventive maintenance by identifying and prioritizing track segments that are more prone to potential track geometry defects. This paper develops a novel machine learning approach that can simultaneously geospatially align track geometry data from multiple inspections accounting for positional errors, and also predict foot-by-foot track geometry change over time. The proposed position correction method considers multiple geometry parameters to provide highly accurate positional information for geometry measurements, which is a prerequisite for foot-by-foot track geometry prediction. A hybrid CNN (convolutional neural network)-LSTM (long short-term memory neural network) machine learning model is developed to account for spatial–temporal dependence with respect to foot-by-foot track geometry change. Specifically, CNN is used to incorporate spatial dependence of track geometry on adjacent segments. LSTM is applied to learn dynamic changes in track geometry data over time to consider the temporal dependence. The hybrid CNN-LSTM model is validated using track inspection data provided by one major freight railroad in the United States. The results show that our positional error correction method can reduce the relative positional error to less than 1 foot with a 99% confidence interval. Our proposed CNN-LSTM model outperforms four other models, including a naïve model (using the last observation), multi-layer perceptron (MLP), plain CNN, and plain LSTM, for both short-term and long-term prediction periods. The proposed hybrid machine learning methodology can be adapted to various other freight and passenger rail lines. The predictive track geometry change information can be used by the industry to plan and prioritize resources for preventative maintenance, yielding benefits in safety and operational efficiency. [ABSTRACT FROM AUTHOR]

• A novel method is developed for real-time monitoring of crack dynamic expansion. • An efficient benchmarking method is established to determine crack expansion. • The YOLOv7 algorithm is improved to enhance the accuracy of crack detection. • A new attention mechanism is introduced in the DeepLabv3+ algorithm to improve crack segmentation. Crack monitoring has been a hot research topic in structural health monitoring. However, the current research on deep learning-based crack image focuses more on cracks at a certain moment and ignores the full-time crack expansion details, which are crucial for more reasonable evaluation and safety quantification of concrete structures. This paper proposes a new method based on the combination of improved You Only Look Once v7 (YOLOv7) algorithm, crack expansion benchmark method, improved DeepLabv3+ algorithm, and image processing technology to monitor the whole process of crack development, including real-time crack recognition and real-time monitoring of crack dynamic expansion. The precision of the improved detection algorithm can be improved by a maximum of 5.34%, and the mean intersection over union (mIoU) of the improved segmentation algorithm can be improved by 0.15%, resulting in better segmentation results. The experimental results show that this method can efficiently and accurately achieve real-time tracking of crack dynamic expansion, especially for monitoring of tiny cracks. [ABSTRACT FROM AUTHOR]

[Display omitted] • A control framework for improving vehicle ride comfort in the post-braking phase is proposed. • Residual reinforcement learning approach is applied. • Real vehicle experiments under different scenarios are performed. • Results prove the superiority of the proposed controller. Owing to increasing urban congestion, ensuring vehicle ride comfort during the post-braking phase has become an essential requirement. However, achieving vehicle ride comfort using current conventional methods is challenging due to the vehicles' complex dynamics. This paper proposes a novel controller with residual reinforcement learning, combining the advantages of the model-free reinforcement learning algorithm, heuristic optimization algorithm, and prior expert knowledge to significantly improve training efficiency. The nonlinear and transient characteristics of the tire and vehicle are modeled to improve the control accuracy. On-vehicle experiments are performed using a skateboard chassis. The experimental results show that the proposed strategy achieves significant improvement in vehicle ride comfort under various braking scenarios. We believe that this technology has the potential to alleviate vehicle discomfort issues in daily life. [ABSTRACT FROM AUTHOR]

[Display omitted] • A controller to prevent secondary crashes after an initial rear-end collision is proposed. • Rule-based switching control is embedded into reinforcement learning. • Pre-collision control and post-collision control are combined. • Results prove the superiority of the proposed controller. Rear-end collisions result in a large number of casualties and property losses, and the serious injury risk in multiple impact accidents is much higher than that in single impact accidents. In this paper, we propose a novel controller to facilitate the prevention of secondary crashes after an initial rear-end collision, which expands the operational horizon of conventional vehicle active safety systems from preventive measures to post-event mitigation measures. Considering the complexity of the problem with multi-object synthesis optimization and vehicle nonlinear dynamics, this study combines the pre-collision control and post-collision control to reduce the initial crash loss and the subsequent control difficulty. The rule-based switching control and drift manipulation are embedded into the reinforcement learning algorithm to improve the training efficiency and control performance. The bench test results validate the superiority of the proposed controller over other strategies and algorithms in different rear-end collision scenarios. [ABSTRACT FROM AUTHOR]

Although substantial progress in safety management performance has been made in the construction industry, continuing fatalities and injuries at workplaces hinder sustainable development of this labor-intensive industry. Many machine learning approaches using different types of data such as text, image, video, and audio were adopted for safety risk analysis at construction sites. Our paper aimed to implement a hybrid review of construction safety research based upon machine learning. This hybrid review focused on various attributes from three perspectives: Nature of dataset, machine learning approach, and research topic. After the review of individual attributes, intra-relationships between attributes in each perspective and inter-relationships between attributes across the three perspectives were determined. According to risk recognition, risk prediction, and risk control, feasible research paths were developed from both intra-relationships and inter-relationships between multiple attributes for reference in future studies. Finally, gaps and opportunities were discussed in detail for research agendas on this subject. This hybrid review contributes to outlining the framework of construction safety management based upon machine learning. It is able to provide new entrants with a systematic idea of promising research trends for the future. Research findings are helpful for academia and industry to fill in the gaps between study and practice in the area of construction safety, in order to assist in sustainable development of the construction industry by use of machine learning. [ABSTRACT FROM AUTHOR]

• Excavation simulation experiment was conducted to assess the mental fatigue status of the subjects. • A non-invasive smart cushion was developed to monitor heartbeat and respiration signals. • The time domain and frequency domain features in HRV and RR were extracted to develop the mental fatigue classification model. • Random forest model achieved high classification performance. Mental fatigue has a considerable influence on the completion of work as well as the safety and health of construction equipment operators who have been exposed to human–machine interaction for an extended period of time. Existing techniques of monitoring and measuring mental fatigue among construction equipment operators are either invasive or rely on subjective judgements, limiting their applicability on construction sites. Therefore, this paper developed a non-invasive smart cushion to evaluate the mental fatigue of construction equipment operators through real-time continuous monitoring of the heartbeat and respiration signals. A laboratory experiment was designed, and twelve participants were recruited to conduct the simulated excavator operation for data acquisition. First, wearable sensors were used as a reference to verify the monitoring accuracy of heartbeat and respiration signals. Excellent correlations were realized between the results of the two monitoring methods. Second, the relevant time domain and frequency domain features were extracted to construct a random forest classification model to discover the relationship between vital signs and subjectively reported mental fatigue status. The results demonstrated that the random forest model had a classification accuracy of 92%, better than other commonly used machine learning algorithms. A combination of heartbeat and respiration features is more effective than a single feature in identifying and classifying mental fatigue among construction equipment operators. The study's findings indicate that it is feasible to apply the smart cushion to identify and classify the mental fatigue of construction equipment operators, therefore contributing to the improvement of their health and safety. [ABSTRACT FROM AUTHOR]

State-of-the-art workflows within Architecture, Engineering, and Construction (AEC) are still caught in sequential planning processes. Digital design tools in this domain often lack proper communication between different stages of design and relevant domain knowledge. Furthermore, decisions made in the early stages of design, where sketching is used to initiate, develop, and communicate ideas, heavily impact later stages, resulting in the need for rapid feedback to the architectural designer so they can proceed with adequate knowledge about design implications. Accordingly, this paper presents research on a novel integrative design framework based on a recently developed 4D sketching interface, targeted for architectural design as a form-finding tool coupled with three modules: (1) a Geometric Modelling module, which utilises Points2Surf as a machine learning model for automatic surface mesh reconstruction from the point clouds produced by sketches, (2) a Material Modelling module, which predicts the mechanical properties of biocomposites based on multiscale micromechanics homogenisation techniques, and (3) a Structural Analysis module, which assesses the mechanical performance of the meshed structure on the basis of the predicted material properties using finite element simulations. The proposed framework is a step towards using material-informed design already in the early stages of design. [ABSTRACT FROM AUTHOR]

Machine learning (ML) and deep learning (DL) are both branches of AI. As a form of AI, ML automatically adapts to changing datasets with minimal human interference. Deep learning is a subset of ML that uses artificial neural networks to imitate the learning process of the human brain. The 'black box' nature of ML and DL makes their inner workings difficult to understand and interpret. Deploying explainable artificial intelligence (XAI) can help explain why and how the output of ML and DL models are generated. As a result, understanding a model's functioning, behavior, and outputs can be garnered, reducing bias and error and improving confidence in decision-making. Despite providing an improved understanding of model outputs, XAI has received limited attention in construction. This paper presents a narrative review of XAI and a taxonomy of precepts and models to raise awareness about its potential opportunities for use in construction. It is envisaged that the opportunities suggested can stimulate new lines of inquiry to help alleviate the prevailing skepticism and hesitancy toward AI adoption and integration in construction. [ABSTRACT FROM AUTHOR]

The ability to process large amounts of data and to extract useful insights from data has revolutionised society. This phenomenon—dubbed as Big Data—has applications for a wide assortment of industries, including the construction industry. The construction industry already deals with large volumes of heterogeneous data; which is expected to increase exponentially as technologies such as sensor networks and the Internet of Things are commoditised. In this paper, we present a detailed survey of the literature, investigating the application of Big Data techniques in the construction industry. We reviewed related works published in the databases of American Association of Civil Engineers (ASCE), Institute of Electrical and Electronics Engineers (IEEE), Association of Computing Machinery (ACM), and Elsevier Science Direct Digital Library. While the application of data analytics in the construction industry is not new, the adoption of Big Data technologies in this industry remains at a nascent stage and lags the broad uptake of these technologies in other fields. To the best of our knowledge, there is currently no comprehensive survey of Big Data techniques in the context of the construction industry. This paper fills the void and presents a wide-ranging interdisciplinary review of literature of fields such as statistics, data mining and warehousing, machine learning, and Big Data Analytics in the context of the construction industry. We discuss the current state of adoption of Big Data in the construction industry and discuss the future potential of such technologies across the multiple domain-specific sub-areas of the construction industry. We also propose open issues and directions for future work along with potential pitfalls associated with Big Data adoption in the industry. [ABSTRACT FROM AUTHOR]

The prefabricated concrete buildings (PCBs）are the booster in the process of construction industrialization and intelligent upgrading. However, its high cost has become one of the restricting factors of further application and promotion of prefabricated concrete buildings. Moreover, the existing investment estimation methods of prefabricated concrete buildings have limited predicting accuracy as well as the ability of adapting dynamic factors. Therefore, to achieve more reliable and reasonable investment estimation of prefabricated concrete buildings, this paper has proposed an investment estimation model of prefabricated concrete buildings based on XGBoost machine learning algorithm. In the proposed model, the construction project cost-significance theory (CS) and analytic hierarchy process (AHP) were used to extract the construction characteristic indices of prefabricated concrete buildings investment estimation. Then the XGBoost machine learning algorithm was implemented to build an investment estimation model of prefabricated concrete buildings that was able to quantify the uncertainty of the confidence and prediction, and to enhance the interpretability of the model. The research conducted in this paper showed that when compared with traditional machine learning methods such as Support vector machine (SVM), Back Propagation Neural Network (BPNN) and Random Forest (RF), XGBoost had better generalization and interpretable ability. The discussion provided in this paper further demonstrated the reliability and feasibility of the proposed model, and provided reliable basis for the investment decision-making of prefabricated concrete building projects. [ABSTRACT FROM AUTHOR]

Ultra-precision machining (UPM) is an advanced manufacturing technology that experiences increasing demand. Therefore, it is necessary to minimize the environmental impacts from its enormous consumptions of resources. Achieving sustainable UPM is still a challenge is it involves complicated influencing relationships among relevant factors like energy consumption, and human health, which could affect sustainable performance. And some influencing relationships between two parameters have not been fully studied yet, which are named as undiscussed two-parameter relationships. Therefore, this paper proposed a new topic discovery model based on social network analysis (SNA) and machine learning approach to discover the undiscussed two-parameter relationships with high potential value in the sustainable UPM research field. By using the link prediction metrics obtained by SNA and principal components analysis in this study, the interactive relationships among the parameters of sustainable ultra-precision machining are determined to discover the potential values of undiscussed two-parameter topics. Then, the k-means algorithm is applied to classify the topics based on the similarity of the metrics results to present the potential value distribution of the undiscussed topics in sustainable UPM. From the metrics results, the topic of the relationship between environmental damage and resource waste was found to be the most valuable potential two-parameter topic in the area of sustainable UPM. This paper also contributes to showing the potential value distribution of undiscussed two-parameter relationships and predicting the sustainable development trend in the UPM sectors. [ABSTRACT FROM AUTHOR]

Abstract This paper presents a machine learning-based approach to power transformer fault diagnosis based on dissolved gas analysis (DGA), a bat algorithm (BA), optimizing the probabilistic neural network (PNN). PNN is a radial basis function feedforward neural network based on Bayesian decision theory, which has a strong fault tolerance and significant advantages in pattern classification. However, one challenge still remains: the performance of PNN is greatly affected by its hidden layer element smooth factor which impacts the classification performance. The proposed approach addresses this challenge by deploying the BA algorithm, a kind of bio-inspired algorithm to optimize PNN. Using the real data collected from a transformer system, we conducted the experiments for validating the performance of the developed method. The experimental results demonstrated that BA is an effective algorithm for optimizing PNN smooth factor and BA-PNN can improve the fault diagnosis performance; in turn, and the machine learning-based model (BA-PNN) can significantly enhance the accuracies of power transformer fault diagnosis. [ABSTRACT FROM AUTHOR]

Abstract 3D scenes reconstructed from point clouds, acquired by either laser scanning or photogrammetry, are subject to data voids generated by occluding objects. Modeling from incomplete data is usually a manual process in which human interpretation plays an essential role. This paper presents a machine learning algorithm based on neural networks capable of recovering point cloud occlusions for surfaces that can be approximated with injective functions. Starting from the point clouds acquired around the occlusion, a set of single-layer feedforward networks with a variable number of neurons is trained and validated with a subset of the original cloud, which is preliminarily decimated using local curvature to reduce CPU cost. The averaged result of the best neural networks is evaluated on a spatial domain that contains the 2D projection of the void, obtaining a complete 3D point cloud for the occluded volume. Criteria for choosing the number of neurons and the activation function for hidden and output layers are illustrated and discussed. Results are presented for both simulated and real occlusions, describing the pros and cons of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract Successful adoption and management of sustainable urban systems hinges on the community embracing these systems. Capturing citizens' ideas, views, and assessments of the built environment will be essential to this goal. In collaborative city planning, these are qualified and valued forms of partial knowledge that should be collectively used to shape the decision making process of urban planning. Among other tools, social media and online social network analytics can provide means to capture elements of such a distributed knowledge. While a structured definition of sustainability (normally dictated in a top-down fashion) may not sufficiently respond well to the pluralist nature of such knowledge acquisition; dealing with the unstructured community inputs, assessments and contributions on social media can be confusing. We can detect fully relevant topics/ideas in community discussions; but they typically suffer from lack of coherence. In this paper, we advocate the use of a semi-structured approach for capturing, analyzing, and interpreting citizens' inputs. Public officials and professionals can develop the main elements (topical aspects) of sustainability, which can act as the skeleton of a taxonomy. It is however, the community inputs/ideas (in our case collected via social media and parsed), that can shape-up that skeleton and augment those topical aspects with adding the required semantic depth. In more specific terms, we collected tweets for four urban infrastructure mega-projects in North America. Then we used a game-with-a-purpose to crowdsource the identification of topics for a training set of tweets. This was then used to train machine learning algorithms to cluster the rest of collected tweets. We studied the semantic (finding the topics) of tweets as well as their sentiment (in terms of being opposing or supportive of a project). Our classification tested different decision trees with different topic hierarchies. We considered/extracted eight different linguistic features in studying contents of a tweet. Finally, we examined the accuracy of three algorithms in classifying tweets according to the sequence in the tree, and based on the extracted features. These are: K-nearest neighbors, Naïve Bayes classifiers and Support Vector Machines (SVM). Respective to our data set, SVM outperformed other algorithms. Semantic analysis was insensitive to the depth/number of linguistic features considered. In contrast, sentiment analysis was enhanced when part of speech (PoS) was tracked. Interestingly, our work shows that considering the topic (semantic) of a tweet helped enhance the accuracy of sentiment analysis: including topical class as a feature in conducting sentiment analysis results in higher accuracies. This could be used as means to detect the evolution of community opinion: that topic-based social networks are evolving within the communities tweeting about urban projects. It could also be used to identify the topics of top priority to the community or the ones that have the widest spread of views. In our case, these were mainly the impacts of the design and engineering features on social issues. [ABSTRACT FROM AUTHOR]

Highlights • Overexertion can cause bodily injury in lift/lower/carry, and push/pull activities. • Machine learning helps recognize activities from body-mounted smartphone data. • In this research, activity duration and frequency are linked to overexertion risk. • Best results are achieved when smartphone is placed on the subject's upper-arm. • The designed algorithm is validated using leave-one-subject-out cross-validation. Abstract Workers in various industries are often subject to challenging physical motions that may lead to work-related musculoskeletal disorders (WMSDs). To prevent WMSDs, health and safety organizations have established rules and guidelines that regulate duration and frequency of labor-intensive activities. In this paper, a methodology is introduced to unobtrusively evaluate the ergonomic risk levels caused by overexertion. This is achieved by collecting time-stamped motion data from body-mounted smartphones (i.e., accelerometer, linear accelerometer, and gyroscope signals), automatically detecting workers' activities through a classification framework, and estimating activity duration and frequency information. This study also investigates various data acquisition and processing settings (e.g., smartphone's position, calibration, window size, and feature types) through a leave-one-subject-out cross-validation framework. Results indicate that signals collected from arm-mounted smartphone device, when calibrated, can yield accuracy up to 90.2% in the considered 3-class classification task. Further post-processing the output of activity classification yields very accurate estimation of the corresponding ergonomic risk levels. This work contributes to the body of knowledge by expanding the current state in workplace health assessment by designing and testing ubiquitous wearable technology to improve the timeliness and quality of ergonomic-related data collection and analysis. [ABSTRACT FROM AUTHOR]

Graphical abstract Abstract Increasing sustainability requirements make evaluating different design options for identifying energy-efficient design ever more important. These requirements demand simulation models that are not only accurate but also fast. Machine Learning (ML) enables effective mimicry of Building Performance Simulation (BPS) while generating results much faster than BPS. Component-Based Machine Learning (CBML) enhances the capabilities of the monolithic ML model. Extending monolithic ML approach, the paper presents deep-learning architectures, component development methods and evaluates their suitability for space exploration in building design. Results indicate that deep learning increases the performance of models over simple artificial neural network models. Methods such as transfer learning and Multi-Task Learning make the component development process more efficient. Testing the deep-learning model on 201 new design cases indicates that its cooling energy prediction (R2: 0.983) is similar to BPS, while errors for heating energy predictions (R2: 0.848) are higher than BPS. Higher heating energy prediction error can be resolved by collecting heating data using better design space sampling methods that cover the heating demand distribution effectively. Given that the accuracy of the deep-learning model for heating predictions can be increased, the major advantage of deep-learning models over BPS is their high computation speed. BPS required 1145 s to simulate 201 design cases. Using the deep-learning model, similar results can be obtained in 0.9 s. High computation speed makes deep-learning models suitable for design space exploration. [ABSTRACT FROM AUTHOR]

Highlights • Automatic classification of WOs enhances and structures occupant-generated data. • Category prediction accuracy 57–86% for all problem categories and 80%+ for the top 3. • Prediction accuracy up to 90% was achieved for the most frequent subcategories. • Class prediction accuracies above 95% were consistently achieved for multiple classes. • When the problem category is known, subcategory prediction accuracy increases above 90% Abstract Occupant-generated work orders are recognized as a good potential data to support Facility Management (FM) activities, however they are unstructured and rarely contain the specific information engineers require to resolve the reported issues. Instead, this often requires multiple trips are often needed to identify the required trade, identify the problem and required parts/tools, and resolve. A key challenge is data quality: free-form (unstructured) text is collected that frequently lacks necessary detail for problem diagnosis. Machine Learning provides new opportunities within the FM domain to improve the quality of information collected through online work order reporting systems by automatically classifying WOs and prompting building occupants with appropriate FM team-developed questions in real time to gather the required specific information in structured form. This paper presents the development, comparison, and application of two sets of supervised machine learning models to perform this classification for WOs generated from occupant complaints. A set of ∼150,000 historical WOs was used for model development and textual classification using with various term and itemset frequency approaches was tested. Classifier prediction accuracies ranged from 46.6% to 81.3% for classification by detailed subcategory; this increased to between 68% (simple term frequency) to 90% (random forest) when the dataset only included the ten most common (accounting for 70% of all WOs) subcategories. Hierarchical classification decreased performance. An FM-BIM integration approach is finally presented using the resultant classifiers to provide facilities management teams with spatio-temporal visualization of the work order categories across a series of buildings to help prioritize and streamline operations and maintenance task assignments. [ABSTRACT FROM AUTHOR]

This paper presents distress classification of class-imbalanced inspection data via correlation-maximizing weighted extreme learning machine (CMWELM). For distress classification, it is necessary to extract semantic features that can effectively distinguish multiple kinds of distress from a small amount of class-imbalanced data. In recent machine learning techniques such as general deep learning methods, since effective feature transformation from visual features to semantic features can be realized by using multiple hidden layers, a large amount of training data are required. However, since the amount of training data of civil structures becomes small, it becomes difficult to perform successful transformation by using these multiple hidden layers. On the other hand, CMWELM consists of two hidden layers. The first hidden layer performs feature transformation, which can directly extract the semantic features from visual features, and the second hidden layer performs classification with solving the class-imbalanced problem. Specifically, in the first hidden layer, the feature transformation is realized by using projections obtained by maximizing the canonical correlation between visual and text features as weight parameters of the hidden layer without designing multiple hidden layers. Furthermore, the second hidden layer enables successful training of our classifier by using weighting factors concerning the class-imbalanced problem. Consequently, CMWELM realizes accurate distress classification from a small amount of class-imbalanced data. [ABSTRACT FROM AUTHOR]

Soil classification is a means of grouping soils into categories according to a shared set of properties or characteristics that will exhibit similar engineering behaviour under loading. Correctly classifying site conditions is an important, costly, and time-consuming process which needs to be carried out at every building site prior to the commencement of construction or the design of foundation systems. This paper presents a means of automating classification for fine-grained soils, using a feed-forward ANN (Artificial Neural Networks) and CPT (Cone Penetration Test) measurements. Thus representing a significant saving of both time and money streamlining the construction process. 216 pairs of laboratory results and CPT tests were gathered from five locations across Northern Croatia and were used to train, test, and validate the ANN models. The resultant Neural Networks were saved and were subjected to a further external verification using CPT data from the Veliki vrh landslide. A test site, which the model had not previously been exposed to. The neural network approach proved extremely adept at predicting both ESCS (European Soil Classification System) and USCS (Unified Soil Classification System) soil classifications, correctly classifying almost 90% of soils. While the soils that were incorrectly classified were only partially misclassified. The model was compared to a previously published model, which was compiled using accepted industry standard soil parameter correlations and was shown to be a substantial improvement, in terms of correlation coefficient, absolute average error, and the accuracy of soil classification according to both USCS and ESCS guidelines. The study confirms the functional link between CPT results, the percentage of fine particles FC , the liquid limit w L and the plasticity index I P. As the training database grows in size, the approach should make soil classification cheaper, faster and less labour intensive. [ABSTRACT FROM AUTHOR]

Numerical weather forecasts, such as meteorological forecasts of precipitation, are inherently uncertain. These uncertainties depend on model physics as well as initial and boundary conditions. Since precipitation forecasts form the input into hydrological models, the uncertainties of the precipitation forecasts result in uncertainties of flood forecasts. In order to consider these uncertainties, ensemble prediction systems are applied. These systems consist of several members simulated by different models or using a single model under varying initial and boundary conditions. However, a too wide uncertainty range obtained as a result of taking into account members with poor prediction skills may lead to underestimation or exaggeration of the risk of hazardous events. Therefore, the uncertainty range of model-based flood forecasts derived from the meteorological ensembles has to be restricted. In this paper, a methodology towards improving flood forecasts by weighting ensemble members according to their skills is presented. The skill of each ensemble member is evaluated by comparing the results of forecasts corresponding to this member with observed values in the past. Since numerous forecasts are required in order to reliably evaluate the skill, the evaluation procedure is time-consuming and tedious. Moreover, the evaluation is highly subjective, because an expert who performs it makes his decision based on his implicit knowledge. Therefore, approaches for the automated evaluation of such forecasts are required. Here, we present a semi-automated approach for the assessment of precipitation forecast ensemble members. The approach is based on supervised machine learning and was tested on ensemble precipitation forecasts for the area of the Mulde river basin in Germany. Based on the evaluation results of the specific ensemble members, weights corresponding to their forecast skill were calculated. These weights were then successfully used to reduce the uncertainties within rainfall-runoff simulations and flood risk predictions. [ABSTRACT FROM AUTHOR]

The prediction method plays crucial roles in accurate precipitation forecasts. Recently, machine learning has been widely used for forecasting precipitation, and the K -nearest neighbor (KNN) algorithm, one of machine learning techniques, showed good performance. In this paper, we propose an improved KNN algorithm, which offers robustness against different choices of the neighborhood size k , particularly in the case of the irregular class distribution of the precipitation dataset. Then, based our improved KNN algorithm, a new precipitation forecast approach is put forward. Extensive experimental results demonstrate that the effectiveness of our proposed precipitation forecast approach based on improved KNN algorithm. [ABSTRACT FROM AUTHOR]

Probabilistic seismic demand models are widely used for structures to establish a relation between the engineering demand parameter (EDP) and ground motion intensity measures (IM). For the complex infrastructures such as dams two challenges in implementation of probabilistic seismic demand models are scarcity of the appropriate real ground motions, and computational limitation to perform hundreds of simulations. This paper addresses both concerns by using a series of stochastic ground motion records as an alternative for real ones, and also employing machine learning algorithms to predict the IM-EDP relation. A 3D concrete arch dam with foundation and reservoir interaction is used as a case study of a demanding system (as opposed to 2D framed building). A large group of about 600 real and stochastic ground motion records are used to analyze the coupled system. In addition, five machine learning algorithms were used to develop the predictive meta-models. This paper also highlights the feasibility of using stochastic ground motions to predict the probabilistic seismic demand meta-models and fragility curves from real records, and vice versa. While the outcomes illustrate promising results, they also show that the existing stochastic ground motion simulation models do not cover all the inherent characteristics of the real records. [ABSTRACT FROM AUTHOR]

Autonomous driving and the machine learning (ML) models developed to achieve it have grown rapidly in importance and complexity respectively, and with them has also grown their carbon footprint due to long training times. Given the importance of climate change, it should be necessary to include the CO 2 impact of ML models explicitly to encourage competition on more than just model quality. This work presents the implementation of two different Convolutional Neural Network (CNN) training approaches, used for autonomous driving by behavioral cloning in a simulated environment and compares their impact on the CO 2 footprint. Using a cloud execution environment and driving data, previously obtained by applying an end-to-end deep learning technique, the first implemented training approach is a classical approach that uses an image generator that carries out the pre-processing and data augmentation during training. The contribution proposed in this paper is a second approach that improves the training by decreasing the training time, carrying out the data augmentation and pre-processing tasks before training the model, storing the result in RAM and then starting the training. The new approach to training presented in this article finishes the training approximately 38 times faster and reduces the carbon footprint impact by approximately 96%. In absolute values, this is a reduction from an average value of approximately 0.1643 CO 2 e q (kg) to 0.007 CO 2 e q (kg). To estimate the impact of CO 2 , the hardware used for the project, the training time and the cloud service provider were all taken into account. [ABSTRACT FROM AUTHOR]

• One-shot-fault learning for fault severity estimation. • Physical domain adaptation for TDM. • Adaptation of a simulation to an experiment with a one-fault-shot example. • Estimation of the gear fault size using machine learning. Gearboxes are integral elements in rotating machines and have a high tendency to fail due to their operation in harsh conditions. A robust method to estimate the fault size of gears is desirable for a successful prognostic process, which is, to date, still unavailable in the literature. The fault size can be estimated by vibration analysis, using signal processing and machine-learning tools. However, the availability of labeled or unlabeled vibration signals from faulty rotating machinery components is rare, making it challenging to apply machine-learning algorithms. Therefore, some physical pre-knowledge should be incorporated in the model for a successful learning process. This can be done by exposing the learning model to simulated data, and by a physical pre-processing procedure. This paper suggests a novel algorithm to overcome the lack of faulty data (labeled and unlabeled), and it is trained on a combination of simulated data and some real data. The algorithm tunes the differences between simulation and experiment using one faulty experimental example, and transfers knowledge from simulation to reality by addressing the transfer function effects. It addresses the transfer function by spectrum background estimation and minimum phase estimation while also selecting features that are invariant to the unmitigated effects of the transfer function. The new algorithm is demonstrated on simulated signals and measured transfer function, and on experimental signals with known fault sizes. The codes and the data of the study are available via the link: https://github.com/omriMatania/one_fault_shot_learning_for_gears_fault_severity_estimation. [ABSTRACT FROM AUTHOR]

Complete and high-quality deformation monitoring data are critical for shield tunnel construction safety and quality. In engineering practices, data missing frequently occurs during instrumentation, adversely impacting further analysis and decision making. Existing imputation methods either ignore the crucial interactions between different parameters during shield tunneling or focus on the global characteristics of deformation data while neglect their local difference. This paper proposes a novel hybrid model, MCCB, combining multi-view matrix completion algorithms, convolutional neural network (CNN), and bidirectional long short-term memory (BiLSTM) algorithms to impute missing deformation values in shield tunnel monitoring data. The performance of the proposed method is verified using bridge deformation data from a practical project in Beijing. Different missing patterns of the bridge deformation data are filled. The experiment results show that the proposed model can effectively learn the various characteristics of the deformation data and outperforms the four selected models and its two sub-models, and can be used to improve the accuracy of the deformation prediction through data imputation. The novelty of this study includes two aspects. First is that the complicated interaction between different parameters and local difference of the data are considered simultaneously. They have not been addressed before by existing studies. Second is that the innovative combination of matrix completion and deep learning algorithm for application in missing deformation values imputation. To our best knowledge, no research on engineering construction has implemented this technique before. [ABSTRACT FROM AUTHOR]

As Building Information Modeling (BIM) workflows are becoming very relevant for the different stages of the project's lifecycle, more data is produced and managed across it. The information and data accumulated in BIM-based projects present an opportunity for analysis and extraction of project knowledge from the inception to the operation phase. In other industries, Machine Learning (ML) has been demonstrated to be an effective approach to automate processes and extract useful insights from different types and sources of data. The rapid development of ML applications, the growing generation of BIM-related data in projects, and the different needs for use of this data present serious challenges to adopt and effectively apply ML techniques to BIM-based projects in the Architecture, Engineering, Construction and Operations (AECO) industry. While research on the use of BIM data through ML has increased in the past decade, it is still in a nascent stage. In order to asses where the industry stands today, this paper carries out a systematic literature review (SLR) identifying and summarizing common emerging areas of application and utilization of ML within the context of BIM-generated data. Moreover, the paper identifies research gaps and trends. Based on the observed limitations, prominent future research directions are suggested, focusing on information architecture and data, applications scalability, and human information interactions. [ABSTRACT FROM AUTHOR]

With the acceleration of the upgrading of the automobile consumption market, artificial intelligence has become an increasingly effective means of enhancing the creative design of automobile appearance modeling. However, when artificial intelligence processes specific design tasks, creativity is primarily based on data drive, resulting in machine-generated design schemes that do not match human-specific psychological intentions. Due to the absence of design knowledge in the process of machine design, there is a data gap between human cognitive thought and machine information processing. This paper aims to structure the human's complex cognitive knowledge of car frontal form, establish the consistency between human and machine cognitive structures, and reduce communication barriers in the process of human–machine hybrid creative design. To achieve this objective, a human–machine hybrid intelligence methodology – a combination of human cognitive mental model, human–machine shared knowledge base, and Generative Adversarial Networks (GAN) – was developed to generate a large number of car frontal forms that are consistent with the design intent. First, we constructed a mental model of human cognition based on three dimensions: design intent, drawing behavior, and functional structure. Second, we created a shared human–machine knowledge base with design Knowledge. This knowledge base contains 12,560 images of car frontal form designs with corresponding morphological semantic labels and 3,140 sketches of car frontal forms drawn by hand. Human–machine shared knowledge base data was utilized in a machine learning training network. In addition, a conditional cross-domain generative adversarial network was developed to investigate the implicit relationship between sketch characteristics, morphological semantics, and image visual effects. Using the suggested method, a large number of images with the specified morphological semantic category and resembling the hand-drawn sketch of a car frontal form can be generated rapidly. In terms of the quality of car frontal form generation, our research is superior to the baseline model according to qualitative and quantitative assessments. In comparison to the designer's output, the human–machine hybrid intelligent generation also demonstrates excellent creative performance. [ABSTRACT FROM AUTHOR]

Data-based models are widely applied in concrete dam health monitoring. However, most existing models are restricted to offline modeling, which cannot continuously track the displacement behavior with dynamic evolution patterns, especially in time-varying environments. In this paper, sequential learning is introduced to establish an online monitoring model for dam displacement behavior. This approach starts by considering the timeliness difference between old and new data using the forgetting mechanism, and a novel adaptive forgetting extreme learning machine (AF-ELM) is presented. A primary predictor based on AF-ELM is then formulated, which aims to sequentially learn the complex nonlinear relationship between dam displacement and main environmental factors. Considering the chaotic characteristics contained in the residual sequence of the primary predictor, a multi-scale residual-error correction (REC) strategy is devised based on divide-and-conquer scheme. Specifically, time-varying filter-based empirical mode decomposition is adopted to decompose the raw chaotic residual-error series into a set of subseries with more stationarity, which are further aggregated and reconstructed by fuzzy entropy theory and suitable approximation criterion. Finally, the corrected residual sequence is superimposed with the preliminary predictions from AF-ELM to generate the required modeling results. The effectiveness of the proposed model is verified and assessed by taking a real concrete dam as an example and comparing prediction performance with state-of-the-art models. The results show that AF-ELM performs better in displacement prediction compared with benchmark models, and the multi-scale REC can effectively identify the valuable information within the residual sequence. The proposed online monitoring model can more closely track the dynamic variations of displacement data, which provides a fire-new solution for dam behavior prediction and analysis. [ABSTRACT FROM AUTHOR]

Drawing the strengths of data science and machine learning, process mining has recently emerged as an effective research approach for process management and its decision support. Bottleneck identification and analysis is a key problem in process mining which is considered a critical component for process improvement. While previous studies focusing on bottlenecks have been reported, visible gaps remain. Most of these studies considered bottleneck identification from local perspectives by quantitative metrics, such as machine operation and resource requirement, which can not be applied to knowledge-intensive processes. Moreover, the root cause of such bottlenecks has not been given enough attention, which limits the impact of process optimisation. This paper proposes an approach that utilises fusion-based clustering and hyperbolic neural network-based knowledge graph embedding for bottleneck identification and root cause analysis. Firstly, a fusion-based clustering is proposed to identify bottlenecks automatically from a global perspective, where the execution frequency of each stage at different periods is calculated to reveal the abnormal stage. Secondly, a process knowledge graph representing tasks, organisations, workforce and relation features as hierarchical and logical patterns is established. Finally, a hyperbolic cluster-based community detection mechanism is researched, based on the process knowledge graph embedding trained by a hyperbolic neural network, to analyse the root cause from a process perspective. Experimental studies using real-world data collected from a multidisciplinary design project revealed the merits of the proposed approach. The execution of the proposed approach is not limited to event logs; it can automatically identify bottlenecks without local quantitative metrics and analyse the causes from a process perspective. [ABSTRACT FROM AUTHOR]

Traditional control methods in automatic train operation (ATO) models have some disadvantages, such as high energy consumption and low riding comfort. To alleviate these shortcomings of the ATO models, this paper presents three data-driven train operation (DTO) models from a new perspective that combines data mining methods with expert knowledge, since the manual driving by experienced drivers can achieve better performance than ATO model in some degree. Based on the experts knowledge that are summarized from experienced train drivers, three DTO models are developed by employing K-nearest neighbor (KNN) and ensemble learning methods, i.e., Bagging-CART (B-CART) and Adaboost.M1-CART (A-CART), into experts systems for train operation. Furthermore, the DTO models are improved via a heuristic train parking algorithm (HPA) to ensure the parking accuracy. With the field data in Chinese Dalian Rapid Rail Line 3 (DRRL3), the effectiveness of the DTO models are evaluated on a simulation platform, and the performance of the proposed DTO models are compared with both ATO and manual driving strategies. The results indicate that the developed DTO models obtain all the merits of the ATO models and the manual driving. That is, they are better than the ATO models in energy consumption and riding comfort, and also outperform the manual driving in stopping accuracy and punctuality. Additionally, the robustness of the proposed model is verified by a number of experiments with some steep gradients and complex speed limits. [ABSTRACT FROM AUTHOR]

Abstract: Although researchers have developed numerous computational approaches to reasoning and knowledge representation, their implementations are always limited to specific applications (e.g. assembly planning, fault diagnosis or production scheduling) for which bespoke knowledge bases or algorithms have been created. However, “cloud computing” has made irrelevant both the physical location and internal processes used by machine intelligence. In other words, the Internet encourages functional processes to be treated as ‘black boxes’ with which users need only be concerned with posing the right question and interpreting the response. The system asking the questions does not need to know how answers are generated, only that they are available in an appropriate time frame. This paper proposes that Crowdsourcing could provide on-line, ‘black-box’, reasoning capabilities that could far exceed the capabilities of current AI technologies (i.e. genetic algorithms, neural-nets, case-based reasoning) in terms of flexibility and scope. This paper describes how Crowdsourcing has been deployed in three different reasoning scenarios to carry out industrial tasks that involve significant amounts of tacit (e.g. unformalised) knowledge. The first study reports the application of Crowdsourcing to identify canonical view of 3D CAD models. The qualitative results suggest that the anonymous, Internet, workforce have a good comprehension of 3D geometry. Having established this basic competence the second experiment assesses the Crowd’s ability to judge the similarity of 3D components. Comparison of the results with published benchmarks shows a high degree of correspondence. Lastly the performance of the Internet labourers is quantified in a 2D nesting task, where their performance is found to be superior to reported computational algorithms. In all these cases results were returned within a couple of hours and the paper concludes that there is potential for broad application of Crowdsourcing to geometric problem solving in CAD/CAM. [Copyright &y& Elsevier]

The colored paintings on the surfaces of ancient Chinese buildings have suffered from the wind and the sun and produced many defects such as paint loss, blurring, and color distortion. Previous methods have not been able to virtually repair them well. Therefore, this paper proposed a virtual restoration method for the weathered beams in the Forbidden City using multiple deep learning algorithms. Instead of using only one technology to restore paintings, this paper divided the painting into 3 parts, i.e., the background, the golden edges, and the dragon patterns, and restored them in different technology. For the background, this paper transformed the problem of unrecognizable color restoration into a semantic segmentation problem using U-Net MobileNet. For the golden edges, this paper used traditional image processing technology to obtain them from the color maps generated by the semantic segmentation algorithm. For the dragon patterns, after sketching the skeletons according to the dragon patterns, the image translation algorithm Pix2pix was applied to generate a realistic dragon pattern. Finally, the three repair results are superimposed to complete the repair. The virtually restored paintings can provide reference and guidance for traditional manual restoration and help the restorers to imagine what they looked like before oxidation, thus alleviating some repetitive work and reducing the complexity of restoration. Besides, each step of the method could form a layer, so that the restorers could overlay or modify them as they wish. [ABSTRACT FROM AUTHOR]

• Survey of papers with historic perspective. • Discussion on implementation of analytics for prognosis and health management. • Implementation of machine learning beyond black-box models. Prognosis and health management plays an important role in the control of costs associated with operating large industrial equipment, such as wind turbines and aircraft. It is only fair that engineers and scientists have vastly researched modeling approaches to support decision making. Motivated by the growing availability of data and computational power as well as the advances in algorithms and methods, modeling frameworks often merge elements of physics, machine learning, and statistical learning. In this paper, we present a review on modeling in support of prognosis and health management of industrial equipment. This survey complements the existing prognosis and health management literature by discussing how modeling strategies are influenced by industry-specific aspects such as maintenance approaches (e.g., reactive, proactive, and predictive), implementation factors (e.g., industry, business model, purpose, development, and deployment), as well as supporting technologies (sensing, repair, and modeling itself). We use the onshore wind energy and civil aviation industries to illustrate how these aforementioned aspects can influence modeling and implementation of prognosis and health management. The literature review is broad and covers contributions over the past 40 years. We close the paper with few topics that can motive research going forward. [ABSTRACT FROM AUTHOR]

• Monitoring and assessing awkward postures help to prevent injuries in construction. • Incremental DNN-based recognition models can continuously learn new postures. • Incremental models can keep the memory of learned postures after adaptation. • Ergonomics assessment with recognized and actual postures shows comparable results. Monitoring and assessing awkward postures is a proactive approach for Musculoskeletal Disorders (MSDs) prevention in construction. Machine Learning models have shown promising results when used in recognition of workers' posture from Wearable Sensors. However, there is a need to further investigate: i) how to enable Incremental Learning, where trained recognition models continuously learn new postures from incoming subjects while controlling the forgetting of learned postures; ii) the validity of ergonomics risk assessment with recognized postures. The research discussed in this paper seeks to address this need through an adaptive posture recognition model– the incremental Convolutional Long Short-Term Memory (CLN) model. The paper discusses the methodology used to develop and validate this model's use as an effective Incremental Learning strategy. The evaluation was based on real construction workers' natural postures during their daily tasks. The CLN model with "shallow" (up to two) convolutional layers achieved high recognition performance (Macro F1 Score) under personalized (0.87) and generalized (0.84) modeling. Generalized CLN model, with one convolutional layer, using the "Many-to-One" Incremental Learning scheme can potentially balance the performance of adaptation and controlling forgetting. Applying the ergonomics rules on recognized and ground truth postures yielded comparable risk assessment results. These findings support that the proposed incremental Deep Neural Networks model has a high potential for adaptive posture recognition. They can be deployed alongside ergonomics rules for effective MSDs risk assessment. [ABSTRACT FROM AUTHOR]

Extreme learning machine (ELM) is widely used in complex industrial problems, especially the online-sequential extreme learning machine (OS-ELM) plays a good role in industrial online modeling. However, OS-ELM requires batch samples to be pre-trained to obtain initial weights, which may reduce the timeliness of samples. This paper proposes a novel model for the online process regression prediction, which is called the Recurrent Extreme Learning Machine (Recurrent-ELM). The nodes between the hidden layers are connected in Recurrent-ELM, thus the input of the hidden layer receives both the information from the current input layer and the previously hidden layer. Moreover, the weights and biases of the proposed model are generated by analysis rather than random. Six regression applications are used to verify the designed Recurrent-ELM, compared with extreme learning machine (ELM), fast learning network (FLN), online sequential extreme learning machine (OS-ELM), and an ensemble of online sequential extreme learning machine (EOS-ELM), the experimental results show that the Recurrent-ELM has better generalization and stability in several samples. In addition, to further test the performance of Recurrent-ELM, we employ it in the combustion modeling of a 330 MW coal-fired boiler compared with FLN, SVR and OS-ELM. The results show that Recurrent-ELM has better accuracy and generalization ability, and the theoretical model has some potential application value in practical application. [ABSTRACT FROM AUTHOR]

Recent advances in physiological computing have been made due to Artificial Intelligence and Machine Learning, which have profoundly begun to influence occupational health and safety (OHS) in construction. Acknowledging the current and future use of physiological computing, we address the following research question in this paper: What developments in physiological computing can be used to improve OHS in construction ? Using a narrative systematic review, we examine studies that have used physiological computing in construction to monitor people OHS. Our review indicates that there is a need for physiological computing systems to be: (1) more accurate; (2) portable and easier to use; (3) generalizable across varying work tasks; and (4) accepted by users and their benefits realized. Considering our observations derived from the prevailing literature and practice, we suggest that future research should aim to mitigate OHS risks by focusing on: (1) development of high-quality database; (2) feature engineering extraction by using an array of machine learning techniques; (3) understanding the context and enacting intervention strategies. The upshot of performing such a review is to provide a signpost for future research in physiological computing of OHS in construction. [ABSTRACT FROM AUTHOR]

[Display omitted] • Automated condition assessment of pipelines is achieved by machine learning models. • Detecting, locating, and quantifying anomalies of pipelines are reviewed. • Operation data, nondestructive testing data, and computer vision data are covered. • SWOT analysis and practical recommendations are discussed. Pipelines carrying energy products play vital roles in economic wealth and public safety, but incidents continue occurring. Condition assessment of pipelines is essential to identify anomalies timely. Advanced sensing technologies obtain informative data for condition assessment, while data analysis by human has limited efficiency, accuracy, and reliability. Advances in machine learning offer exciting opportunities for automated condition assessment with minimum human intervention. This paper reviews machine learning approaches to detect, classify, locate, and quantify pipeline anomalies based on intelligent interpretation of routine operation data, nondestructive testing data, and computer vision data. Statistics and uncertainties of performance metrics of machine learning approaches are discussed. An analysis on strengths, weaknesses, opportunities, and threats (SWOT) is performed. Guides for practitioners to perform automated pipeline condition assessment are recommended. This review provide insights into the machine learning approaches for automated pipeline condition assessment. The SWOT analysis will support decision making in the pipeline industry. [ABSTRACT FROM AUTHOR]

Subsoil schematisations are a paramount activity for the execution of infrastructure earthworks. Currently, subsoil schematisations are laborious, and mostly based on correlation and interpolation of available geotechnical and geophysical in-situ data. Geotechnical in-situ testing tends to be accurate but merely provides local information, while geophysical investigations are often performed to image the subsurface, and provide valuable insight for areas. Yet, there is not always a clear way to reflect geophysical properties to soil schematisation. This paper presents a data fusion methodology to perform subsoil schematisation and parametrisation. This methodology makes use of available geotechnical, geological and geophysical data sets, and combines them by means of machine learning algorithms (Neural Networks and Random Forest). The data fusion method is applied on two case studies. The first case study concerns a large area of a flood defence line, and the second case study concerns a small polder area next to a dike. The accuracy of the data fusion algorithm is assessed by comparing its results against a validation data set, that has not been exposed to the data fusion algorithm. The performance of the two algorithms and of the parameters that govern each algorithm is discussed. The results show that both Neural Networks and Random Forest are suitable to perform subsoil schematisations. The analyses show that the Random Forest leads to a lower error on the validation data set. However, Random Forest fails to predict the occurrence of thin clay layers in the second case study, while Neural Networks are successful at it. The data fusion methodology shows the potential to enhance the subsoil schematisation procedure, by increasing the schematisation spatial resolution. [ABSTRACT FROM AUTHOR]

Context : An important task in civil engineering is the detection of collisions of a 3D model with an environment representation. Existing methods using the structure gauge provide an insufficient measure because the model either rotates or because the trajectory makes tight turns through narrow passages. This is the case in either automotive assembly lines or in narrow train tunnels. Objective : Given two point clouds, one of the environment and one of a model and a trajectory with six degrees of freedom along which the model moves through the environment, find all colliding points of the environment with the model within a certain clearance radius. Method : This paper presents two collision detection (CD) methods called kd-CD and kd-CD-simple and two penetration depth (PD) calculation methods called kd-PD and kd-PD-fast . All four methods are based on searches in a k -d tree representation of the environment. The creation of the k -d tree, its search methods and other features will be explained in the scope of their use to detect collisions and calculate depths of penetration. Results : The algorithms are benchmarked by moving the point cloud of a train wagon with 2.5 million points along the point cloud of a 1144 m long train track through a narrow tunnel with overall 18.92 million points. Points where the wagon collides with the tunnel wall are visually highlighted with their penetration depth. With a safety margin of 5 cm kd-PD-simple finds all colliding points on its trajectory which is sampled into 19,392 positions in 77 s on a standard desktop machine of 1.6 GHz. Conclusion : The presented methods for collision detection and penetration depth calculation are shown to solve problems for which the structure gauge is an insufficient measure. The underlying k -d tree is shown to be an effective data structure for the required look-up operations. [ABSTRACT FROM AUTHOR]

Timely and accurate monitoring of onsite construction operations can bring an immediate awareness on project specific issues. It provides practitioners with the information they need to easily and quickly make project control decisions. Despite their importance, the current practices are still time-consuming, costly, and prone to errors. To facilitate the process of collecting and analyzing performance data, researchers have focused on devising methods that can semi-automatically or automatically assess ongoing operations both at project level and operation level. A major line of work has particularly focused on developing computer vision techniques that can leverage still images, time-lapse photos and video streams for documenting the work in progress. To this end, this paper extensively reviews these state-of-the-art vision-based construction performance monitoring methods. Based on the level of information perceived and the types of output, these methods are mainly divided into two categories (namely project level: visual monitoring of civil infrastructure or building elements vs. operation level: visual monitoring of construction equipment and workers). The underlying formulations and assumptions used in these methods are discussed in detail. Finally the gaps in knowledge that need to be addressed in future research are identified. [ABSTRACT FROM AUTHOR]

Existing physics-based modeling approaches do not have a good compromise between performance and computational efficiency in predicting the seismic response of reinforced concrete (RC) frames, where high-fidelity models (e.g., fiber-based modeling method) have reasonable predictive performance but are computationally demanding, while more simplified models (e.g., shear building model) are the opposite. This paper proposes a novel artificial intelligence (AI)-enhanced computational method for seismic response prediction of RC frames which can remedy these problems. The proposed AI-enhanced method incorporates an AI technique with a shear building model, where the AI technique can directly utilize the real-world experimental data of RC columns to determine the lateral stiffness of each column in the target RC frame while the structural stiffness matrix is efficiently formulated via the shear building model. Therefore, this scheme can enhance prediction accuracy due to the use of real-world data while maintaining high computational efficiency due to the incorporation of the shear building model. Two data-driven seismic response solvers are developed to implement the proposed approach based on a database including 272 RC column specimens. Numerical results demonstrate that compared to the experimental data, the proposed method outperforms the fiber-based modeling approach in both prediction capability and computational efficiency and is a promising tool for accurate and efficient seismic response prediction of structural systems. [ABSTRACT FROM AUTHOR]

Accurate prediction of electricity consumption is essential for providing actionable insights to decision-makers for managing volume and potential trends in future energy consumption for efficient resource management. A single model might not be sufficient to solve the challenges that result from linear and non-linear problems that occur in electricity consumption prediction. Moreover, these models cannot be applied in practice because they are either not interpretable or poorly generalized. In this paper, a stacking ensemble model for short-term electricity consumption is proposed. We experimented with machine learning and deep models like Random Forests, Long Short Term Memory, Deep Neural Networks, and Evolutionary Trees as our base models. Based on the experimental observations, two different ensemble models are proposed, where the predictions of the base models are combined using Gradient Boosting and Extreme Gradient Boosting (XGB). The proposed ensemble models were tested on a standard dataset that contains around 500,000 electricity consumption values, measured at periodic intervals, over the span of 9 years. Experimental validation revealed that the proposed ensemble model built on XGB reduces the training time of the second layer of the ensemble by a factor of close to 10 compared to the state-of-the-art , and also is more accurate. An average reduction of approximately 39% was observed in the Root mean square error. [ABSTRACT FROM AUTHOR]

Abstract: The field of building energy management, which monitors and analyses the energy use of buildings with the aim to control and reduce energy expenditure, is seeing a rapid evolution. Automated meter reading approaches, harvesting data at hourly or even half-hourly intervals, create a large pool of data which needs analysis. Computer analysis by means of machine learning techniques allows automated processing of this data, invoking expert analysis where anomalies are detected. However, machine learning always requires a historical dataset to train models and develop a benchmark to define what constitutes an anomaly. Computer analysis by means of building performance simulation employs physical principles to predict energy behaviour, and allows the assessment of the behaviour of buildings from a pure modelling background. This paper explores how building simulation approaches can be fused into energy management practice, especially with a view to the production of artificial bespoke benchmarks where historical profiles are not available. A real accommodation block, which is subject to monitoring, is used to gather an estimation of the accuracy of this approach. The findings show that machine learning from simulation models has a high internal accuracy; comparison with actual metering data shows prediction errors in the system (20%) but still achieves a substantial improvement over industry benchmark values. [Copyright &y& Elsevier]

Transfer learning (TL) is a machine learning (ML) method in which knowledge is transferred from the existing models of related problems to the model for solving the problem at hand. Relational TL enables the ML models to transfer the relationship networks from one domain to another. However, it has two critical issues. One is determining the proper way of extracting and expressing relationships among data features in the source domain such that the relationships can be transferred to the target domain. The other is how to do the transfer procedure. Knowledge graphs (KGs) are knowledge bases that use data and logic to graph-structured information; they are helpful tools for dealing with the first issue. The proposed relational feature transfer learning algorithm (RF-TL) embodies an extended structural equation modelling (SEM) as a method for constructing KGs. Additionally, in fields such as medicine, economics, and law related to people's lives and property safety and security, the knowledge of domain experts is a gold standard. This paper introduces the causal analysis and counterfactual inference in the TL domain that directs the transfer procedure. Different from traditional feature-based TL algorithms like transfer component analysis (TCA) and CORelation Alignment (CORAL), RF-TL not only considers relations between feature items but also utilizes causality knowledge, enabling it to perform well in practical cases. The algorithm was tested on two different healthcare-related datasets — sleep apnea questionnaire study data and COVID-19 case data on ICU admission — and compared its performance with TCA and CORAL. The experimental results show that RF-TL can generate better transferred models that give more accurate predictions with fewer input features. [ABSTRACT FROM AUTHOR]