Showing papers in "Advanced Engineering Informatics in 2020"

TL;DR: This is a foundational study that formalises and categorises the existing usage of AR and VR in the construction industry and provides a roadmap to guide future research efforts.

TL;DR: An industry-led case study demonstrates how to turn conventional appliances to smart products and systems (SPS) by utilising the state-of-the-art Industry 4.0 technologies.

TL;DR: Comparison was made with the segmentation results produced by other automatic classical methods, revealing that the results made by the anomaly map outperform other segmentation methods, in terms of precision, recall, F1 measure and F2 measure, without severe under- and over-segmentation.

TL;DR: The developments of computer vision studies used to identify unsafe behaviour from 2D images that arises on construction sites are reviewed and in light of advances made with deep learning, its integration with computer vision to support BBS is examined.

TL;DR: This research introduces and evaluates a series of convolutional neural network (CNN) models for ground object detection from aerial views of disaster’s aftermath that are capable of recognizing critical ground assets including building roofs (both damaged and undamaged), vehicles, vegetation, debris, and flooded areas.

TL;DR: An intelligent approach is presented for FM classification and bearing capacity prediction of RC columns based on the ensemble machine learning techniques and shows that ensemble learning (especially AdaBoost) has better performance than single learning.

TL;DR: The results showed that uneven splits of crowd flow motivated participants under mental stress to follow the majority of the crowd over the course of evacuation, and such consistency could be reinforced by stronger directional information conveyed by the crowd flow as well as positive feedback from the outcomes of previous wayfinding decisions.

TL;DR: A smart surface inspection system is proposed using faster R-CNN as a CNN-based object detection method that can efficiently identify defects in complex product images and the cloud-edge computing framework can provide fast computation speed and evolving algorithm models.

TL;DR: An improved deep learning-based approach to automatically classify near-miss information contained within safety reports using Bidirectional Transformers for Language Understanding (BERT) is developed, designed to pre-train deep bi-directional representations by jointly extracting context features in all layers.

TL;DR: Results show that by employing a new integrated solution of predictive model-based quality inspection in industrial manufacturing by utilizing Machine Learning techniques and Edge Cloud Computing technology, inspection volumes can be reduced significantly and thus economic advantages can be generated.

TL;DR: An integrated detection framework of solder joint defects in the context of Automatic Optical Inspection (AOI) of Printed Circuit Boards (PCBs) is proposed and an active learning method was proposed to reduce the labeling workload when a large labeled training database is not easily available.

TL;DR: It is demonstrated that a hybrid light gradient boosting and natural gradient boosting model provides the most desirable construction cost estimates in terms of the accuracy metrics, uncertainty estimates, and training speed.

TL;DR: The proposed system, Telerobotic Operation based on Auto-reconstructed Remote Scene (TOARS), utilizes a deep learning algorithm to automatically detect objects in the captured scene, along with their physical properties, based on the point cloud data.

TL;DR: A new framework based on small labeled infrared thermal images and enhanced convolutional neural network (ECNN) transferred from Convolutional auto-encoder (CAE) to outperforms the current mainstream methods in intelligent fault diagnosis of rotor-bearing systems.

TL;DR: A BIM-based simulation framework that implements the Fire Dynamic Simulator and agent-based modeling for simulating fire growth and evacuation performance for different building layout scenarios is developed and contributes to building fire safety management by enabling to minimize both injuries/fatalities and property loss.

TL;DR: A text-mining based global supply chain risk management framework involving two phases, a risk hierarchy and sentiment analysis results can improve the understanding of regionalglobal supply chain risks and provide guidance in supplier selection.

TL;DR: A novel and robust framework that combines deep learning and text mining technologies that provide the ability to analyse hazard records automatically, enabling managers to understand their patterns of manifestation and therefore put in place strategies to prevent them from reoccurring is presented.

TL;DR: Evaluated ensemble learning potential in predicting microbially induced concrete corrosion in sewer systems from the data mining (DM) perspective indicates that the prediction ability of the random forests DM model is superior to the other ensemble learners, followed by the ensemble Bag-CHAID method.

TL;DR: This study released road-damage image data from the road maintenance agency in Zhubei, Hsinchu County, Taiwan for research and other uses, increasing the limited amount of published image data sources and positively impacting future scholarly research into road damage detection.

TL;DR: A robust real-time health monitoring framework for detecting performance anomalies, which may impact system safety during flight operations, with high accuracy and generalized applicability is proposed.

TL;DR: A hybrid deep neural network, combining the bidirectional long short-term memory (Bi-LSTM) and the conditional random field (CRF) for the automatic extraction of the qualitative construction procedural constraints is explored, demonstrating the good performance of the end-to-endDeep neural network in the extraction of construction procedural constraint constraints.

TL;DR: A BIM (Building Information Model)-based robotic assembly model that contains all the required information for planning was proposed, and the implementation details were described to improve the planning efficiency of robotic brick assembly without affecting accuracy.

TL;DR: The proposed AutoSM, unlike previous EA-based automatic surrogate model selection methods, is not a black box and is interpretable, and can find the promising surrogate model and associated hyper-parameter in 9 times less than other automatic selection approaches while maintaining the same accuracy and robustness in surrogatemodel selection.

TL;DR: An IoT edge computing enabled collaborative tracking architecture is developed to offload the computation pressure and realize distributed decision making and a supervised learning of genetic tracking method is innovatively presented to ensure tracking accuracy and effectiveness.

TL;DR: In this paper, the authors propose an integrated framework to match DLT design options with desired characteristics of a use case, and analyse the use cases using the new framework, which can guide future implementers toward more connected and structured thinking between the technological properties of DLT and use cases in construction.

TL;DR: A new QFD approach integrating picture fuzzy linguistic sets (PFLS) and the evaluation based on distance from average solution (EDAS) method is proposed for the determination of ranking order of ECs.

TL;DR: This paper proposed algorithms using inter-subject transfer learning for EEG-based mental fatigue recognition, which did not need a calibration, and explored the influence of the number of EEG channels on the algorithms’ accuracy.

TL;DR: A novel reinforcement learning based optimizer with the integration of deep-Q network (DQN) and particle swarm optimization (PSO) is proposed to improve the extreme learning machine (ELM) based tunneling-induced settlement prediction model.

TL;DR: A new approach called Cox proportional hazard deep learning (CoxPHDL) is proposed to tackle the issues of data sparsity and data censoring that are common in the analysis of operational maintenance data and offers an integrated solution by taking advantage of deep learning and reliability analysis.

TL;DR: A combination deformation prediction model considering both quantitative evaluation of influencing factors and hysteresis correction is presented in order to further improve estimation accuracy and provide an alternative method for predicting and analyzing dam deformation and evolution behavior of other similar hydraulic structures.