Content uploaded by Taofeek Akinosho
Author content
All content in this area was uploaded by Taofeek Akinosho on Oct 09, 2020
Content may be subject to copyright.
Deep Learning in the Construction Industry: A Review of Present Status and Future Innovations
Abstract
The construction industry is known to be overwhelmed with resource planning, risk management and logistic challenges which often result in design defects, project delivery delays, cost overruns and contractual disputes. These challenges have instigated research in the application of advanced machine learning algorithms such as deep learning to help with diagnostic and prescriptive analysis of causes and preventive measures. However, the publicity created by tech firms like Google, Facebook and Amazon about Artificial Intelligence and applications to unstructured data is not the end of the field. There abound many applications of deep learning, particularly within the construction sector in areas such as site planning and management, health and safety and construction cost prediction, which are yet to be explored. The overall aim of this article was to review existing studies that have applied deep learning to prevalent construction challenges like structural health monitoring, construction site safety, building occupancy modelling and energy demand prediction. To the best of our knowledge, there is currently no extensive survey of the applications of deep learning techniques within the construction industry. This review would inspire future research into how best to apply image processing, computer vision, natural language processing techniques of deep learning to numerous challenges in the industry. Limitations of deep learning such as the black box challenge, ethics and GDPR, cybersecurity and cost, that can be expected by construction researchers and practitioners when adopting some of these techniques were also discussed.
... The use of artificial intelligence in CAD and BIM systems such as Revit and ArchiCAD marks a notable advancement in architectural design (Akinosho et al., 2020). These systems provide sophisticated modeling and simulation that can be augmented with artificial intelligence-based optimization techniques (Urbieta et al., 2023). ...
The integration of artificial intelligence (AI) into architectural floorplan layout planning represents a transformative advancement in spatial design, offering new avenues for efficiency, adaptability, and personalization. This study investigates the application of advanced AI methodologies—including deep learning, generative models, reinforcement learning, and hybrid systems—in addressing the inherent limitations of traditional and rule-based layout design approaches. Through a systematic literature review of peer-reviewed publications from 2020 to 2025, the research synthesizes current innovations, identifies thematic trends, and evaluates model performance across diverse design contexts. The findings highlight the capacity of AI to overcome key constraints such as spatial inefficiencies, lack of flexibility, and limited adaptability to user requirements. Results indicate that hybrid models combining generative adversarial networks, graph-based reasoning, and human-in-the-loop mechanisms offer superior design control and responsiveness to functional, aesthetic, and regulatory standards. Furthermore, the integration of AI into Building Information Modeling (BIM) and Computer-Aided Design (CAD) platforms enhances real-time decision-making, ensuring scalability and user-centered outcomes. While current models demonstrate high efficacy, challenges persist regarding generalization, dataset diversity, and computational resource demands. The study concludes that the continued evolution of AI in floorplan design necessitates the development of universally adaptable algorithms, enriched annotated datasets, and interactive systems that facilitate collaboration between human designers and AI agents.
... Selain itu, kami akan menerapkan teknik statistik untuk menentukan korelasi dan pengaruh antar variabel risiko dalam proyek. [8] Penting untuk memodelkan hubungan antar risiko dalam proyek konstruksi berkelanjutan ini guna mengidentifikasi risiko yang dapat mempengaruhi proyek secara keseluruhan. Dalam konteks ini, risiko dapat meliputi risiko lingkungan, risiko keuangan, risiko regulasi, risiko teknis, risiko keamanan, dan risiko ketergantungan pada pihak ketiga. ...
Konstruksi berkelanjutan telah menjadi fokus utama dalam industri konstruksi karena dampaknya yang positif terhadap lingkungan dan masyarakat. Namun, proyek konstruksi berkelanjutan juga menghadapi berbagai risiko yang dapat mempengaruhi keberhasilan proyek. Oleh karena itu, penting untuk memahami dan mengelola risiko ini dengan baik. Metode yang digunakan dalam penelitian ini adalah analisis risiko yang melibatkan identifikasi, analisis, dan penilaian risiko yang mungkin terjadi dalam proyek konstruksi berkelanjutan. Selain itu, hubungan antar risiko tersebut juga dianalisis menggunakan teknik statistik untuk menentukan korelasi dan pengaruh antar variabel. Penelitian ini bertujuan untuk memodelkan hubungan antara risiko pada proyek konstruksi berkelanjutan di Perumahan Satria Residen Tulungagung. Hasil penelitian menunjukkan adanya beberapa risiko utama yang mempengaruhi proyek konstruksi berkelanjutan di Perumahan Satria Residen Tulungagung, Melalui analisis risiko dalam perbandingan kelompok risiko “Engineer” menunjukkan bahwasanya risiko “kurang bertangung jawab” memiliki kepentingan yang sama dengan risiko “kurang kompeten”. Sedangkan pada kelompok risiko “desain penyebab risiko” risiko “ketidaksesuaian antara gambar” sedikit lebih penting dari pada risiko “kemungkinan perubahan desain. Hasil penelitian ini dapat digunakan oleh pengembang, manajer proyek, dan pihak terkait lainnya dalam merencanakan dan mengelola proyek konstruksi berkelanjutan dengan lebih efektif.
... La innovación es un proceso que se adopta por diferentes actores de acuerdo con su nivel de impacto, en donde ha de diferenciarse entre innovadores, adoptantes tempranos, mayorías tempranas, mayorías tardías y rezagados (Curtis, 2020). Detacando aspectos como la compatibilidadm, la ventaja relativa, la complejidad, entre otros, que terminan por condicionar la adopción de la innovación por parte de la sociedad (Akinosho et al., 2020). Dicho lo anterior, la inversión en I+D se fundamenta en procesos sistémicos orientados a la generación de nuevos saberes, desarrollo tecnológico y mejora de productos y procesos. ...
Introducción: Colombia ocupa el puesto 67 de 133 economías en el GII, aspecto que se une a la baja inversión en innovación que es tan solo del 0,20% de su PIB. Metodología: el presente estudio usa un enfoque cuantitativo y alcance correlacional, en el cual se utilizó una fuente de información secundaria, una muestra de 9304 empresas de servicios y 20 variables numéricas sobre número de innovaciones, dinero invertido en TIC, dinero invertido en capacitación y cantidad de colaboradores dedicados a ACTI. Resultados: se comprueba que a mayor cantidad de colaboradores dedicados a ACTI con títulos de doctorado, maestría, especialización, grado y tecnología se incrementa la cantidad de innovaciones; sin embargo, a mayor dinero invertido en capacitaciones no formales, así como a mayor personal dedicado a ACTI con estudios únicamente de secundaria o sin ningún tipo de estudio formal reducen la cantidad de innovaciones que pueden generar las empresas de este sector. Discusión: El estudio contradice la utilidad de las TIC en la generación de innovaciones, dado que esta resultó no significativa sobre la variable dependiente. Conclusiones: El estudio concluye que personal con mayor nivel de formación es clave para la generación de innovaciones en el sector de comercio y servicios de Colombia.
The article analyzes the existing international experience in the field of monitoring capital construction projects using information modeling technologies, artificial intelligence, etc. It is noted that the works under consideration do not take into account the possibility of using photogrammetry methods to identify defects, which is the subject of the presented study. At the first stage, a sample of publications collected by the keywords "buildings" and "defect monitoring" for the period from 2014 to 2024 in the international Scopus database was used to form a cluster map. In total, 757 publications are presented in the sample. Next, an analysis of the publications presented in the Scopus database is carried out, the main publications corresponding to the research topic are highlighted. For the second stage, photographs of defects in capital construction projects (CCP) are taken, and the possibility of identifying CCP defects based on files in the "point cloud" format obtained from photographs corresponding to various states of the construction project is substantiated.
Steel is an integral part of the infrastructure. Nowadays, steel has become a popular material in infrastructure due to its reduced processing time and easy fitting. Rust damage is one of the major causes of the degradation of steel infrastructure, and effective monitoring of rust is crucial for evaluating the damage to steel structures. The purpose of this research is to develop a near real-time rust recognition method for steel structure health monitoring using transformers-based Convolutional Neural Networks, e.g., SegFormer. The proposed transformer-based CNN rust recognition method is a simple, efficient, and powerful solution for semantic segmentation, which utilizes transformers and lightweight multilayer perceptron (MLP) decoders to reduce processing time and enhance accuracy. Conventional manual rust inspection is time-consuming, sometimes dangerous, and prone to human mistakes. The transformer-based CNN method could tackle these problems and minimize the costly manpower requirement for inspection. Most deep learning models for semantic segmentation were developed solely based on convolutional architecture and required explicit positional encoding. Consequently, the resolution variance between the testing and training phases led to reduced performance. In contrast, SegFormer is a hybrid architecture that integrates components from both transformers and convolutional neural networks (CNNs) and does not require positional encoding. In this research, the proposed method adopts the “light PyTorch” backend to significantly reduce the required processing time and develop a near real-time rust recognition method. It is able to quickly process rust segmentation at the image pixel level, and provide fast rusting deterioration assessment of steel structures.
The construction industry is a high hazard industry. Accidents frequently occur, and part of them are closely relate to workers who are not certified to carry out specific work. Although workers without a trade certificate are restricted entry to construction sites, few ad-hoc approaches have been commonly employed to check if a worker is carrying out the work for which they are certificated. This paper proposes a novel framework to check whether a site worker is working within the constraints of their certification. Our framework comprises key video clips extraction, trade recognition and worker competency evaluation. Trade recognition is a new proposed method through analyzing the dynamic spatiotemporal relevance between workers and non-worker objects. We also improved the identification results by analyzing, comparing, and matching multiple face images of each worker obtained from videos. The experimental results demonstrate the reliability and accuracy of our deep learning-based method to detect workers who are carrying out work for which they are not certified to facilitate safety inspection and supervision.
Equipment and workers are two important resources in the construction industry. Performance monitoring of these resources would help project managers improve the productivity rates of construction jobsites and discover potential performance issues. A typical construction workface monitoring system consists of four major levels: location tracking, activity recognition, activity tracking, and performance monitoring. These levels are employed to evaluate work sequences over time and also assess the workers’ and equipment’s well-being and abnormal edge cases. Results of an automated performance monitoring system could be used to employ preventive measures to minimize operating/repair costs and downtimes. The authors of this paper have studied the feasibility of implementing a wide range of technologies and computational techniques for automated activity recognition and tracking of construction equipment and workers. This paper provides a comprehensive review of these methods and techniques as well as describes their advantages, practical value, and limitations. Additionally, a multifaceted comparison between these methods is presented, and potential knowledge gaps and future research directions are discussed.
Work-related musculoskeletal disorders (WMSDs) are the leading cause of the nonfatal injuries for construction workers, and a worker’s overexertion is a major source of such WMSDs. Pushing, pulling, and carrying movements—which are all activities largely associated with physical loads—account for 35% of WMSDs. However, most previous studies have focused on the identification of non-ergonomic postures, and there has been limited effort expended on measuring a worker’s exposures to the physical loads caused by materials or tools during construction tasks. With the advantage of using a wearable inertial measurement sensor to monitor a worker’s bodily movements, this study investigates the feasibility of identifying various physical loading conditions by analyzing a worker’s lower body movements. In the experiment with laboratory settings, workers performed a load carrying task by moving concrete bricks. A bidirectional long short-term memory algorithm is employed to classify physical load levels; this approach achieved 74.6 to 98.6% accuracy and 0.59 to 0.99 F-score in classification. The results demonstrate the feasibility of the proposed approach in identifying the states of physical loads. The findings of this study contribute to the literature on classifying ergonomically at-risk workers and on preventing WMSDs in high physical demand occupations, thereby helping enhance the health and safety of the construction workplace.
The leading causes of construction fatalities include traumatic brain injuries (resulted from fall and electrocution) and collisions (resulted from struck by objects). As a preventive step, the U.S. Occupational Safety and Health Administration (OSHA) requires that contractors enforce and monitor appropriate usage of personal protective equipment (PPE) of workers (e.g., hard hat and vest) at all times. This paper presents three deep learning (DL) models built on You-Only-Look-Once (YOLO) architecture to verify PPE compliance of workers; i.e., if a worker is wearing hard hat, vest, or both, from image/video in real-time. In the first approach, the algorithm detects workers, hats, and vests and then, a machine learning model (e.g., neural network and decision tree) verifies if each detected worker is properly wearing hat or vest. In the second approach, the algorithm simultaneously detects individual workers and verifies PPE compliance with a single convolutional neural network (CNN) framework. In the third approach, the algorithm first detects only the workers in the input image which are then cropped and classified by CNN-based classifiers (i.e., VGG-16, ResNet-50, and Xception) according to the presence of PPE attire. All models are trained on an in-house image dataset that is created using crowd-sourcing and web-mining. The dataset, named Pictor-v3, contains ~1,500 annotated images and ~4,700 instances of workers wearing various combinations of PPE components. It is found that the second approach achieves the best performance, i.e., 72.3% mean average precision (mAP), in real-world settings, and can process 11 frames per second (FPS) on a laptop computer which makes it suitable for real-time detection, as well as a good candidate for running on light-weight mobile devices. The closest alternative in terms of performance (67.93% mAP) is the third approach where VGG-16, ResNet-50, and Xception classifiers are assembled in a Bayesian framework. However, the first approach is the fastest among all and can process 13 FPS with 63.1% mAP. The crowed-sourced Pictor-v3 dataset and all trained models are publicly available to support the design and testing of other innovative applications for monitoring safety compliance, and advancing future research in automation in construction.
Although heavy equipment is an indispensable resource in many construction projects, it is often underutilized. Inefficient usage patterns and frequent idling contribute to increased emissions and project costs. Efforts to improve usage patterns often begin with activity tracking. Recent research into automated activity tracking has leveraged sensing devices and Internet-of-Things (IoT) frameworks to power machine learning models that can predict the behaviors of monitored equipment. However, shallow machine learning models require complex manual feature engineering that could be further automated with more recent deep learning approaches. Deep learning approaches not only increase automation but also promise improved accuracies by avoiding biases introduced by manual feature design. This paper proposes a construction equipment activity recognition framework that uses deep learning architectures to predict the activities of heavy construction equipment monitored via accelerometers and applies this framework to a roller compactor and an excavator performing real work. The performance of a simple baseline convolutional neural network (CNN) is compared to a hybrid network that contains both convolutional and recurrent long short-term memory (LSTM) layers. The hybrid model outperforms the baseline model in all instances studied. In the task of classifying the activities of the roller compactor, the hybrid model achieves a validation accuracy of 77.1% when presented with six activities and a validation accuracy of 96.2% when distinguishing only direction. In the task of classifying seven activities of the excavator, the hybrid model achieves a validation accuracy of 77.6%, with some confusion between isolated activities and a Various category that includes elements of the isolated activities. With the Various category removed, the hybrid model achieves a validation accuracy of 90.7%. This study demonstrates that deep learning frameworks can model the activities of construction equipment with high accuracy. In particular, this work shows that convolutional and LSTM layers can each form effective parts of deep learning models that characterize equipment activities based on accelerometer data, and furthermore that these components can produce more effective models when combined. The findings of this study can be leveraged by researchers and industry professionals to develop reliable automated activity recognition systems for tracking and monitoring equipment performance and for measuring the productivity and the efficiency of the work performed.
Hardhats play an essential role in protecting construction individuals from accidents. However, wearing hardhats is not strictly enforced among workers due to all kinds of reasons. To enhance construction sites safety, the majority of existing works monitor the presence and proper use of hardhats through multi-stage data processing, which come with limitations on adaption and generalizability. In this paper, a one-stage system based on convolutional neural network is proposed to automatically monitor whether construction personnel are wearing hardhats and identify the corresponding colors. To facilitate the study, this work constructs a new and publicly available hardhat wearing detection benchmark dataset, which consists of 3174 images covering various on-site conditions. Then, features from different layers with different scales are fused discriminately by the proposed reverse progressive attention to generate a new feature pyramid, which will be fed into the Single Shot Multibox Detector (SSD) to predict the final detection results. The proposed system is trained by an end-to-end scheme. The experimental results demonstrate that the proposed system is effective under all kinds of on-site conditions, which can achieve 83.89% mAP (mean average precision) with the input size 512 × 512.
The construction industry around the globe has unsatisfactory occupational health and safety records. One of the major reasons is attributed to high physical demands and hostile working environments. Construction work always requires workers to work for a long duration without sufficient breaks to recover from overexertion and to work under harsh climatic conditions and/or in confined workspaces. Such circumstances can increase the risk of physical fatigue. Traditionally, fatigue monitoring in the construction domain relies on self-reporting or subjective questionnaires. These methods require the manual collection of responses and are impractical for continuous fatigue monitoring. Some researchers have used on-body sensors for fatigue monitoring (such as heart rate monitors and surface electromyography (sEMG) sensors). Although these devices appear to be promising, they are intrusive, requiring sensors to be attached to the worker's body. Such on-body sensors are uncomfortable to wear and could easily cause irritation. Considering the limitations of these methodologies, the current research proposes a novel non-intrusive method to monitor the whole-body physical fatigue with computer vision for construction workers. A computer vision-based 3D motion capture algorithm was developed to model the motion of various body parts using an RGB camera. A fatigue assessment model was developed using the 3D model data from the developed motion capture algorithm and biomechanical analysis. The experiment showed that the proposed physical fatigue assessment method could provide joint-level physical fatigue assessments automatically. Then, a series of experiments demonstrated the potential of the method in assessing the physical fatigue level of different construction task conditions such as site layout and the work-rest schedules.
Automated, real-time, and reliable equipment activity recognition on construction sites can help to minimize idle time, improve operational efficiency, and reduce emissions. Previous efforts in activity recognition of construction equipment have explored different classification algorithms anm accelerometers and gyroscopes. These studies utilized pattern recognition approaches such as statistical models (e.g., hidden-Markov models); shallow neural networks (e.g., Artificial Neural Networks); and distance algorithms (e.g., K-nearest neighbor) to classify the time-series data collected from sensors mounted on the equipment. Such methods necessitate the segmen-tation of continuous operational data with fixed or dynamic windows to extract statistical features. This heuristic and manual feature extraction process is limited by human knowledge and can only extract human-specified shallow features. However, recent developments in deep neural networks, specifically recurrent neural network (RNN), presents new opportunities to classify sequential time-series data with recurrent lateral connections. RNN can automatically learn high-level representative features through the network instead of being manually designed , making it more suitable for complex activity recognition. However, the application of RNN requires a large training dataset which poses a practical challenge to obtain from real construction sites. Thus, this study presents a data-augmentation framework for generating synthetic time-series training data for an RNN-based deep learning network to accurately and reliably recognize equipment activities. The proposed methodology is validated by generating synthetic data from sample datasets, that were collected from two earthmoving operations in the real world. The synthetic data along with the collected data were used to train a long short-term memory (LSTM)-based RNN. The trained model was evaluated by comparing its performance with traditionally used classification algorithms for construction equipment activity recognition. The deep learning framework presented in this study outperformed the traditionally used machine learning classification algorithms for activity recognition regarding model accuracy and generalization.
The rapid development of the construction industry in China has introduced unprecedented quality-related problems in the country’s building industry. In response to this issue, the government has established various complaint channels to report quality problems. Therefore, building quality complaints (BQCs) need to be classified and solved by respective agencies or departments rapidly for avoiding adverse impact on the safety, health, and well-being of people. However, the current process of classifying BQCs is labor intensive, time consuming, and error prone. An automatic complaint classification is required to improve the effectiveness and efficiency of complaint handling, but studies on this issue are limited. Prevailing text classification research in construction has focused on utilizing conventional shallow machine learning. By contrast, this study explores a novel convolutional neural network (CNN)- based approach that incorporates a deep-learning method to automatically classify the short texts contained within BQCs. The presented approach enables capturing the semantic features in BQC texts and automatic classification of the BQCs into predefined categories. After the model optimization, tests are conducted to examine the practical application of the text classification approach compared with Bayes-based and support vector machine classifiers. Results indicate that the developed CNN-based approach performs well in the Chinese BQC classification with limited manual intervention and few complicated feature engineering.
With tunnel boring machines (TBMs) widely used in tunnel construction, the adaptable adjustment of TBM operating status has become a research focus. Since the prediction of tunnel geological conditions is still challenging before excavation, the prediction of important TBM operating parameters plays an important role in the research on TBM adaptable adjustment. In this paper, we use three kinds of recurrent neural networks (RNNs), including traditional RNNs, long-short term memory (LSTM) networks and gated recurrent unit (GRU) networks, to deal with the real-time prediction of TBM operating parameters based on TBM in-situ operating data. The experimental results show that the proposed three kinds of RNN-based predictors can provide accurate prediction values of some important TBM operating parameters during next period including the torque, the velocity , the thrust and the chamber pressure. We also make a comparison with several classical regression models (e.g., support vector regression (SVR), random forest (RF) and Lasso) which actually cannot act as real-time predictors in a real sense, and the comparative experiments show that the proposed RNN-based predictors outperform the regression models in most cases. The feasibility of RNNs for the real-time prediction of TBM operating parameters indicates that RNNs can afford the analysis and the forecasting of the time-continuous in-situ data collected from various construction equipments.