Wei-Lun Tsai , Phuong-Linh Le , Wang-Fat Ho , Nai-Wen Chi , Jacob J. Lin , Shuai Tang , Shang-Hsien Hsieh
{"title":"Construction safety inspection with contrastive language-image pre-training (CLIP) image captioning and attention","authors":"Wei-Lun Tsai , Phuong-Linh Le , Wang-Fat Ho , Nai-Wen Chi , Jacob J. Lin , Shuai Tang , Shang-Hsien Hsieh","doi":"10.1016/j.autcon.2024.105863","DOIUrl":"10.1016/j.autcon.2024.105863","url":null,"abstract":"<div><div>Traditional safety inspections require significant human effort and time to capture site photos and textual descriptions. While standardized forms and image captioning techniques have been explored to improve inspection efficiency, compiling reports with both visual and text data remains challenging due to the multiplicity of safety-related knowledge. To assist inspectors in evaluating violations more efficiently, this paper presents an image-language model, utilizing Contrastive Language-Image Pre-training (CLIP) fine-tuning and prefix captioning to automatically generate safety observations. A user-friendly mobile phone application has been created to streamline safety report documentation for site engineers. The language model successfully classifies nine violation types with an average accuracy of 73.7%, outperforming the baseline model by 41.8%. Experiment participants confirmed that the mobile application is helpful for safety inspections. This automated framework simplifies safety documentation by identifying violation scenes through images, improves overall safety performance, and supports the digital transformation of construction sites.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"169 ","pages":"Article 105863"},"PeriodicalIF":9.6,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Signs on glasses: LiDAR data voids, hotspot effect, and reflection artifacts","authors":"Tung Sum Fong , Wai Yeung Yan","doi":"10.1016/j.autcon.2024.105877","DOIUrl":"10.1016/j.autcon.2024.105877","url":null,"abstract":"<div><div>A key challenge in terrestrial laser scanning (TLS) based as-built survey is the presence of data voids, reflection artifacts, and hotspot effect on glasses. This paper investigates the effects of scanning range, illumination condition, instrument height, and spatial offset of an occluded object between the scanner and glasses with respect to these data artifacts. Experimental results show that ordinary float glass encounters a high percentage of data voids (<span><math><mo>></mo></math></span> 93%) and reflective glass backscatters laser pulses with greater than 50%. Hotspot effect is found notable at azimuth scanning angle between <span><math><mo>±</mo></math></span> 2 and 5° on both types of glasses. Ordinary float glass never results in any reflection artifacts, while they likely emerge when the scanner is set up at a certain height (<span><math><mo>></mo></math></span> 0.7 m) and a closer range (<span><math><mo>≤</mo></math></span> 7.5 m) on the reflective glass. These findings shed light on the future algorithmic development of glass detection, classification, and defects removal.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"169 ","pages":"Article 105877"},"PeriodicalIF":9.6,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684174","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yuan Cao , Shifan Li , Geoffrey Qiping Shen , Hongyu Chen , Yang Liu
{"title":"Intelligent dynamic control of shield parameters using a hybrid algorithm and digital twin platform","authors":"Yuan Cao , Shifan Li , Geoffrey Qiping Shen , Hongyu Chen , Yang Liu","doi":"10.1016/j.autcon.2024.105882","DOIUrl":"10.1016/j.autcon.2024.105882","url":null,"abstract":"<div><div>This paper presents a digital twin (DT) platform integrated with an online optimization algorithm that combines Bayesian Optimization (BO), Categorical Boosting (CatBoost), and the Nondominated Sorting Genetic Algorithm (NSGA)-III. The platform enables multi-objective dynamic optimization of shield parameters under varying geological conditions. Using the Wuhan Metro as a case study, the effectiveness of the method is validated. The results demonstrate that: (1) the DT model accurately estimates shield machine performance, with an R<sup>2</sup> of no less than 0.957 on the test set across three geological conditions; (2) the online optimization significantly enhances shield machine performance, with a comprehensive optimization improvement of over 25 % across all conditions; (3) comparison of the constructed algorithm's accuracy, along with Shapley additive explanations, confirms the accuracy, interpretability, and universality of the proposed algorithm.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"169 ","pages":"Article 105882"},"PeriodicalIF":9.6,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142698961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bo Xiao , Yifan Wang , Yongpan Zhang , Chen Chen , Amos Darko
{"title":"Automated daily report generation from construction videos using ChatGPT and computer vision","authors":"Bo Xiao , Yifan Wang , Yongpan Zhang , Chen Chen , Amos Darko","doi":"10.1016/j.autcon.2024.105874","DOIUrl":"10.1016/j.autcon.2024.105874","url":null,"abstract":"<div><div>Daily reports are important in construction management, informing project teams about status, enabling timely resolutions of delays and budget issues, and serving as official records for disputes and litigation. However, current practices are manual and time-consuming, requiring engineers to physically visit sites for observations. To fill this gap, this paper proposes an automated framework to generate daily construction reports from on-site videos by integrating ChatGPT and computer vision (CV)-based methods. The framework utilizes CV methods to analyze video footage and extract relevant productivity and activity information, which is then fed into ChatGPT using proper prompts to generate daily reports. A web application is developed to implement and validate the framework on a real construction site in Hong Kong, generating daily reports over a month. This research enhances construction management by significantly reducing documentation efforts through generative artificial intelligence, with potential applications in jobsite safety management, quality reporting, and stakeholder communication.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105874"},"PeriodicalIF":9.6,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Automated physics-based modeling of construction equipment through data fusion","authors":"Liqun Xu , Dharmaraj Veeramani , Zhenhua Zhu","doi":"10.1016/j.autcon.2024.105880","DOIUrl":"10.1016/j.autcon.2024.105880","url":null,"abstract":"<div><div>Physics-based simulations are essential for designing autonomous construction equipment, but preparing models is time-consuming, requiring the integration of mechanical and geometric data. Current automatic modeling methods for modular robots are inadequate for construction equipment. This paper explores automating the modeling process by integrating mechanical data into 3D computer-aided design (CAD) models. A template library is developed with hierarchy and joint templates specific for equipment. During model generation, appropriate templates are selected based on the equipment type. Unspecified joint template data is extracted from technical specifications using a large language model (LLM). The 3D CAD model is then converted into a Universal Scene Description (USD) model. Users can adjust the part names and hierarchy within the USD model to align with the hierarchy template, and joint data is automatically integrated, resulting in a simulation-ready model. This method reduces modeling time by over 87 % compared to manual methods, while maintaining accuracy.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105880"},"PeriodicalIF":9.6,"publicationDate":"2024-11-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Automated rule-based safety inspection and compliance checking of temporary guardrail systems in construction","authors":"K.W. Johansen , J. Teizer , C. Schultz","doi":"10.1016/j.autcon.2024.105849","DOIUrl":"10.1016/j.autcon.2024.105849","url":null,"abstract":"<div><div>The construction industry records more hazards compared to any other sector. Protective equipment, such as guardrail systems, is essential for protecting workers from deadly falls but may quickly become incompliant after installation. Yet, many construction projects do not have the resources to dedicate personnel to perform the inspection as frequently as needed. Therefore, this paper proposes an automated rule-based inspection and compliance-checking system that can assist the responsible personnel in detecting faulty guardrails in live work environments. The classification approach utilizes safety design and mimics the steps of human guardrailing compliance assessment, which enforces simplicity and transparency, allowing the human domain expert to remain in control. Even under scarce data availability, this first-of-a-kind classification approach is reliable and scalable and successfully classifies 21 predefined and 9 validation scenarios of guardrail systems for fall protection.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105849"},"PeriodicalIF":9.6,"publicationDate":"2024-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"3D Pixelwise damage mapping using a deep attention based modified Nerfacto","authors":"Geontae Kim, Youngjin Cha","doi":"10.1016/j.autcon.2024.105878","DOIUrl":"10.1016/j.autcon.2024.105878","url":null,"abstract":"<div><div>Recent advancements in structural health monitoring have highlighted the necessity for accurate three-dimensional (3D) damage mapping on digital twins, moving beyond traditional methods such as photogrammetry, which frequently struggle to capture intricate planar surfaces. To address this limitation, this paper proposes a new advanced 3D reconstruction method and its integration with 3D damage mapping techniques. As the 3D reconstruction method, an Attention-based Modified Nerfacto (ABM-Nerfacto) model is developed, and is integrated with an advanced damage segmentation method. Using a three-span continuous bridge with concrete piers as an example structure, and concrete cracks as the example damage, the state-of-the-art STRNet is utilized for crack segmentation. Through extensive parametric studies and comparative evaluations, the proposed ABM-Nerfacto model was demonstrated to produce high-quality 3D reconstructions and corresponding damage mappings for this bridge system. This integrated approach provides a promising solution for comprehensive 3D digital twin-based structural health monitoring.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105878"},"PeriodicalIF":9.6,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684188","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sehwan Chung , Jungyeon Kim , Joonwoo Baik , Seokho Chi , Du Yon Kim
{"title":"Identifying issues in international construction projects from news text using pre-trained models and clustering","authors":"Sehwan Chung , Jungyeon Kim , Joonwoo Baik , Seokho Chi , Du Yon Kim","doi":"10.1016/j.autcon.2024.105875","DOIUrl":"10.1016/j.autcon.2024.105875","url":null,"abstract":"<div><div>The uncontrollable nature of international construction projects requires continuous monitoring of issues in the host country. News articles can provide relevant information to monitor the issues, but the manual investigation of substantial news text is impractical. This paper proposes a framework to automatically collect information related to the host country's business environments from news text, consisting of three modules: (1) web scraping for collecting news text; (2) text embedding using a pre-trained language model; and (3) text clustering for extracting essential issues. Applying this framework to real-world news demonstrated its proficiency in identifying significant issues, outperforming the existing similar methods in terms of: (1) the accuracy of issue identification; (2) the quality of identified issues; and (3) the degree of human intervention. This paper contributes to the body of knowledge by showcasing the utility of news text in gathering information and identifying issues about the host country during international projects.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105875"},"PeriodicalIF":9.6,"publicationDate":"2024-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142684229","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Anja Kunic, Davide Angeletti, Giuseppe Marrone, Roberto Naboni
{"title":"Design and construction automation of reconfigurable timber slabs","authors":"Anja Kunic, Davide Angeletti, Giuseppe Marrone, Roberto Naboni","doi":"10.1016/j.autcon.2024.105872","DOIUrl":"10.1016/j.autcon.2024.105872","url":null,"abstract":"<div><div>Structural adaptivity and readiness for change are some of the key enablers of resilient and sustainable architecture. This paper presents an approach to the design and construction of reconfigurable timber slabs, termed ReconWood Slabs, which integrate a stress-driven design approach and cyber-physical construction processes to enhance data-informed circularity. Using advanced computational design tools, the research outlines a workflow for generating optimised slab configurations that balance structural performance with material reusability. The slabs, composed of modular beams connected via reversible steel fasteners, are designed for easy disassembly and reconfiguration, promoting material reuse across multiple building lifecycles. The paper demonstrates the system's potential through the construction of three slab structures employing Mixed Reality for material data tracking and assembly. The three structures store nearly a ton of CO2eq in reusable parts.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105872"},"PeriodicalIF":9.6,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142652728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yun Chen , Gengyang Lu , Ke Wang , Shu Chen , Chenfei Duan
{"title":"Knowledge graph for safety management standards of water conservancy construction engineering","authors":"Yun Chen , Gengyang Lu , Ke Wang , Shu Chen , Chenfei Duan","doi":"10.1016/j.autcon.2024.105873","DOIUrl":"10.1016/j.autcon.2024.105873","url":null,"abstract":"<div><div>With the increasing demand for water conservancy engineering (WCE), the number of safety accidents during construction has continued to rise, requiring an urgent improvement in construction safety. The existing safety management regulations for water conservancy construction engineering (WCCE) comprise a considerable amount of text, with cross-references between different standards severely reducing their use efficiency. To address this issue, this paper proposes an ALBERT-BiLSTM-CRF model based on textual data from WCCE safety management standards. ALBERT, a lightweight pretrained language model, is integrated with the BiLSTM-CRF to construct an intelligent text entity recognition method. Association rules are used to extract entity relationships, and a knowledge graph representing the WCCE safety management standards is established. The results show that the ALBERT-BiLSTM-CRF algorithm improves the precision, with a recognition accuracy exceeding 85 %. Case studies validate that the constructed knowledge graph can quickly query safety standard knowledge, aiding in the generation of safety measures.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"168 ","pages":"Article 105873"},"PeriodicalIF":9.6,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142652729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}