Zhe Sun , Junbo Li , Ioannis Brilakis , Svetlana Besklubova , Bin Liang , Zhansheng Liu
{"title":"Visual-semantic alignment for automatic structural defect detection and diagnosis of prestressed concrete bridges","authors":"Zhe Sun , Junbo Li , Ioannis Brilakis , Svetlana Besklubova , Bin Liang , Zhansheng Liu","doi":"10.1016/j.autcon.2025.106522","DOIUrl":"10.1016/j.autcon.2025.106522","url":null,"abstract":"<div><div>Defect detection and diagnosis are vital for ensuring safe operations of in-service bridges. However, detecting diverse defects from limited, low-quality image datasets remains challenging. Besides, interpreting identified bridge defects into knowledge for diagnosing bridge health conditions is also difficult. This paper develops a visual-semantic alignment tool for automatic bridge defect detection and diagnosis through computer vision and semantic analysis. The proposed visual-semantic alignment tool contains 1) an enhanced YOLOv10-based detection model for capturing bridge defects; 2) a semantic extraction model for extracting defect information from bridge inspection reports; and 3) a Graph Neural Network (GNN)-based diagnosis model for reasoning structural health conditions. Results show that the developed method achieves 91.7 % precision in defect detection and 81.3 % precision in defect diagnosis. Results indicate that aligning visual with semantic information could support effective bridge defect detection and diagnosis. Future research will focus on advancing computational efficiency to support in-situ bridge inspections.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106522"},"PeriodicalIF":11.5,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145027273","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}
Yi Da , Yangming Gao , Yuanyuan Li , Dan Ren , Kai Liu , Ana Bras , Andy Shaw
{"title":"Advances in smart technologies and materials for automated asphalt pavement inspection: Toward transport infrastructure digitalisation","authors":"Yi Da , Yangming Gao , Yuanyuan Li , Dan Ren , Kai Liu , Ana Bras , Andy Shaw","doi":"10.1016/j.autcon.2025.106523","DOIUrl":"10.1016/j.autcon.2025.106523","url":null,"abstract":"<div><div>The digitalisation of transport infrastructure helps extend pavement service life by enabling timely maintenance based on real-time data from automated inspection. This paper aims to review recent advancements in smart technologies and materials for the structural health monitoring (SHM) of asphalt pavements. Smart monitoring technologies are discussed by analysing their capability in the real-time automated inspection of early-stage pavement internal distress. Furthermore, smart pavement materials, particularly self-sensing asphalt materials, are reviewed in terms of their functionalities, fabrication and electrical characterisation. Finally, applications and challenges of self-sensing asphalt pavements are evaluated, including their implementation, engineering performance, and life cycle assessment. It is concluded that self-sensing asphalt materials provide an effective solution for real-time automated inspection of the early-stage internal distress in pavements. Artificial intelligence (AI) can facilitate the practical implementation of self-sensing asphalt pavement systems integrated with smart materials, management information systems, and intelligent control systems.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106523"},"PeriodicalIF":11.5,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145046633","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}
Yingwen Yu , Edward Verbree , Peter van Oosterom , Uta Pottgiesser , Yuyang Peng , Florent Poux
{"title":"From comparison to integration: A workflow evaluation of 3D Gaussian splatting and LiDAR point cloud for modern architectural heritage","authors":"Yingwen Yu , Edward Verbree , Peter van Oosterom , Uta Pottgiesser , Yuyang Peng , Florent Poux","doi":"10.1016/j.autcon.2025.106509","DOIUrl":"10.1016/j.autcon.2025.106509","url":null,"abstract":"<div><div>This paper investigates the role of 3D Gaussian Splatting (3DGS) within point cloud–dominated workflows for modern architectural heritage digitization. While 3DGS enables real-time, photorealistic visualization, its integration into LiDAR-based documentation pipelines remains underexplored. Using Bouwpub, a modern heritage building in the Netherlands, as a case study, the paper compares 3DGS and LiDAR across data acquisition and preservation, visualization, semantic segmentation, and dissemination. Results show that 3DGS offers superior visual expressiveness and user responsiveness, whereas LiDAR provides greater structural accuracy and segmentation reliability. Based on these findings, two integration strategies are proposed: a Blender-based multi-angle rendering workflow and a Level of Detail 3DGS (LOD3DGS) pipeline. Moving from isolated assessment to applied integration, the study positions 3DGS as a complementary visualization and dissemination module rather than a replacement. This hybrid approach supports immersive, scalable, and semantically enriched digital heritage systems, offering new directions for enhancing both expert documentation and public engagement.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106509"},"PeriodicalIF":11.5,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020196","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":"Generative AI approaches for architectural design automation","authors":"Adeer Khan , Seongju Chang , Hojong Chang","doi":"10.1016/j.autcon.2025.106506","DOIUrl":"10.1016/j.autcon.2025.106506","url":null,"abstract":"<div><div>This review examines the potential and challenges of Generative Artificial Intelligence (AI) in automated building design within architectural practice. A comprehensive analysis of advanced generative models is conducted to evaluate their performance across eight architectural criteria. The qualitative assessment indicates that hybrid approaches combining diffusion models with autoregressive techniques provide the most promising outcomes for architectural applications. Despite advancements, significant challenges remain, including scalability limitations, fragmented workflow integration, and the lack of standardized evaluation frameworks. Potential solutions are identified through interdisciplinary collaboration and strategic research directions, such as developing unified evaluation metrics, enhancing model adaptability, integrating energy-optimized design generation for sustainability, and incorporating designer input in AI-driven workflows. This review provides a structured evaluation of current generative design approaches while proposing a roadmap for future research that bridges the gap between AI innovation and practical architectural implementation, ultimately advancing the field toward more efficient, creative, and sustainable building design automation.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106506"},"PeriodicalIF":11.5,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020193","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}
Taegeon Kim , Seokhwan Kim , Wei-Chih Chern , Somin Park , Daeho Kim , Hongjo Kim
{"title":"Optimizing large vision-language models for context-aware construction safety assessment","authors":"Taegeon Kim , Seokhwan Kim , Wei-Chih Chern , Somin Park , Daeho Kim , Hongjo Kim","doi":"10.1016/j.autcon.2025.106510","DOIUrl":"10.1016/j.autcon.2025.106510","url":null,"abstract":"<div><div>This paper presents a context-aware large vision-language model (LVLM) for automated construction site safety assessment, addressing the limitations of existing models in domain-specific hazard recognition. It introduces a framework that combines domain-specific image-text data generation, vision encoder fine-tuning for improved object recognition, and Low-Rank Adaptation (LoRA)-based model adjustment for context-aware safety reasoning. The model was evaluated on 400 images from 10 hazardous situations, demonstrating superior performance in the image captioning task (average ROUGE-L: 0.3852, SPICE: 0.3615, SBERT-based similarity: 0.7484). For safety assessment, the fine-tuned model achieved 94.25 % accuracy in predicting safety status, significantly outperforming GPT-4 V (53.25 %) and LLaVA 1.5 (48 %). The quality of textual justifications was assessed using both GPT-4 V-based and expert-based evaluations of relevance and preference. In both settings, the fine-tuned model received the highest scores, demonstrating robust and context-aware safety reasoning. These findings confirm that domain-specific fine-tuning enhances safety classification and hazard interpretation, advancing construction site monitoring.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106510"},"PeriodicalIF":11.5,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020191","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}
Lichao Nie, Zhenhua Pan, Shixun Jia, Zhi-Qiang Li, Yue Xiao
{"title":"Real-time detection of water-bearing structures in tunneling using the long-electrode while-drilling resistivity method","authors":"Lichao Nie, Zhenhua Pan, Shixun Jia, Zhi-Qiang Li, Yue Xiao","doi":"10.1016/j.autcon.2025.106528","DOIUrl":"10.1016/j.autcon.2025.106528","url":null,"abstract":"<div><div>In tunnel engineering, water-related hazards such as water inrush and mud pose severe safety risks, especially in water-rich or karst environments. Traditional resistivity methods have limitations in real-time data acquisition and construction coordination. This paper proposes a long electrode source while-drilling (LEWD) resistivity method. Utilizing the drill rod as a current electrode and synchronously acquiring observation data from the tunnel face electrode array, LEWD enables continuous, automated monitoring without interrupting excavation. A 3D finite element forward model is optimized using Cholesky decomposition to improve computational efficiency, while a time-windowed iterative inversion strategy enhances the stability and accuracy of resistivity imaging. Based on this framework, numerical simulations verify the method's ability to accurately identify low-resistivity anomalies under various conditions. Furthermore, field tests in the Xianglu Mountain tunnel in China demonstrate that the LEWD method can effectively detect water-bearing structures ahead of the tunnel face. This method provides effective technical support for water hazard prevention in tunnel construction.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106528"},"PeriodicalIF":11.5,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020192","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}
Proboste Martinez Mathias , Mora Serrano Javier , Muñoz La Rivera Felipe
{"title":"Virtual worlds in AECO operations: Towards a human-centric framework","authors":"Proboste Martinez Mathias , Mora Serrano Javier , Muñoz La Rivera Felipe","doi":"10.1016/j.autcon.2025.106529","DOIUrl":"10.1016/j.autcon.2025.106529","url":null,"abstract":"<div><div>The AECO industry faces persistent challenges, including fragmentation, inefficiencies, and limited human-centric integration in digital workflows. While BIM and Digital Twins (DTs) have advanced digitalization, they often lack the immersive, multi-user, and interactive environments crucial for Industry 5.0 and Construction 5.0. This paper presents a comprehensive review of Virtual Worlds (VWs) and their potential for the AECO sector. Based on this review, a human-centric methodological and technological framework is proposed, structured across four synergistic layers: Data & Context; Processing & Simulation; Interaction & Visualization; and Integration & Connectivity. This framework integrates Extended Reality (XR), Artificial Intelligence (AI), and Gaming Technology with existing BIM and DTs data. Its applicability is conceptually illustrated through five research prototypes, demonstrating its potential to guide advancements in collaboration, simulation, visualization, and human-machine interaction in AECO. This work contributes a structured vision and roadmap for fostering more efficient, collaborative, and human-augmented digital practices.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106529"},"PeriodicalIF":11.5,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145020195","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}
Dong Zhao , Hongyu Hu , Cornelia Asiedu-Kwakyewaa , Noah Durst , Janice Beecher , Linlang He , Lei Shu
{"title":"Electric vehicle charging infrastructure design: Expertise, methods, and challenges","authors":"Dong Zhao , Hongyu Hu , Cornelia Asiedu-Kwakyewaa , Noah Durst , Janice Beecher , Linlang He , Lei Shu","doi":"10.1016/j.autcon.2025.106521","DOIUrl":"10.1016/j.autcon.2025.106521","url":null,"abstract":"<div><div>This review synthesizes the research landscape on electric vehicle charging infrastructure design to identify key methods, challenges, and future directions relevant to the architecture, engineering, and construction (AEC) industry. The analysis reveals that research is dominated by two problems: station location planning and charging technology innovation, which are primarily framed using Optimization Modeling and driven by traffic and geospatial data. A key contribution of this review is the clarification of the methodological hierarchy, from these high-level formulation approaches down to the specific computational techniques (e.g., Mixed-Integer Programming) and executable algorithms (e.g., Genetic Algorithm) used to find solutions. Critically, the review identifies the need for a strategic shift from reactively accommodating demand to proactive planning that uses infrastructure to guide market growth. Future work must resolve the core engineering trade-off between network capacity expansion and technological efficiency and integrate social and regional equity as a formal constraint in system design.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106521"},"PeriodicalIF":11.5,"publicationDate":"2025-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145007512","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}
Wei Wei , Yujie Lu , Ruihan Bai , Lijian Zhong , Yufan Chen , Yijun Lin , Jack C.P. Cheng
{"title":"Enhanced MEP construction progress tracking using panoramic mobile positioning and optimized pipeline segmentation","authors":"Wei Wei , Yujie Lu , Ruihan Bai , Lijian Zhong , Yufan Chen , Yijun Lin , Jack C.P. Cheng","doi":"10.1016/j.autcon.2025.106487","DOIUrl":"10.1016/j.autcon.2025.106487","url":null,"abstract":"<div><div>Efficient progress management is pivotal for the successful delivery of MEP construction projects. While non-intrusive methods, such as image recognition, hold promise in enhancing progress management efficiency, several challenges in MEP scenarios, particularly irregular and sparse features, may constrain the progress recognition accuracy. This paper proposes a framework for automated MEP construction progress tracking, which integrates panoramic mobile positioning, instance segmentation, and Unreal Engine to compare virtual (as-planned) and real (as-built) MEP construction scenes. Moreover, quantitative progress can be estimated by identifying and mapping unfinished construction components. Remarkably, dynamic snake convolution is introduced to enhance the component segmentation accuracy, specifically for irregular and sparse features. The framework is successfully applied to an MEP construction project in China, achieving commendable component segmentation (mAP<sub>50</sub> = 85 %) and progress recognition accuracy. This paper provides theoretical references for vision-based MEP construction progress tracking, offering practical insights for intelligent construction inspection in continuous spaces.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106487"},"PeriodicalIF":11.5,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004675","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}
Tan Li , Hong Wang , Jiasheng Tan , Lingjie Kong , Haoran Zhang , Dongxu Pan , Zhihao Zhao
{"title":"Intelligent quality assessment of concrete vibration using computer vision and large language models","authors":"Tan Li , Hong Wang , Jiasheng Tan , Lingjie Kong , Haoran Zhang , Dongxu Pan , Zhihao Zhao","doi":"10.1016/j.autcon.2025.106507","DOIUrl":"10.1016/j.autcon.2025.106507","url":null,"abstract":"<div><div>The monitoring of concrete vibration quality is crucial for ensuring construction quality. This paper proposes a monitoring method that combines computer vision and Large Language Model (LLM). First, an unsupervised shadow removal method is used to optimize image quality. Next, a multi-head classification model is applied to conduct a multi-dimensional comprehensive assessment of vibration quality. After that, the classification results are mapped to natural language information through a key-value image-to-text mapping method. Finally, the natural language is used for inference in the LLM to generate real-time feedback. Experimental results show that the proposed method achieves an accuracy of 94.45 % in classifying the vibration quality. Additionally, by combining image classification results with LLM for logical reasoning and feedback generation, the system can provide detailed descriptions of compaction quality and corresponding solutions. This research has been successfully applied in real-world projects and is expected to promote the intelligent development of construction operations.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"180 ","pages":"Article 106507"},"PeriodicalIF":11.5,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004676","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}