Automation in Construction最新文献

英文中文

Multimodal data fusion for welding defect detection using ensemble deep learning 基于集成深度学习的焊接缺陷检测多模态数据融合

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-12-08 DOI: 10.1016/j.autcon.2025.106694

Shiqiang Tang , Feilong Fei , Limao Zhang , Jinfeng Yu

This study proposes a multimodal deep learning model for high-precision automated detection of resistance spot welding defects. A dual-input weight-sharing network is employed to process the surface images of the weld nugget, while infrared images and welding parameter data are processed by two additional base models. The outputs of these base models are fused using Dempster–Shafer theory, yielding the ensemble multimodal deep learning model (EMMDL). Validation on a welding dataset reveals that: (1) EMMDL achieves an accuracy of 91.6 %, significantly outperforming base models with single modality; (2) Dual-input and weight sharing increases classification accuracy by 7.87 % and enhances robustness in small sample scenarios; (3) The model uses more information from infrared images when identifying bad samples. By integrating complementary multimodal information, EMMDL overcomes blind spots of single-source methods and provides interpretable decision support for industrial quality control.

本文提出了一种多模态深度学习模型，用于电阻点焊缺陷的高精度自动检测。采用双输入权值共享网络对焊核表面图像进行处理，红外图像和焊接参数数据由另外两个基本模型进行处理。这些基本模型的输出使用Dempster-Shafer理论进行融合，产生集成多模态深度学习模型（EMMDL）。在焊接数据集上的验证表明：(1)EMMDL的准确率达到91.6%，显著优于单一模态的基础模型；(2)双输入和权值共享使分类准确率提高了7.87%，增强了小样本场景下的鲁棒性；(3)该模型在识别不良样品时更多地利用了红外图像的信息。通过整合互补的多模态信息，EMMDL克服了单源方法的盲点，为工业质量控制提供了可解释的决策支持。

引用次数: 0

Leveraging large language models for BIM-based automated compliance checking 利用大型语言模型进行基于bim的自动遵从性检查

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-12-05 DOI: 10.1016/j.autcon.2025.106707

Odin Iversen, Lizhen Huang

Current methods of checking regulatory compliance in the architecture, engineering, construction, and operations (AECO) industry are mostly manual, time consuming and error prone. This paper, using design science research (DSR), proposes an artifact that leverages a large language model (LLM) for automated compliance checking (ACC) to directly interpret regulations, extract BIM data, execute checks, and generate detailed reports. For rule interpretation, the artifact achieves high F1-scores (97% for classification, 100% for dependency identification). For building model preparation, it correctly selected data extraction tools in 97% of cases. In rule execution, it demonstrated 97,7% accuracy and significantly outperformed a naive baseline, which highlights the need for a structured framework. Finally, the artifact generated detailed reports that included the LLM’s reasoning. The key finding is that an LLM-based reasoning engine enables a holistic approach that overcomes the manual rule digitization bottleneck in traditional ACC systems.

在架构、工程、构造和操作（AECO）行业中，当前检查法规遵从性的方法大多是手动的、耗时且容易出错的。本文利用设计科学研究（DSR），提出了一种工件，该工件利用大型语言模型（LLM）进行自动合规性检查（ACC），直接解释法规，提取BIM数据，执行检查并生成详细报告。对于规则解释，工件达到了很高的f1分数（97%用于分类，100%用于依赖标识）。对于构建模型的准备，它在97%的情况下正确选择了数据提取工具。在规则执行中，它显示了97,7%的准确性，并且显著优于朴素基线，这突出了对结构化框架的需求。最后，工件生成了包含LLM推理的详细报告。关键发现是，基于llm的推理引擎实现了一种整体方法，克服了传统ACC系统中手动规则数字化的瓶颈。

引用次数: 0

Cascaded and upgradable smart contracts for blockchain-aided construction business process management 用于区块链辅助建筑业务流程管理的级联和可升级智能合约

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-12-05 DOI: 10.1016/j.autcon.2025.106695

Xuling Ye , Xingyu Tao , Jack C.P. Cheng , Markus König

Collaboration in Construction Business Process Management (CBPM) often suffers from inefficiency, fragmentation, and security concerns. Blockchain and Smart Contract (SC) offer potential solutions by enabling automation, transparency, and tamper-resistant records. However, adoption remains limited due to two critical gaps: (1) insufficient automation, as current SCs lack cascaded (interdependent) execution, and (2) insufficient adaptability, as existing SCs are non-upgradable, limiting responsiveness to workflow changes. This paper proposes a SC-CBPM framework addressing these gaps through three objectives: (1) Automate CBPM tasks and processes; (2) Develop Cascaded SCs to link interdependent tasks and enforce access control; (3) Develop Upgradable SCs to allow updates without data loss. The framework is validated through two scenarios: BIM-based design collaboration and payment automation, demonstrating feasibility and acceptable computational workability. Performance is evaluated through gas consumption and latency, ensuring deployment readiness. The main contribution is advancing blockchain from a static record-keeping tool to an adaptive workflow automation mechanism.

构建业务流程管理（CBPM）中的协作经常受到效率低下、碎片化和安全性问题的困扰。区块链和智能合约（SC）通过实现自动化、透明度和防篡改记录提供了潜在的解决方案。然而，由于两个关键的差距，采用仍然有限：(1)自动化不足，因为当前的sc缺乏级联（相互依赖）执行；(2)适应性不足，因为现有的sc不可升级，限制了对工作流程变化的响应。本文提出了一个SC-CBPM框架，通过三个目标来解决这些差距：(1)自动化CBPM任务和流程；(2)开发级联sc以连接相互依赖的任务并实施访问控制；(3)开发可升级的sc，允许更新而不丢失数据。该框架通过两个场景进行验证：基于bim的设计协作和支付自动化，展示了可行性和可接受的计算可行性。性能通过气体消耗和延迟来评估，确保部署就绪。其主要贡献是将区块链从静态记录保存工具提升为自适应工作流自动化机制。

{"title":"Cascaded and upgradable smart contracts for blockchain-aided construction business process management","authors":"Xuling Ye , Xingyu Tao , Jack C.P. Cheng , Markus König","doi":"10.1016/j.autcon.2025.106695","DOIUrl":"10.1016/j.autcon.2025.106695","url":null,"abstract":"<div><div>Collaboration in Construction Business Process Management (CBPM) often suffers from inefficiency, fragmentation, and security concerns. Blockchain and Smart Contract (SC) offer potential solutions by enabling automation, transparency, and tamper-resistant records. However, adoption remains limited due to two critical gaps: (1) insufficient automation, as current SCs lack cascaded (interdependent) execution, and (2) insufficient adaptability, as existing SCs are non-upgradable, limiting responsiveness to workflow changes. This paper proposes a SC-CBPM framework addressing these gaps through three objectives: (1) Automate CBPM tasks and processes; (2) Develop Cascaded SCs to link interdependent tasks and enforce access control; (3) Develop Upgradable SCs to allow updates without data loss. The framework is validated through two scenarios: BIM-based design collaboration and payment automation, demonstrating feasibility and acceptable computational workability. Performance is evaluated through gas consumption and latency, ensuring deployment readiness. The main contribution is advancing blockchain from a static record-keeping tool to an adaptive workflow automation mechanism.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"182 ","pages":"Article 106695"},"PeriodicalIF":11.5,"publicationDate":"2025-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684077","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}

引用次数: 0

Cross-domain decision support system for spoil dumpsite selection in mega transportation infrastructure projects 大型交通基础设施项目垃圾堆放场选择的跨域决策支持系统

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-12-03 DOI: 10.1016/j.autcon.2025.106697

Long Li , Jing Yang , Yulong Li , Guobin Wu , Shengxi Zhang

Spoil dumpsite selection (SDS), a representative complex decision problem in mega transportation infrastructure (MTI) projects, is crucial for ensuring the project sustainability. However, conventional single-domain or purely data-driven methods cannot fully address the multi-objective conflicts and cross-domain knowledge heterogeneity in SDS. To bridge this gap, this paper proposes a cross-domain decision support system (CDDS) with three progressive modules: (1) criteria system and alternatives identification, (2) domain division and ontology representation, and (3) two-stage knowledge fusion, together forming a systematic decision process. Case study of a mega railway project in western China demonstrates that CDDS can produce viable and robust results, verifying its effectiveness and applicability. Applying this tool, project stakeholders can enhance the interpretability of their decisions in complex environments. Furthermore, this system expands the theoretical and methodological boundaries of knowledge fusion decision and can guide practical complex system engineering decisions in similar contexts.

垃圾堆放场选择是大型交通基础设施项目中具有代表性的复杂决策问题，是保证项目可持续性的关键问题。然而，传统的单领域或纯数据驱动方法不能完全解决SDS中的多目标冲突和跨领域知识异质性问题。为了弥补这一缺陷，本文提出了一个跨领域决策支持系统（CDDS），该系统由三个渐进模块组成：(1)标准体系和备选方案识别，(2)领域划分和本体表示，(3)两阶段知识融合，共同形成一个系统的决策过程。以中国西部某大型铁路项目为例，验证了CDDS方法的有效性和适用性。应用此工具，项目干系人可以在复杂的环境中增强其决策的可解释性。此外，该系统扩展了知识融合决策的理论和方法边界，可以指导类似环境下的实际复杂系统工程决策。

{"title":"Cross-domain decision support system for spoil dumpsite selection in mega transportation infrastructure projects","authors":"Long Li , Jing Yang , Yulong Li , Guobin Wu , Shengxi Zhang","doi":"10.1016/j.autcon.2025.106697","DOIUrl":"10.1016/j.autcon.2025.106697","url":null,"abstract":"<div><div>Spoil dumpsite selection (SDS), a representative complex decision problem in mega transportation infrastructure (MTI) projects, is crucial for ensuring the project sustainability. However, conventional single-domain or purely data-driven methods cannot fully address the multi-objective conflicts and cross-domain knowledge heterogeneity in SDS. To bridge this gap, this paper proposes a cross-domain decision support system (CDDS) with three progressive modules: (1) criteria system and alternatives identification, (2) domain division and ontology representation, and (3) two-stage knowledge fusion, together forming a systematic decision process. Case study of a mega railway project in western China demonstrates that CDDS can produce viable and robust results, verifying its effectiveness and applicability. Applying this tool, project stakeholders can enhance the interpretability of their decisions in complex environments. Furthermore, this system expands the theoretical and methodological boundaries of knowledge fusion decision and can guide practical complex system engineering decisions in similar contexts.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"182 ","pages":"Article 106697"},"PeriodicalIF":11.5,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145658681","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}

引用次数: 0

Sensor data-driven decision support system for real-time optimization and impact assessment in concrete construction 基于传感器数据驱动的混凝土施工实时优化与影响评估决策支持系统

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-11-29 DOI: 10.1016/j.autcon.2025.106685

David Boix-Cots , Tai Ikumi , Nikola Tošić , Albert de la Fuente

This paper introduces a sensor data-driven decision support system for calculating both economic and environmental impacts of implementing the maturity method. The system integrates data from wireless temperature sensors embedded in concrete with a four-phase workflow and a dedicated Impact Assessment Methodology (IAM). This combination enables construction teams to assess both economic and environmental impacts of early-age concrete behaviour, supporting decisions such as formwork removal timing or concrete mix adjustment. The proposed methodology was applied to a real-world viaduct construction project involving 691 m³ of concrete and 50 wireless sensors. The results demonstrated significant optimization potential compared with the standard method, including cost savings of 48.15 €/m³, 1.59 kg CO₂-eq/m³ of avoided emissions, and a reduction of 0.031 m³ of water per cubic meter of concrete. The system provides a transparent and replicable framework with potential applicability to a wide range of construction contexts, from building projects to large-scale infrastructure works.

本文介绍了一个传感器数据驱动的决策支持系统，用于计算实施成熟度法的经济和环境影响。该系统集成了嵌入混凝土中的无线温度传感器的数据，采用四阶段工作流程和专用的影响评估方法（IAM）。这种组合使施工团队能够评估早期混凝土行为的经济和环境影响，支持诸如拆除模板时间或混凝土混合调整等决策。所提出的方法应用于一个实际的高架桥建设项目，该项目涉及691立方米的混凝土和50个无线传感器。与标准方法相比，结果显示出显著的优化潜力，包括节省48.15欧元/立方米的成本，避免1.59千克二氧化碳当量/立方米的排放，每立方米混凝土减少0.031立方米的水。该系统提供了一个透明和可复制的框架，具有广泛的建筑环境的潜在适用性，从建筑项目到大型基础设施工程。

{"title":"Sensor data-driven decision support system for real-time optimization and impact assessment in concrete construction","authors":"David Boix-Cots , Tai Ikumi , Nikola Tošić , Albert de la Fuente","doi":"10.1016/j.autcon.2025.106685","DOIUrl":"10.1016/j.autcon.2025.106685","url":null,"abstract":"<div><div>This paper introduces a sensor data-driven decision support system for calculating both economic and environmental impacts of implementing the maturity method. The system integrates data from wireless temperature sensors embedded in concrete with a four-phase workflow and a dedicated Impact Assessment Methodology (IAM). This combination enables construction teams to assess both economic and environmental impacts of early-age concrete behaviour, supporting decisions such as formwork removal timing or concrete mix adjustment. The proposed methodology was applied to a real-world viaduct construction project involving 691 m<sup>3</sup> of concrete and 50 wireless sensors. The results demonstrated significant optimization potential compared with the standard method, including cost savings of 48.15 €/m<sup>3</sup>, 1.59 kg CO₂-eq/m<sup>3</sup> of avoided emissions, and a reduction of 0.031 m<sup>3</sup> of water per cubic meter of concrete. The system provides a transparent and replicable framework with potential applicability to a wide range of construction contexts, from building projects to large-scale infrastructure works.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"181 ","pages":"Article 106685"},"PeriodicalIF":11.5,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145613939","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}

引用次数: 0

SLAM-centric visual inspection of civil infrastructure 以slam为中心的民用基础设施目视检查

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-11-29 DOI: 10.1016/j.autcon.2025.106682

Nicholas Charron , Jake McLaughlin , Sriram Narasimhan

Existing robot-aided inspection methods suffer from inconsistent map accuracy, unreliable defect measurements, and platform-specific designs. This paper investigates whether a SLAM-centric framework can enable precise, repeatable, and platform-agnostic visual inspections. The framework integrates lidar–camera–inertial SLAM, offline trajectory refinement, inspection-map generation, defect extraction from imagery, and 3D ray-tracing to project defects into a unified map. The approach confirms that accurate defect localization, dimensional quantification, and dense inspection maps can be produced in real-world scenarios. This finding benefits infrastructure owners and inspectors by providing an end-to-end solution for robot-aided inspections that enable faster, safer, and more objective assessments compared to current qualitative workflows. The released datasets and software establish a foundation for future research on long-term defect monitoring and inspection automation.

现有的机器人辅助检测方法存在地图精度不一致、缺陷测量不可靠以及平台特定设计等问题。本文研究了一个以slam为中心的框架是否能够实现精确的、可重复的和平台无关的视觉检查。该框架集成了激光雷达-相机-惯性SLAM、离线轨迹优化、检查图生成、从图像中提取缺陷以及3D光线跟踪，将缺陷投影到统一的地图中。该方法证实了在真实的场景中可以产生精确的缺陷定位、尺寸量化和密集的检查图。与目前的定性工作流程相比，这一发现为机器人辅助检查提供了端到端解决方案，使基础设施所有者和检查人员受益，从而实现更快、更安全、更客观的评估。发布的数据集和软件为未来长期缺陷监测和检测自动化的研究奠定了基础。

引用次数: 0

Automated safety risk assessment for crane operations using cascade learning 基于级联学习的起重机作业安全风险自动评估

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-11-27 DOI: 10.1016/j.autcon.2025.106686

Mudasir Hussain , Zhongnan Ye , Hung-Lin Chi , Shu-Chien Hsu

Construction machinery enhances productivity and ensures project timelines. However, equipment failure poses significant risks, including injuries, fatalities, and financial losses. Traditional safety assessments rely on manual reporting and are prone to errors, delays, and inconsistencies. This paper introduced a cascade learning technique for automated safety risk assessment in crane operations, ensuring reliable, accurate, and adaptable evaluations. The cascade model detects cranes, classifies safety statuses and activities, and computes risk values using confidence scores and impact factors. A risk threshold of 0.52 triggers real-time alerts for intervention. Video-feed analysis supports continuous monitoring and documentation. Expert validation confirmed the practicality of the risk-quantification models. The model achieved 92.10 % precision in crane detection, 99.25 % accuracy in safety classification, and 99.47 % accuracy in activity classification, with an inference time of 0.70 s. This approach enhances Smart Site Safety System (4S) technologies, automates safety assessments, and contributes to improved construction safety standards.

工程机械提高了生产效率，保证了项目进度。然而，设备故障会带来重大风险，包括伤害、死亡和经济损失。传统的安全评估依赖于手工报告，容易出现错误、延迟和不一致。介绍了一种用于起重机作业安全风险自动评估的级联学习技术，保证了评估的可靠性、准确性和适应性。级联模型检测起重机，对安全状态和活动进行分类，并使用置信度评分和影响因子计算风险值。风险阈值0.52触发实时干预警报。视频馈送分析支持持续监控和记录。专家验证证实了风险量化模型的实用性。该模型在起重机检测、安全分类和活动分类方面的准确率分别达到92.10%、99.25%和99.47%，推理时间为0.70 s。这种方法增强了智能工地安全系统（4S）技术，使安全评估自动化，并有助于提高建筑安全标准。

{"title":"Automated safety risk assessment for crane operations using cascade learning","authors":"Mudasir Hussain , Zhongnan Ye , Hung-Lin Chi , Shu-Chien Hsu","doi":"10.1016/j.autcon.2025.106686","DOIUrl":"10.1016/j.autcon.2025.106686","url":null,"abstract":"<div><div>Construction machinery enhances productivity and ensures project timelines. However, equipment failure poses significant risks, including injuries, fatalities, and financial losses. Traditional safety assessments rely on manual reporting and are prone to errors, delays, and inconsistencies. This paper introduced a cascade learning technique for automated safety risk assessment in crane operations, ensuring reliable, accurate, and adaptable evaluations. The cascade model detects cranes, classifies safety statuses and activities, and computes risk values using confidence scores and impact factors. A risk threshold of 0.52 triggers real-time alerts for intervention. Video-feed analysis supports continuous monitoring and documentation. Expert validation confirmed the practicality of the risk-quantification models. The model achieved 92.10 % precision in crane detection, 99.25 % accuracy in safety classification, and 99.47 % accuracy in activity classification, with an inference time of 0.70 s. This approach enhances Smart Site Safety System (4S) technologies, automates safety assessments, and contributes to improved construction safety standards.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"181 ","pages":"Article 106686"},"PeriodicalIF":11.5,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145611907","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}

引用次数: 0

Generalizing fatigue prediction models for construction workers: Cross-experiment evaluation with transfer learning across thermal and load conditions 建筑工人疲劳预测模型的推广：热负荷条件下迁移学习的交叉实验评估

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-11-25 DOI: 10.1016/j.autcon.2025.106680

Sharjeel Anjum , Muhammad Khan , Chukwuma Nnaji , Ashrant Aryal , Amanda S. Koh

Physical fatigue among construction workers is a major safety concern, impacting both health and productivity. Machine learning (ML) models for fatigue monitoring often struggle with generalizing across varying work conditions and populations. This paper advances fatigue monitoring automation by (1) developing ML models trained under diverse temperature and load conditions (Dataset 1), (2) evaluating generalizability on unseen construction-related data (Dataset 2), and (3) proposing transfer learning-based fine-tuning to enhance models' adaptability while reducing the need for large datasets. Initial accuracies on Dataset 1 were 87.5 % (RFC), 89.7 % (XGBoost), and 92 % (FatigueNet); however, these dropped sharply to 40 % (RFC, XGBoost) and 29 % (FatigueNet) under the generalizability test. When trained from scratch on combined datasets, RFC and FatigueNet achieved 47 % and 60 % accuracy, highlighting challenges with generalization. Transfer learning improved FatigueNet's accuracy to 82 % and RFC's to 87 %. These results demonstrate transfer learning's potential for real-time fatigue monitoring and construction site safety.

建筑工人的身体疲劳是一个主要的安全问题，影响健康和生产力。用于疲劳监测的机器学习（ML）模型通常难以在不同的工作条件和人群中进行泛化。本文通过(1)开发在不同温度和负载条件下训练的ML模型（数据集1），(2)评估未见建筑相关数据（数据集2）的泛化性，以及(3)提出基于迁移学习的微调以增强模型的适应性，同时减少对大型数据集的需求来推进疲劳监测自动化。数据集1的初始准确度为87.5% (RFC), 89.7% （XGBoost）和92% (FatigueNet)；然而，在通用性测试中，这些数据急剧下降到40% （RFC, XGBoost）和29% （FatigueNet）。当在组合数据集上从零开始训练时，RFC和FatigueNet的准确率分别达到47%和60%，这凸显了泛化的挑战。迁移学习将FatigueNet的准确率提高到82%，RFC的准确率提高到87%。这些结果证明了迁移学习在实时疲劳监测和施工现场安全方面的潜力。

{"title":"Generalizing fatigue prediction models for construction workers: Cross-experiment evaluation with transfer learning across thermal and load conditions","authors":"Sharjeel Anjum , Muhammad Khan , Chukwuma Nnaji , Ashrant Aryal , Amanda S. Koh","doi":"10.1016/j.autcon.2025.106680","DOIUrl":"10.1016/j.autcon.2025.106680","url":null,"abstract":"<div><div>Physical fatigue among construction workers is a major safety concern, impacting both health and productivity. Machine learning (ML) models for fatigue monitoring often struggle with generalizing across varying work conditions and populations. This paper advances fatigue monitoring automation by (1) developing ML models trained under diverse temperature and load conditions (Dataset 1), (2) evaluating generalizability on unseen construction-related data (Dataset 2), and (3) proposing transfer learning-based fine-tuning to enhance models' adaptability while reducing the need for large datasets. Initial accuracies on Dataset 1 were 87.5 % (RFC), 89.7 % (XGBoost), and 92 % (FatigueNet); however, these dropped sharply to 40 % (RFC, XGBoost) and 29 % (FatigueNet) under the generalizability test. When trained from scratch on combined datasets, RFC and FatigueNet achieved 47 % and 60 % accuracy, highlighting challenges with generalization. Transfer learning improved FatigueNet's accuracy to 82 % and RFC's to 87 %. These results demonstrate transfer learning's potential for real-time fatigue monitoring and construction site safety.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"181 ","pages":"Article 106680"},"PeriodicalIF":11.5,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145593092","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}

引用次数: 0

Machine learning-based automatic detection and prediction of cracks and corrosion using spatiotemporal measurements from distributed fiber optic sensors 基于机器学习的裂缝和腐蚀的自动检测和预测，利用分布式光纤传感器的时空测量

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-11-24 DOI: 10.1016/j.autcon.2025.106679

Sina Poorghasem, Yiming Liu, Zhan Jiang, Jinxin Chen, Yi Bao

Monitoring and predicting damages of civil infrastructure are essential for safe and efficient operation and maintenance. This paper presents a digital twin-based approach for automatic detection and prediction of cracks and corrosion utilizing spatiotemporal measurements of strains from distributed fiber optic sensors. Generative machine learning techniques are used to improve the quantity and quality of datasets used to develop damage detection and prediction models. The performance of the approach was evaluated using laboratory experiments through case studies on reinforced concrete beams and steel pipes. Results demonstrated that cracks and corrosion were detected accurately (accuracy>0.98) and efficiently (latency = 0.17 ms). Predictions of strain distributions were performed 7 min ahead for cracks and 21 h ahead for corrosion. The effects of sensing parameters on performance were investigated, enabling sensor configuration optimization. The presented approach advances the ability to monitor and predict damages based on advanced machine learning and distributed fiber optic sensing techniques.

监测和预测民用基础设施的损害是保证民用基础设施安全、高效运行和维护的必要条件。本文提出了一种基于数字孪生的方法，利用分布式光纤传感器的应变时空测量来自动检测和预测裂缝和腐蚀。生成式机器学习技术用于提高用于开发损伤检测和预测模型的数据集的数量和质量。通过钢筋混凝土梁和钢管的实例试验，对该方法的性能进行了评价。结果表明，裂纹和腐蚀检测准确（精度>；0.98），有效（延迟= 0.17 ms）。裂纹和腐蚀分别提前7分钟和21小时预测应变分布。研究了传感参数对性能的影响，实现了传感器配置的优化。该方法基于先进的机器学习和分布式光纤传感技术，提高了监测和预测损伤的能力。

{"title":"Machine learning-based automatic detection and prediction of cracks and corrosion using spatiotemporal measurements from distributed fiber optic sensors","authors":"Sina Poorghasem, Yiming Liu, Zhan Jiang, Jinxin Chen, Yi Bao","doi":"10.1016/j.autcon.2025.106679","DOIUrl":"10.1016/j.autcon.2025.106679","url":null,"abstract":"<div><div>Monitoring and predicting damages of civil infrastructure are essential for safe and efficient operation and maintenance. This paper presents a digital twin-based approach for automatic detection and prediction of cracks and corrosion utilizing spatiotemporal measurements of strains from distributed fiber optic sensors. Generative machine learning techniques are used to improve the quantity and quality of datasets used to develop damage detection and prediction models. The performance of the approach was evaluated using laboratory experiments through case studies on reinforced concrete beams and steel pipes. Results demonstrated that cracks and corrosion were detected accurately (accuracy>0.98) and efficiently (latency = 0.17 ms). Predictions of strain distributions were performed 7 min ahead for cracks and 21 h ahead for corrosion. The effects of sensing parameters on performance were investigated, enabling sensor configuration optimization. The presented approach advances the ability to monitor and predict damages based on advanced machine learning and distributed fiber optic sensing techniques.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"181 ","pages":"Article 106679"},"PeriodicalIF":11.5,"publicationDate":"2025-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145593113","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}

引用次数: 0

Mobile robotic rebar cage assembly via imitation learning 基于模仿学习的移动机器人钢筋笼组装

IF 11.5 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Automation in Construction

Pub Date : 2025-11-24 DOI: 10.1016/j.autcon.2025.106671

Tao Sun , Beining Han , Jimmy Wu , Szymon Rusinkiewicz , Yi Shao

Manipulation remains a key bottleneck in achieving fully autonomous rebar cage assembly. Existing solutions based on rail-guided systems are expensive, poorly scalable, and limited in capability. This paper introduces a framework that leverages a mobile manipulator and uses visual servoing together with imitation learning (IL) to address complex rebar manipulation tasks. The framework enables autonomous execution of two challenging manipulation tasks: (a) tight-fit rebar slot insertion and (b) rebar tying at complex intersection nodes within cages. Using only low-cost RGB cameras, the proposed approach achieves over 90% success rate for over 20 rollouts on both tasks. A highlight is the integration of a segmentation module and a reinsertion strategy that improves tight-fit insertion performance by 41.7% over the baseline and significantly improves robustness to background changes. Notably, the system requires neither depth sensors nor explicit geometric modeling, and supports rapid deployment in novel environments. This paper establishes a foundation for extending autonomy to broader rebar manipulation scenarios. Qualitative results are available on the project website¹.

操作仍然是实现全自动钢筋笼组装的关键瓶颈。现有的基于轨道制导系统的解决方案价格昂贵，可扩展性差，而且能力有限。本文介绍了一种利用移动机械手，结合视觉伺服和模仿学习（IL）来解决复杂钢筋操纵任务的框架。该框架能够自动执行两项具有挑战性的操作任务：(a)紧密配合的螺纹钢槽插入和(b)在笼内复杂的交叉节点绑扎螺纹钢。仅使用低成本的RGB相机，所提出的方法在两个任务上进行超过20次的推出，成功率超过90%。一个亮点是分割模块和插入策略的集成，在基线基础上提高了41.7%的紧密配合插入性能，并显着提高了对背景变化的鲁棒性。值得注意的是，该系统既不需要深度传感器，也不需要明确的几何建模，并且支持在新环境中快速部署。本文为将自治扩展到更广泛的螺纹钢操作场景奠定了基础。定性结果可在项目网站上查阅。

{"title":"Mobile robotic rebar cage assembly via imitation learning","authors":"Tao Sun , Beining Han , Jimmy Wu , Szymon Rusinkiewicz , Yi Shao","doi":"10.1016/j.autcon.2025.106671","DOIUrl":"10.1016/j.autcon.2025.106671","url":null,"abstract":"<div><div>Manipulation remains a key bottleneck in achieving fully autonomous rebar cage assembly. Existing solutions based on rail-guided systems are expensive, poorly scalable, and limited in capability. This paper introduces a framework that leverages a mobile manipulator and uses visual servoing together with imitation learning (IL) to address complex rebar manipulation tasks. The framework enables autonomous execution of two challenging manipulation tasks: (a) tight-fit rebar slot insertion and (b) rebar tying at complex intersection nodes within cages. Using only low-cost RGB cameras, the proposed approach achieves over 90% success rate for over 20 rollouts on both tasks. A highlight is the integration of a segmentation module and a reinsertion strategy that improves tight-fit insertion performance by 41.7% over the baseline and significantly improves robustness to background changes. Notably, the system requires neither depth sensors nor explicit geometric modeling, and supports rapid deployment in novel environments. This paper establishes a foundation for extending autonomy to broader rebar manipulation scenarios. Qualitative results are available on the project website<span><span><sup>1</sup></span></span>.</div></div>","PeriodicalId":8660,"journal":{"name":"Automation in Construction","volume":"181 ","pages":"Article 106671"},"PeriodicalIF":11.5,"publicationDate":"2025-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145593116","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}

引用次数: 0

首页上一页

下一页尾页

类型

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

Automation in Construction

全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.

﹀