首页 > 最新文献

Journal of Industrial Information Integration最新文献

英文 中文
APPD:An Auditable and Privacy-Preserving Data Sharing scheme for Cloud-assisted Industrial Internet of Things APPD:面向云辅助工业物联网的可审计且隐私保护的数据共享方案
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-02-03 DOI: 10.1016/j.jii.2026.101084
Yanbo Yang , Yating Guo , Jiawei Zhang , Zhuo Ma , Jianfeng Ma
Cloud-assisted Industrial Internet of Things (IIoT) is prevalent in offering high quality industrial service by accommodating a huge volume of industrial data to eliminate the heavy burden of resource-limited smart devices and providing convenient industrial data sharing services for participants. However, the outsourced industrial data in remote cloud contain strongly sensitive information of manufacturing and are essential for decisions with analysis. Unauthorized access by malicious users or even destruction to these data will cause severe privacy leakage or manufacturing negligence. Thus, access control, privacy preserving and data integrity are of great significance to industrial data sharing in Cloud-assisted IIoT. Although Ciphertext-Policy Attribute-Based Encryption (CP-ABE) is a powerful tool for cloud data sharing, it incurs several limitations when used in industry field. Many existing schemes lack the ability to deal with data integrity violation, malicious user revocation and user privacy leakage of cleartext access policy simultaneously. Meanwhile, the key escrow is also an important security risk. As a countermeasure, in this paper, we propose an Auditable and Privacy Preserving Data Sharing Framework (APPD) for Cloud-assisted IIoT. In our framework, we devise a novel decentralized CP-ABE scheme with large universe and data auditing to achieve both fine-grained access control with key escrow resistance over unbounded attributes and data integrity guarantee. The full policy hiding and user revocation mechanisms are employed to prevent user privacy from being leaked by access policy and malicious users. At last, we present detailed formal security analysis for our proposal and the thorough performance assessment also demonstrates its feasible in IIoT application.
云辅助工业物联网(IIoT)通过容纳大量工业数据,消除资源有限的智能设备的沉重负担,为参与者提供便捷的工业数据共享服务,从而提供高质量的工业服务。然而,远程云中的外包工业数据包含制造业的高度敏感信息,对分析决策至关重要。恶意用户未经授权访问甚至破坏这些数据将造成严重的隐私泄露或制造疏忽。因此,访问控制、隐私保护和数据完整性对于云辅助工业物联网中的工业数据共享具有重要意义。基于密文策略属性的加密技术(cipher - policy - Attribute-Based Encryption, CP-ABE)是一种强大的云数据共享工具,但在实际应用中存在一定的局限性。现有的许多方案缺乏同时处理明文访问策略的数据完整性破坏、恶意用户撤销和用户隐私泄露的能力。同时,密钥托管也是一个重要的安全风险。作为对策,本文提出了一种可审计且保护隐私的云辅助工业物联网数据共享框架(APPD)。在我们的框架中,我们设计了一种具有大范围和数据审计的新型分散CP-ABE方案,以实现具有无界属性的密钥托管阻力的细粒度访问控制和数据完整性保证。采用全策略隐藏和用户撤销机制,防止用户隐私被访问策略和恶意用户泄露。最后,我们对我们的提案进行了详细的正式安全分析,并进行了全面的性能评估,也证明了其在工业物联网应用中的可行性。
{"title":"APPD:An Auditable and Privacy-Preserving Data Sharing scheme for Cloud-assisted Industrial Internet of Things","authors":"Yanbo Yang ,&nbsp;Yating Guo ,&nbsp;Jiawei Zhang ,&nbsp;Zhuo Ma ,&nbsp;Jianfeng Ma","doi":"10.1016/j.jii.2026.101084","DOIUrl":"10.1016/j.jii.2026.101084","url":null,"abstract":"<div><div>Cloud-assisted Industrial Internet of Things (IIoT) is prevalent in offering high quality industrial service by accommodating a huge volume of industrial data to eliminate the heavy burden of resource-limited smart devices and providing convenient industrial data sharing services for participants. However, the outsourced industrial data in remote cloud contain strongly sensitive information of manufacturing and are essential for decisions with analysis. Unauthorized access by malicious users or even destruction to these data will cause severe privacy leakage or manufacturing negligence. Thus, access control, privacy preserving and data integrity are of great significance to industrial data sharing in Cloud-assisted IIoT. Although Ciphertext-Policy Attribute-Based Encryption (CP-ABE) is a powerful tool for cloud data sharing, it incurs several limitations when used in industry field. Many existing schemes lack the ability to deal with data integrity violation, malicious user revocation and user privacy leakage of cleartext access policy simultaneously. Meanwhile, the key escrow is also an important security risk. As a countermeasure, in this paper, we propose an Auditable and Privacy Preserving Data Sharing Framework (APPD) for Cloud-assisted IIoT. In our framework, we devise a novel decentralized CP-ABE scheme with large universe and data auditing to achieve both fine-grained access control with key escrow resistance over unbounded attributes and data integrity guarantee. The full policy hiding and user revocation mechanisms are employed to prevent user privacy from being leaked by access policy and malicious users. At last, we present detailed formal security analysis for our proposal and the thorough performance assessment also demonstrates its feasible in IIoT application.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101084"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146110394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Prior knowledge-embedded first-layer interpretable paradigm for rail transit vehicle human–computer collaboration fault monitoring 基于先验知识的轨道交通车辆人机协同故障监测第一层可解释范式
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-02-06 DOI: 10.1016/j.jii.2026.101068
Chao He , Hongmei Shi , Jing-Xiao Liao , Bin Liu , Qiuhai Liu , Jianbo Li , Zujun Yu
Rail transit vehicles endure large loads, high speeds, and harsh environment, leading to component failure. The first-layer interpretable paradigm (FLIP) embeds human prior knowledge into smart equipment, which is one of intelligent paradigms guided by customized manufacturing and embodied intelligence. It consists of first-layer interpretable modules, backbones, loss metrics. However, existing efforts rely on single-source information, an absence of interpretable backbones, an inability to feature fusion, thereby struggling with multi-excitation, coupled signals. To bridge this gap, a FLIP-based one-stage multi-view capsule fusion network (PIFCapsule) is proposed. Firstly, a signal processing prior-empowered first-layer interpretable module is devised to realize automatic parameter optimization and highlight the complementarity between multi-view features from different signal processing algorithms. Secondly, an interpretable capsule network serves as the backbone. To overcome the inefficiency and shortage of information fusion, an efficient attention fusion routing (AFR) is proposed to reduce the parameters (about 5.72 times) and the complexity (about 2.93 times) in contrast to the vanilla capsule-based networks. In response to the lack of physics-based constraints during training, a noise threshold amplitude ratio (NTAR) is posed as a regularization, which enhances weak periodic transient pulses by suppressing learned noises. The effectiveness and reliability are verified through three real-world rail transit vehicle datasets: PIFCapsule outperforms the state-of-the-art by 6.77% in accuracy with only ten samples. Given the lightweight nature, it holds substantial promise to be deployed in intelligent edge devices. Code is available at https://github.com/liguge/PIFCapsule.
轨道交通车辆承受大载荷、高速度和恶劣环境,导致部件失效。第一层可解释范式(FLIP)将人类的先验知识嵌入到智能设备中,是一种以定制制造和具体智能为指导的智能范式。它由第一层可解释模块、主干、损耗指标组成。然而,现有的工作依赖于单源信息,缺乏可解释的主干,无法特征融合,因此难以处理多激励,耦合信号。为了弥补这一缺陷,提出了一种基于flip的单级多视点胶囊融合网络(PIFCapsule)。首先,设计信号处理优先级第一层可解释模块,实现参数自动优化,突出不同信号处理算法多视图特征之间的互补性;其次,一个可解释的胶囊网络作为主干。为了克服信息融合的低效率和不足,提出了一种高效的注意力融合路由(AFR),与基于香草胶囊的网络相比,其参数减少了约5.72倍,复杂度减少了约2.93倍。针对训练过程中缺乏物理约束的问题,提出了噪声阈值幅值比(NTAR)作为一种正则化方法,通过抑制学习到的噪声来增强弱周期瞬态脉冲。通过三个真实的轨道交通车辆数据集验证了有效性和可靠性:PIFCapsule仅用10个样本就比最先进的精度高出6.77%。考虑到轻量级的特性,它在智能边缘设备中部署的前景非常可观。代码可从https://github.com/liguge/PIFCapsule获得。
{"title":"Prior knowledge-embedded first-layer interpretable paradigm for rail transit vehicle human–computer collaboration fault monitoring","authors":"Chao He ,&nbsp;Hongmei Shi ,&nbsp;Jing-Xiao Liao ,&nbsp;Bin Liu ,&nbsp;Qiuhai Liu ,&nbsp;Jianbo Li ,&nbsp;Zujun Yu","doi":"10.1016/j.jii.2026.101068","DOIUrl":"10.1016/j.jii.2026.101068","url":null,"abstract":"<div><div>Rail transit vehicles endure large loads, high speeds, and harsh environment, leading to component failure. The first-layer interpretable paradigm (FLIP) embeds human prior knowledge into smart equipment, which is one of intelligent paradigms guided by customized manufacturing and embodied intelligence. It consists of first-layer interpretable modules, backbones, loss metrics. However, existing efforts rely on single-source information, an absence of interpretable backbones, an inability to feature fusion, thereby struggling with multi-excitation, coupled signals. To bridge this gap, a FLIP-based one-stage multi-view capsule fusion network (PIFCapsule) is proposed. Firstly, a signal processing prior-empowered first-layer interpretable module is devised to realize automatic parameter optimization and highlight the complementarity between multi-view features from different signal processing algorithms. Secondly, an interpretable capsule network serves as the backbone. To overcome the inefficiency and shortage of information fusion, an efficient attention fusion routing (AFR) is proposed to reduce the parameters (about 5.72 times) and the complexity (about 2.93 times) in contrast to the vanilla capsule-based networks. In response to the lack of physics-based constraints during training, a noise threshold amplitude ratio (NTAR) is posed as a regularization, which enhances weak periodic transient pulses by suppressing learned noises. The effectiveness and reliability are verified through three real-world rail transit vehicle datasets: PIFCapsule outperforms the state-of-the-art by 6.77% in accuracy with only ten samples. Given the lightweight nature, it holds substantial promise to be deployed in intelligent edge devices. Code is available at <span><span>https://github.com/liguge/PIFCapsule</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101068"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Navigating digitalization and global value chains: Empirical insights from the Chinese manufacturing industry 数字化导航与全球价值链:来自中国制造业的实证分析
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-01-31 DOI: 10.1016/j.jii.2026.101083
Qingyu Zhang , Fakhar Shahzad , Chiranjibe Jana , Nikola Ivkovic , Gerhard-Wilhelm Weber
In a rapidly digitalized and globalized world, enterprises understand how digitalization shapes the global value chain (GVC) to remain competitive. Previous studies have examined digitalization, trade openness, research and development (R&D) investment, foreign direct investment (FDI), and infrastructure quality, leaving a gap in understanding the integrated determinants of GVC. This study aims to fill this research gap by examining the integrated impact of digitalization on GVC. Unlike previous studies, this study develops a holistic framework that captures a multidimensional analysis of the interaction between digitalization and GVC participation. This study used panel data models to achieve the desired outcomes from China’s manufacturing sector, and the results were obtained using Machine Learning Techniques. This study shows that manufacturing, domestic and foreign digitalization, research and development, productivity, and GVC participation all improve a GVC’s position; however, foreign direct investment hampers this improvement. Trade openness, financial growth, and infrastructure all positively impact the relationship between digitalization and the GVC position. By explicitly integrating digital technologies with broader economic and institutional factors, these findings offer a comprehensive understanding of the drivers of GVC competitiveness and provide actionable insights for the manufacturing sectors of emerging economies undergoing rapid digital transformation.
在快速数字化和全球化的世界中,企业了解数字化如何塑造全球价值链(GVC)以保持竞争力。之前的研究考察了数字化、贸易开放、研发投资、外国直接投资(FDI)和基础设施质量,在理解全球价值链的综合决定因素方面存在空白。本研究旨在通过研究数字化对全球价值链的综合影响来填补这一研究空白。与以往的研究不同,本研究开发了一个整体框架,对数字化与全球价值链参与之间的相互作用进行了多维分析。本研究使用面板数据模型来实现中国制造业的预期结果,并使用机器学习技术获得结果。研究表明,制造业、国内外数字化、研发、生产力和全球价值链参与都提高了全球价值链的地位;然而,外国直接投资阻碍了这种改善。贸易开放、金融增长和基础设施都对数字化与全球价值链地位之间的关系产生积极影响。通过明确地将数字技术与更广泛的经济和制度因素相结合,这些研究结果提供了对全球价值链竞争力驱动因素的全面理解,并为正在快速数字化转型的新兴经济体的制造业提供了可操作的见解。
{"title":"Navigating digitalization and global value chains: Empirical insights from the Chinese manufacturing industry","authors":"Qingyu Zhang ,&nbsp;Fakhar Shahzad ,&nbsp;Chiranjibe Jana ,&nbsp;Nikola Ivkovic ,&nbsp;Gerhard-Wilhelm Weber","doi":"10.1016/j.jii.2026.101083","DOIUrl":"10.1016/j.jii.2026.101083","url":null,"abstract":"<div><div>In a rapidly digitalized and globalized world, enterprises understand how digitalization shapes the global value chain (GVC) to remain competitive. Previous studies have examined digitalization, trade openness, research and development (R&amp;D) investment, foreign direct investment (FDI), and infrastructure quality, leaving a gap in understanding the integrated determinants of GVC. This study aims to fill this research gap by examining the integrated impact of digitalization on GVC. Unlike previous studies, this study develops a holistic framework that captures a multidimensional analysis of the interaction between digitalization and GVC participation. This study used panel data models to achieve the desired outcomes from China’s manufacturing sector, and the results were obtained using Machine Learning Techniques. This study shows that manufacturing, domestic and foreign digitalization, research and development, productivity, and GVC participation all improve a GVC’s position; however, foreign direct investment hampers this improvement. Trade openness, financial growth, and infrastructure all positively impact the relationship between digitalization and the GVC position. By explicitly integrating digital technologies with broader economic and institutional factors, these findings offer a comprehensive understanding of the drivers of GVC competitiveness and provide actionable insights for the manufacturing sectors of emerging economies undergoing rapid digital transformation.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101083"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146095830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Industrial information integration through autonomous AI agents: paradoxical effects on transparency, dehumanization, and responsible retail operations 通过自主人工智能代理的工业信息集成:对透明度、非人性化和负责任的零售运营的矛盾影响
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-02-03 DOI: 10.1016/j.jii.2026.101089
Shaofeng Wang , Hao Zhang
Industrial information integration increasingly relies on autonomous artificial intelligence agents to synthesize operational data, coordinate cross-functional processes, and execute real-time decisions in retail environments. This study investigates the paradoxical effects of AI-enabled industrial informatization on operational transparency, algorithmic dehumanization, and responsible performance outcomes. Through a mixed-methods analysis of 419 cross-border e-commerce firms employing structural equation modeling, importance-performance mapping, fuzzy-set qualitative comparative analysis, and executive interviews, we uncover counterintuitive dynamics in human-technology information integration. Results demonstrate that AI agent autonomy enhances operational transparency and reduces algorithmic dehumanization by creating audit-ready information architectures and filtering impersonal computational tasks from human-centric work. However, human-centric governance mechanisms introduce bureaucratic friction that paradoxically weakens these positive integration effects. We identify two mechanisms—"enlightened autonomy," wherein sophisticated systems necessitate built-in information traceability, and "humanity-enabling filters," wherein automation redirects human attention toward relational tasks. These findings challenge prevailing assumptions about AI opacity in industrial information systems and reveal that governance structures intended to safeguard ethical integration may inadvertently undermine the benefits of advanced industrial informatization. The study contributes to industrial information integration theory by demonstrating that autonomous technological systems, when properly architected for information transparency, can foster more responsible and human-compatible industrial operations than traditional management approaches.
在零售环境中,工业信息集成越来越依赖自主人工智能代理来综合运营数据、协调跨职能流程并执行实时决策。本研究探讨了人工智能工业信息化对运营透明度、算法非人性化和负责任绩效结果的矛盾影响。本文采用结构方程模型、重要性-绩效映射、模糊集定性比较分析和高管访谈等方法对419家跨境电子商务企业进行了综合分析,揭示了人-技术信息集成的反直觉动态。结果表明,人工智能代理自主性通过创建审计就绪的信息架构和从以人为中心的工作中过滤非个人计算任务,提高了操作透明度,减少了算法的非人性化。然而,以人为中心的治理机制引入了官僚摩擦,矛盾地削弱了这些积极的整合效应。我们确定了两种机制——“开明的自治”,其中复杂的系统需要内置的信息可追溯性,以及“人性化的过滤器”,其中自动化将人类的注意力重定向到关系任务上。这些发现挑战了关于工业信息系统中人工智能不透明的普遍假设,并揭示了旨在维护道德整合的治理结构可能会无意中破坏先进工业信息化的好处。该研究通过证明自主技术系统,当适当地构建信息透明度时,可以促进比传统管理方法更负责任和更人性化的工业运营,从而为工业信息集成理论做出贡献。
{"title":"Industrial information integration through autonomous AI agents: paradoxical effects on transparency, dehumanization, and responsible retail operations","authors":"Shaofeng Wang ,&nbsp;Hao Zhang","doi":"10.1016/j.jii.2026.101089","DOIUrl":"10.1016/j.jii.2026.101089","url":null,"abstract":"<div><div>Industrial information integration increasingly relies on autonomous artificial intelligence agents to synthesize operational data, coordinate cross-functional processes, and execute real-time decisions in retail environments. This study investigates the paradoxical effects of AI-enabled industrial informatization on operational transparency, algorithmic dehumanization, and responsible performance outcomes. Through a mixed-methods analysis of 419 cross-border e-commerce firms employing structural equation modeling, importance-performance mapping, fuzzy-set qualitative comparative analysis, and executive interviews, we uncover counterintuitive dynamics in human-technology information integration. Results demonstrate that AI agent autonomy enhances operational transparency and reduces algorithmic dehumanization by creating audit-ready information architectures and filtering impersonal computational tasks from human-centric work. However, human-centric governance mechanisms introduce bureaucratic friction that paradoxically weakens these positive integration effects. We identify two mechanisms—\"enlightened autonomy,\" wherein sophisticated systems necessitate built-in information traceability, and \"humanity-enabling filters,\" wherein automation redirects human attention toward relational tasks. These findings challenge prevailing assumptions about AI opacity in industrial information systems and reveal that governance structures intended to safeguard ethical integration may inadvertently undermine the benefits of advanced industrial informatization. The study contributes to industrial information integration theory by demonstrating that autonomous technological systems, when properly architected for information transparency, can foster more responsible and human-compatible industrial operations than traditional management approaches.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101089"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146110395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A delphi-informed fuzzy multi-criteria decision framework for prioritizing sustainable development goals in industrial strategy: an application to the paint and coatings sector 工业战略中可持续发展目标优先排序的德尔菲信息模糊多标准决策框架:在油漆和涂料行业的应用
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-02-05 DOI: 10.1016/j.jii.2026.101091
Seyed Pendar Toufighi , Reza Tafazoli , Sara Habibi , Jan Vang
Industrial sustainability decision-making requires the integration of heterogeneous, uncertain, and expert-based information into actionable strategic priorities. Despite the widespread adoption of the Sustainable Development Goals (SDGs), managers still lack robust decision-support tools to prioritize SDGs and translate them into concrete industrial strategies under uncertainty. To address this gap, this study develops a Delphi-informed fuzzy multi-criteria decision analysis framework that integrates the fuzzy BWM and fuzzy TOPSIS to support SDG prioritization and sustainability-oriented capability ranking. The framework is empirically applied to the paint and coatings industry, a resource- and energy-intensive sector facing increasing sustainability pressures. Based on expert consensus, the results identify SDG 8 (decent work and economic growth), SDG 7 (affordable and clean energy), and SDG 9 (industry, innovation, and infrastructure) as the most critical priorities. Among alternative strategies, sustainability-oriented branding and innovation management emerge as the most effective capabilities for advancing these goals. The study contributes scientifically by operationalizing SDGs as computable decision criteria, integrating stakeholder legitimacy and resource-based perspectives into a unified decision-intelligence framework, and demonstrating how fuzzy techniques can support sustainability-driven industrial strategy formulation under uncertainty.
工业可持续发展决策需要整合异构的、不确定的和基于专家的信息到可操作的战略优先事项。尽管可持续发展目标(sdg)被广泛采用,但管理人员仍然缺乏强大的决策支持工具来优先考虑可持续发展目标,并在不确定性下将其转化为具体的产业战略。为了解决这一差距,本研究开发了一个基于德尔菲的模糊多准则决策分析框架,该框架集成了模糊BWM和模糊TOPSIS,以支持可持续发展目标优先级和以可持续发展为导向的能力排名。该框架是经验适用于油漆和涂料行业,资源和能源密集型行业面临越来越大的可持续发展压力。基于专家共识,结果确定了可持续发展目标8(体面工作和经济增长)、可持续发展目标7(负担得起的清洁能源)和可持续发展目标9(工业、创新和基础设施)为最关键的优先事项。在备选战略中,以可持续发展为导向的品牌和创新管理成为推进这些目标的最有效能力。该研究通过将可持续发展目标作为可计算的决策标准进行操作,将利益相关者合法性和基于资源的观点整合到统一的决策智能框架中,并展示了模糊技术如何在不确定的情况下支持可持续驱动的产业战略制定,从而在科学上做出了贡献。
{"title":"A delphi-informed fuzzy multi-criteria decision framework for prioritizing sustainable development goals in industrial strategy: an application to the paint and coatings sector","authors":"Seyed Pendar Toufighi ,&nbsp;Reza Tafazoli ,&nbsp;Sara Habibi ,&nbsp;Jan Vang","doi":"10.1016/j.jii.2026.101091","DOIUrl":"10.1016/j.jii.2026.101091","url":null,"abstract":"<div><div>Industrial sustainability decision-making requires the integration of heterogeneous, uncertain, and expert-based information into actionable strategic priorities. Despite the widespread adoption of the Sustainable Development Goals (SDGs), managers still lack robust decision-support tools to prioritize SDGs and translate them into concrete industrial strategies under uncertainty. To address this gap, this study develops a Delphi-informed fuzzy multi-criteria decision analysis framework that integrates the fuzzy BWM and fuzzy TOPSIS to support SDG prioritization and sustainability-oriented capability ranking. The framework is empirically applied to the paint and coatings industry, a resource- and energy-intensive sector facing increasing sustainability pressures. Based on expert consensus, the results identify SDG 8 (decent work and economic growth), SDG 7 (affordable and clean energy), and SDG 9 (industry, innovation, and infrastructure) as the most critical priorities. Among alternative strategies, sustainability-oriented branding and innovation management emerge as the most effective capabilities for advancing these goals. The study contributes scientifically by operationalizing SDGs as computable decision criteria, integrating stakeholder legitimacy and resource-based perspectives into a unified decision-intelligence framework, and demonstrating how fuzzy techniques can support sustainability-driven industrial strategy formulation under uncertainty.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101091"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Deep acoustic–visual fusion for robust material recognition in intelligent robotic perception 基于深度声视融合的智能机器人感知材料鲁棒识别
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-01-28 DOI: 10.1016/j.jii.2026.101074
Bo Zhu , Tao Geng , Jia Zhang , Jianlei Cui , Boxin Ren
Accurate material recognition is crucial for intelligent robotic perception, enabling autonomous interaction, grasping, and navigation in complex environments. While traditional single-modality approaches often lack comprehensive information, which limits their performance, multimodal methods that combine acoustic and visual data provide a more robust solution by leveraging complementary cues. However, existing techniques face challenges in effectively integrating these modalities, resulting in suboptimal recognition accuracy under certain conditions. To address these limitations, we propose M3CNet, a novel multimodal material classification network that incorporates adaptive frequency filtering, dual-branch feature fusion, cross-attention, and modality fusion attention. The adaptive frequency filtering block dynamically optimizes acoustic frequency bands to enhance the extraction of discriminative features. Meanwhile, the dual-branch feature fusion block captures local and global visual features at multiple scales, improving texture representation. To strengthen inter-modal relationships, the cross-attention block enables mutual reinforcement between acoustic and visual features, while the modality fusion attention block adaptively balances the contributions of each modality at both the channel and spatial levels. This ensures robustness even in the presence of incomplete or noisy data. Extensive experiments on multiple multimodal texture datasets demonstrate that M3CNet consistently outperforms other methods in accuracy, precision, and recall.
准确的材料识别对于智能机器人感知至关重要,可以在复杂环境中实现自主交互、抓取和导航。传统的单模态方法通常缺乏全面的信息,这限制了它们的性能,而结合声学和视觉数据的多模态方法通过利用互补线索提供了更强大的解决方案。然而,现有技术在有效整合这些模式方面面临挑战,导致在某些条件下识别精度不理想。为了解决这些限制,我们提出了M3CNet,一个新的多模态材料分类网络,它结合了自适应频率滤波、双分支特征融合、交叉注意和模态融合注意。自适应频率滤波块动态优化声频带,增强识别特征的提取。同时,双分支特征融合块在多尺度上捕获局部和全局视觉特征,提高了纹理表征。为了加强多模态关系,交叉注意块使声学和视觉特征之间的相互强化,而模态融合注意块在通道和空间层面上自适应地平衡每种模态的贡献。这确保了即使在存在不完整或有噪声的数据时也具有鲁棒性。在多个多模态纹理数据集上的大量实验表明,M3CNet在准确率、精密度和召回率方面始终优于其他方法。
{"title":"Deep acoustic–visual fusion for robust material recognition in intelligent robotic perception","authors":"Bo Zhu ,&nbsp;Tao Geng ,&nbsp;Jia Zhang ,&nbsp;Jianlei Cui ,&nbsp;Boxin Ren","doi":"10.1016/j.jii.2026.101074","DOIUrl":"10.1016/j.jii.2026.101074","url":null,"abstract":"<div><div>Accurate material recognition is crucial for intelligent robotic perception, enabling autonomous interaction, grasping, and navigation in complex environments. While traditional single-modality approaches often lack comprehensive information, which limits their performance, multimodal methods that combine acoustic and visual data provide a more robust solution by leveraging complementary cues. However, existing techniques face challenges in effectively integrating these modalities, resulting in suboptimal recognition accuracy under certain conditions. To address these limitations, we propose <span><math><msup><mrow><mi>M</mi></mrow><mrow><mn>3</mn></mrow></msup></math></span>CNet, a novel multimodal material classification network that incorporates adaptive frequency filtering, dual-branch feature fusion, cross-attention, and modality fusion attention. The adaptive frequency filtering block dynamically optimizes acoustic frequency bands to enhance the extraction of discriminative features. Meanwhile, the dual-branch feature fusion block captures local and global visual features at multiple scales, improving texture representation. To strengthen inter-modal relationships, the cross-attention block enables mutual reinforcement between acoustic and visual features, while the modality fusion attention block adaptively balances the contributions of each modality at both the channel and spatial levels. This ensures robustness even in the presence of incomplete or noisy data. Extensive experiments on multiple multimodal texture datasets demonstrate that <span><math><msup><mrow><mi>M</mi></mrow><mrow><mn>3</mn></mrow></msup></math></span>CNet consistently outperforms other methods in accuracy, precision, and recall.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101074"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A nested goal programming model integrated with an improved genetic bee colony algorithm supported by machine learning methods 基于机器学习的改进遗传蜂群算法的嵌套目标规划模型
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-01-26 DOI: 10.1016/j.jii.2026.101082
N. Gosheh Dezfouli, Behnam Vahdani, E. Mehdizadeh, H.R. Gholami
Formulating engine oil additives is challenging because it requires simultaneously optimizing production efficiency, cost, and compliance with strict quality standards. This study presents an advanced optimization framework for 10W-40 API SL engine oil that combines a nested goal programming model with machine learning (ML) techniques to predict production rates and quality metrics that cannot be expressed in closed-form equations. To address the inability of conventional ML approaches to generate novel additive combinations, we propose an enhanced genetic bee colony algorithm incorporating arithmetic crossover, Makinen–Periaux–Toivanen mutation operators, and a Cauchy distribution-based local search. These modifications significantly improve the algorithm’s ability to explore and evaluate new formulations. The resulting framework achieves 98.76% of nominal production capacity—very close to the theoretical optimum—while reducing quality-related costs by an average of 20.44%. These results represent substantial improvements in production efficiency, cost savings, and overall formulation quality, providing a powerful and practical tool for the engine oil industry.
配制机油添加剂是一项具有挑战性的工作,因为它需要同时优化生产效率、成本,并符合严格的质量标准。本研究提出了一种先进的10W-40 API SL机油优化框架,该框架将嵌套目标规划模型与机器学习(ML)技术相结合,可以预测无法用封闭形式方程表示的生产率和质量指标。为了解决传统机器学习方法无法生成新的加性组合的问题,我们提出了一种增强的遗传蜂群算法,该算法结合了算术交叉、Makinen-Periaux-Toivanen突变算子和基于Cauchy分布的局部搜索。这些修改显著提高了算法探索和评估新公式的能力。最终的框架实现了98.76%的名义产能——非常接近理论最优——同时平均降低了20.44%的质量相关成本。这些结果代表了生产效率、成本节约和整体配方质量的大幅提高,为发动机润滑油行业提供了一个强大而实用的工具。
{"title":"A nested goal programming model integrated with an improved genetic bee colony algorithm supported by machine learning methods","authors":"N. Gosheh Dezfouli,&nbsp;Behnam Vahdani,&nbsp;E. Mehdizadeh,&nbsp;H.R. Gholami","doi":"10.1016/j.jii.2026.101082","DOIUrl":"10.1016/j.jii.2026.101082","url":null,"abstract":"<div><div>Formulating engine oil additives is challenging because it requires simultaneously optimizing production efficiency, cost, and compliance with strict quality standards. This study presents an advanced optimization framework for 10W-40 API SL engine oil that combines a nested goal programming model with machine learning (ML) techniques to predict production rates and quality metrics that cannot be expressed in closed-form equations. To address the inability of conventional ML approaches to generate novel additive combinations, we propose an enhanced genetic bee colony algorithm incorporating arithmetic crossover, Makinen–Periaux–Toivanen mutation operators, and a Cauchy distribution-based local search. These modifications significantly improve the algorithm’s ability to explore and evaluate new formulations. The resulting framework achieves 98.76% of nominal production capacity—very close to the theoretical optimum—while reducing quality-related costs by an average of 20.44%. These results represent substantial improvements in production efficiency, cost savings, and overall formulation quality, providing a powerful and practical tool for the engine oil industry.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101082"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146048129","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
FedAHPIP: Federated Learning with Adaptive Hot Parameter Identification and Personalized Anchoring for multi-agent collaboration FedAHPIP:基于自适应热参数识别和个性化锚定的多智能体协作的联邦学习
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-02-05 DOI: 10.1016/j.jii.2026.101087
Cangming Liang , Zulong Diao , Xin Wang , Yingzi Huo , Kuanching Li , Dacheng He , Wei Liang
Smart manufacturing seeks to achieve collective intelligence through collaboration. However, such collaboration must be secure and personalized to handle heterogeneous industrial agents. Federated learning offers a promising paradigm for this setting but faces two fundamental challenges: privacy leakage through gradient inversion attacks (e.g., DLG) and data heterogeneity requiring personalized models. To address these challenges, we propose FedAHPIP, a federated learning framework that integrates secure aggregation with personalized learning. Our approach includes an adaptive hot parameter identification mechanism that dynamically identifies sensitive parameters (hot parameters) based on their update momentum, layer semantics, and potential label leakage risks. By focusing encryption on these hot parameters, FedAHPIP drastically reduces the privacy leakage surface. We also develop a personalized anchoring strategy that allows each agent to retain its critical parameters while assimilating knowledge from the global model, effectively balancing personalization and collaboration. Extensive experiments on benchmark and industrial datasets demonstrate that FedAHPIP achieves superior personalized accuracy under extreme non-IID settings, provides robust security against DLG attacks, and maintains minimal computational overhead. FedAHPIP thus offers a practical solution for trustworthy collective intelligence in smart manufacturing environments.
智能制造寻求通过协作实现集体智能。然而,这种协作必须是安全和个性化的,以处理异构的工业代理。联邦学习为这种设置提供了一个有前途的范例,但面临两个基本挑战:通过梯度反转攻击(例如,DLG)泄露隐私,以及需要个性化模型的数据异构性。为了应对这些挑战,我们提出了FedAHPIP,这是一个将安全聚合与个性化学习集成在一起的联邦学习框架。我们的方法包括一个自适应热参数识别机制,该机制根据敏感参数的更新势头、层语义和潜在的标签泄漏风险动态识别敏感参数(热参数)。通过将加密重点放在这些热门参数上,FedAHPIP大大减少了隐私泄漏面。我们还开发了一种个性化的锚定策略,允许每个代理在从全局模型中吸收知识的同时保留其关键参数,有效地平衡个性化和协作。在基准和工业数据集上进行的大量实验表明,FedAHPIP在极端的非iid设置下实现了卓越的个性化准确性,提供了针对DLG攻击的强大安全性,并保持了最小的计算开销。因此,FedAHPIP为智能制造环境中值得信赖的集体智能提供了一个实用的解决方案。
{"title":"FedAHPIP: Federated Learning with Adaptive Hot Parameter Identification and Personalized Anchoring for multi-agent collaboration","authors":"Cangming Liang ,&nbsp;Zulong Diao ,&nbsp;Xin Wang ,&nbsp;Yingzi Huo ,&nbsp;Kuanching Li ,&nbsp;Dacheng He ,&nbsp;Wei Liang","doi":"10.1016/j.jii.2026.101087","DOIUrl":"10.1016/j.jii.2026.101087","url":null,"abstract":"<div><div>Smart manufacturing seeks to achieve collective intelligence through collaboration. However, such collaboration must be secure and personalized to handle heterogeneous industrial agents. Federated learning offers a promising paradigm for this setting but faces two fundamental challenges: privacy leakage through gradient inversion attacks (e.g., DLG) and data heterogeneity requiring personalized models. To address these challenges, we propose FedAHPIP, a federated learning framework that integrates secure aggregation with personalized learning. Our approach includes an adaptive hot parameter identification mechanism that dynamically identifies sensitive parameters (hot parameters) based on their update momentum, layer semantics, and potential label leakage risks. By focusing encryption on these hot parameters, FedAHPIP drastically reduces the privacy leakage surface. We also develop a personalized anchoring strategy that allows each agent to retain its critical parameters while assimilating knowledge from the global model, effectively balancing personalization and collaboration. Extensive experiments on benchmark and industrial datasets demonstrate that FedAHPIP achieves superior personalized accuracy under extreme non-IID settings, provides robust security against DLG attacks, and maintains minimal computational overhead. FedAHPIP thus offers a practical solution for trustworthy collective intelligence in smart manufacturing environments.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101087"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A multi-level multi-source digital twin model for performance enhancement and optimization decision-making in precision milling machines 精密铣床性能提升与优化决策的多级多源数字孪生模型
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-01-23 DOI: 10.1016/j.jii.2026.101080
Yang Xie , Shulong Mei , Fei Wang , Chaoyong Zhang
The transition of CNC machining toward digitalization and low-carbon manufacturing is essential for the advancement of intelligent production. However, conventional parameter configuration methods fail to balance efficiency and sustainability. To overcome this limitation, this study proposes an intelligent optimization framework that integrates digital twin (DT) technology with multi-objective optimization. A multi-level virtual machine tool model is established to enable operational condition mapping and structural response modeling of key machining parameters. A Simulation Augmentation Collaboration Mechanism (SACM) is further introduced, in which the DT generates high-fidelity distribution information to guide a Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) in producing realistic samples under critical operating conditions. These augmented data iteratively refine the model, significantly enhancing predictive generalization. An Improved Meta-Learning-Optimized XGBoost (IMeta-XGBoost) model is then established to predict three performance indicators: spindle energy consumption, specific cutting energy, and machining noise. A Predicted-Fitness-Guided Multi-Objective Deep Q-Network (PF-MO-DQN) is then employed for global optimization, followed by entropy-weighted TOPSIS to determine the optimal machining parameters experimental validation demonstrates reductions of 8.95% in spindle energy consumption, 18.03% in specific cutting energy, and 10.15% in machining noise, confirming significant improvements in energy efficiency, productivity, and noise mitigation. This work provides a robust and scalable approach for multi-objective optimization in complex machining environments.
数控加工向数字化和低碳制造的过渡是推进智能生产的必要条件。然而,传统的参数配置方法无法平衡效率和可持续性。为了克服这一限制,本研究提出了一种将数字孪生(DT)技术与多目标优化相结合的智能优化框架。建立了多级虚拟机床模型,实现了运行工况映射和关键加工参数的结构响应建模。进一步介绍了一种仿真增强协作机制(SACM),其中DT生成高保真的分布信息,指导WGAN-GP在关键操作条件下生成真实样本。这些增强的数据迭代地改进了模型,显著增强了预测泛化。然后建立改进的元学习优化XGBoost (i - meta -XGBoost)模型,预测主轴能耗、切削比能量和加工噪声三个性能指标。然后采用预测适应度引导的多目标深度q -网络(PF-MO-DQN)进行全局优化,然后采用熵加权TOPSIS来确定最优加工参数。实验验证表明,主轴能耗降低了8.95%,比切削能量降低了18.03%,加工噪声降低了10.15%,证实了能源效率、生产率和噪声缓解方面的显着提高。这项工作为复杂加工环境下的多目标优化提供了一种鲁棒性和可扩展性的方法。
{"title":"A multi-level multi-source digital twin model for performance enhancement and optimization decision-making in precision milling machines","authors":"Yang Xie ,&nbsp;Shulong Mei ,&nbsp;Fei Wang ,&nbsp;Chaoyong Zhang","doi":"10.1016/j.jii.2026.101080","DOIUrl":"10.1016/j.jii.2026.101080","url":null,"abstract":"<div><div>The transition of CNC machining toward digitalization and low-carbon manufacturing is essential for the advancement of intelligent production. However, conventional parameter configuration methods fail to balance efficiency and sustainability. To overcome this limitation, this study proposes an intelligent optimization framework that integrates digital twin (DT) technology with multi-objective optimization. A multi-level virtual machine tool model is established to enable operational condition mapping and structural response modeling of key machining parameters. A Simulation Augmentation Collaboration Mechanism (SACM) is further introduced, in which the DT generates high-fidelity distribution information to guide a Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) in producing realistic samples under critical operating conditions. These augmented data iteratively refine the model, significantly enhancing predictive generalization. An Improved Meta-Learning-Optimized XGBoost (IMeta-XGBoost) model is then established to predict three performance indicators: spindle energy consumption, specific cutting energy, and machining noise. A Predicted-Fitness-Guided Multi-Objective Deep Q-Network (PF-MO-DQN) is then employed for global optimization, followed by entropy-weighted TOPSIS to determine the optimal machining parameters experimental validation demonstrates reductions of 8.95% in spindle energy consumption, 18.03% in specific cutting energy, and 10.15% in machining noise, confirming significant improvements in energy efficiency, productivity, and noise mitigation. This work provides a robust and scalable approach for multi-objective optimization in complex machining environments.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101080"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
BIM-based construction scheduling optimization through graph neural network-driven spatial semantic reasoning 基于bim的图神经网络驱动空间语义推理施工调度优化
IF 10.4 1区 计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS Pub Date : 2026-05-01 Epub Date: 2026-02-07 DOI: 10.1016/j.jii.2026.101088
Wenbo Zhang , Zhengtao Wang , Zhenqing Yang , Zhi Cai
With the widespread adoption of Building Information Modeling (BIM) in construction engineering, its embedded spatial semantic information provides a valuable foundation for intelligent task scheduling. However, existing methods often fail to fully exploit deep spatial relationships or adapt to dynamically evolving scenarios influenced by multiple constraints. To address these challenges, this paper proposes a construction scheduling optimization framework based on Graph Neural Network (GNN)-driven spatial semantic reasoning. First, a three-dimensional semantic graph is constructed from BIM data by integrating geometric, topological, and attribute information to explicitly represent relationships such as containment, embedding, and contact among components. Then, an enhanced GraphSAGE model is employed to learn implicit precedence dependencies, replacing static rule templates and improving the adaptability and generalization of scheduling logic. Furthermore, a Dynamic Layering Scheduling Algorithm (DLSA) is designed to exploit spatial adjacency and parallel constructability, enabling structured and controllable scheduling. A multi-dimensional priority model is also incorporated, accounting for component criticality, spatial position, and resource conflicts to dynamically generate construction sequences. Experiments conducted on three representative BIM project models demonstrate that the proposed framework improves resource utilization (RU) from 68.24% to 87.93% and reduces abnormal construction events (AC) from 185 to 41, representing a 19.69 percentage point gain in efficiency and a 77.8% reduction in conflicts. These results confirm the effectiveness, scalability, and industrial applicability of integrating GNN-driven reasoning with BIM for intelligent construction scheduling optimization.
随着建筑信息模型(BIM)在建筑工程中的广泛应用,其嵌入的空间语义信息为智能任务调度提供了宝贵的基础。然而,现有的方法往往不能充分利用深层空间关系或适应受多种约束影响的动态变化情景。为了解决这些问题,本文提出了一个基于图神经网络(GNN)驱动的空间语义推理的施工调度优化框架。首先,通过整合几何、拓扑和属性信息,从BIM数据中构建三维语义图,明确表示组件之间的包容、嵌入、接触等关系。然后,利用增强的GraphSAGE模型学习隐式优先依赖关系,取代静态规则模板,提高调度逻辑的适应性和泛化性。利用空间邻接性和并行可构造性,设计了动态分层调度算法(DLSA),实现了调度的结构化和可控。该方法采用多维优先级模型,考虑构件的临界性、空间位置和资源冲突等因素,动态生成构造序列。在三个具有代表性的BIM项目模型上进行的实验表明,所提出的框架将资源利用率(RU)从68.24%提高到87.93%,将异常施工事件(AC)从185个减少到41个,效率提高了19.69个百分点,冲突减少了77.8%。这些结果证实了将gnn驱动推理与BIM集成用于智能施工调度优化的有效性、可扩展性和工业适用性。
{"title":"BIM-based construction scheduling optimization through graph neural network-driven spatial semantic reasoning","authors":"Wenbo Zhang ,&nbsp;Zhengtao Wang ,&nbsp;Zhenqing Yang ,&nbsp;Zhi Cai","doi":"10.1016/j.jii.2026.101088","DOIUrl":"10.1016/j.jii.2026.101088","url":null,"abstract":"<div><div>With the widespread adoption of Building Information Modeling (BIM) in construction engineering, its embedded spatial semantic information provides a valuable foundation for intelligent task scheduling. However, existing methods often fail to fully exploit deep spatial relationships or adapt to dynamically evolving scenarios influenced by multiple constraints. To address these challenges, this paper proposes a construction scheduling optimization framework based on Graph Neural Network (GNN)-driven spatial semantic reasoning. First, a three-dimensional semantic graph is constructed from BIM data by integrating geometric, topological, and attribute information to explicitly represent relationships such as containment, embedding, and contact among components. Then, an enhanced GraphSAGE model is employed to learn implicit precedence dependencies, replacing static rule templates and improving the adaptability and generalization of scheduling logic. Furthermore, a Dynamic Layering Scheduling Algorithm (DLSA) is designed to exploit spatial adjacency and parallel constructability, enabling structured and controllable scheduling. A multi-dimensional priority model is also incorporated, accounting for component criticality, spatial position, and resource conflicts to dynamically generate construction sequences. Experiments conducted on three representative BIM project models demonstrate that the proposed framework improves resource utilization (RU) from 68.24% to 87.93% and reduces abnormal construction events (AC) from 185 to 41, representing a 19.69 percentage point gain in efficiency and a 77.8% reduction in conflicts. These results confirm the effectiveness, scalability, and industrial applicability of integrating GNN-driven reasoning with BIM for intelligent construction scheduling optimization.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"51 ","pages":"Article 101088"},"PeriodicalIF":10.4,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146138393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
Journal of Industrial Information Integration
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.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1