Journal of Industrial Information Integration最新文献

{"title":"Digital Twin-driven cross-domain fault diagnosis for axial piston pumps via deep transfer learning under small-sample condition","authors":"Chunjie Ma ,&nbsp;Ping Yan ,&nbsp;Bocheng Wang ,&nbsp;Lin Gao ,&nbsp;Junlei Du ,&nbsp;Lianqiang Feng ,&nbsp;Han Zhou","doi":"10.1016/j.jii.2025.100966","DOIUrl":"10.1016/j.jii.2025.100966","url":null,"abstract":"<div><div>Axial piston pump is a complex and typical thermal-fluid-structural coupled system. Its reliability directly affects the operational stability of the complex hydraulic system. It faces challenges including scarce fault samples and data distribution discrepancies across operating conditions. Regarding the problem that traditional methods fail to effectively integrate and utilize multi-source information, resulting in incomplete description of fault information, this paper proposes an intelligent cross-domain industrial information integration fault diagnosis method that integrates Digital Twin and adversarial transfer. Firstly, a multi-domain coupled Digital Twin model is constructed to generate multi-source fault simulation information data. The model employs co-simulation of multi-body dynamics and hydraulic systems to ensure the physical fidelity of fault information. Multi-source fused Gramian Angular Summation Fields feature encoding is designed to map multidimensional signals into two-dimensional spatiotemporal correlation images, thereby integrating and enhancing the representation of information. Secondly, an improved Auxiliary Classifier Generative Adversarial Network with multiple generators is adopted to align the distributions of simulated and measured data, with a dynamic optimization strategy employed to enhance generation quality. Finally, a Multi-scale Attention Domain Adversarial Transfer Network is constructed, combining a Gradient Reversal Layer and Conditional Maximum Mean Discrepancy to suppress the cross-domain distribution differences between the simulation and the experimental data. The experiment shows that by integrating experimental and simulation data, the proposed method achieves an average accuracy of over 98 % in cross-condition fault diagnosis tasks under unknown conditions, showing significant improvement over traditional transfer learning methods. Ablation studies validate the effectiveness of each module, providing a novel approach for complex hydraulic system fault diagnosis under small-sample scenarios.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100966"},"PeriodicalIF":10.4,"publicationDate":"2025-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145261934","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}

{"title":"Enabling interoperable human–AI teaming for automation in construction and manufacturing via Digital Twins and Sliding Work Sharing ontologies","authors":"Pantelis Karapanagiotis ,&nbsp;Kolitha Kottagaha W.M. ,&nbsp;Diego Rovere ,&nbsp;Jos A.C. Bokhorst ,&nbsp;Andrea Valdata ,&nbsp;Christos Emmanouilidis","doi":"10.1016/j.jii.2025.100962","DOIUrl":"10.1016/j.jii.2025.100962","url":null,"abstract":"<div><div>This paper introduces an ontology system to support dynamic, explainable, and human-centric collaboration between humans and artificial intelligence-enabled non-human agents in cyber–physical environments. In this setting, Digital Twins (digital models of physical systems or processes that mirror their real-time state) and Human Digital Twins (digital representations of individual humans, including their physiological or cognitive states) may provide information to enable an appropriate dynamic allocation of the work that can be shared by humans and AI actors (i.e., sliding work sharing). A novel upper-level Sliding Work Sharing ontology is defined to support semantic interoperability and reasoning across diverse domains, facilitating sliding work sharing in complex environments. The ontology is grounded in Industry 5.0 concepts and built upon the Industrial Ontology Foundry core ontology. It extends conventional scheduling ontologies by incorporating key constructs for Digital Twins, Human Digital Twins, and dynamic task flows. We validate the ontology through two use cases from the domains of automation in construction and manufacturing. The collaborative construction case involves robots and humans, while the manufacturing one integrates legacy systems, artificial intelligence actors, and human planners. The developed ontology system is evaluated for its coverage and expressiveness through a novel Retrieval-Augmented Generation based methodology, applied on diverse Large Language Models to derive competency questions from external sources. This approach enhances conventional ontology validation techniques with a scalable and unbiased alternative. Logical consistency is confirmed using a range of standard reasoners. Our results demonstrate that the Sliding Work Sharing ontology has considerable flexibility and potential to advance human–AI teaming in future work environments.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100962"},"PeriodicalIF":10.4,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145181251","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}

{"title":"OPC UA-based three-layer architecture for aggregated microgrids integrating edge cloud computing and IEC 62264","authors":"Changdae Lee ,&nbsp;Hyukyoon Kwon ,&nbsp;Young Il Lee","doi":"10.1016/j.jii.2025.100965","DOIUrl":"10.1016/j.jii.2025.100965","url":null,"abstract":"<div><div>This paper presents a three-layer architecture based on Open Platform Communications Unified Architecture (OPC UA) to address interoperability challenges in aggregated microgrid systems, including protocol heterogeneity and latency mismatches. Designed to meet the latency requirements of two key energy management strategies — high-step (hourly BESS scheduling) and low-step (real-time BESS adjustments) — the architecture segments communication into the Microgrid Aggregation Layer, Communication Platform Layer, and Distributed Microgrid Layer. The proposed architecture achieves seamless and scalable data exchange while ensuring compatibility with heterogeneous devices and supporting flexible operations by leveraging the OPC UA platform, Fieldbus protocols, edge cloud computing, and IEC 62264 standards. Compared to IEC 61850, OPC UA offers broader interoperability, dynamic semantic modeling, seamless OT/IT integration, and robust TLS/AES256-based security, making it well-suited for secure cloud-integrated microgrids. Additionally, OPC UA-based communication operates through dynamic information models. These models enable flexible and adaptive structuring of device data in the microgrid ecosystem. This paper also defines Cloud and Microgrid Component Information Models, specifically designed for microgrid environments. These dynamic models enable selective data updates and hierarchical structuring of information, reducing unnecessary network traffic and improving responsiveness across the architecture. Validation on a real-world testbed demonstrates up to 17.86-fold latency reduction and compliance with Industry 4.0 benchmarks, highlighting the effectiveness of the architecture in enabling scalable, real-time microgrid management and establishing a robust foundation for practical energy systems.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100965"},"PeriodicalIF":10.4,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145181270","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}

{"title":"Interoperability of AI-enhanced digital twins","authors":"Umar Memon,&nbsp;Wolfgang Mayer,&nbsp;Matt Selway,&nbsp;Markus Stumptner","doi":"10.1016/j.jii.2025.100961","DOIUrl":"10.1016/j.jii.2025.100961","url":null,"abstract":"<div><div>Interoperability is one of the biggest challenges when multiple digital twins are used in collaboration. Although attempts to standardise and define interfaces have made significant progress, real interoperability is still difficult to achieve. It is due to unstated assumptions, contextual factors, and quality characteristics not covered by conventional methods. This paper presents a composition framework that uses a meta-model to capture contextual factors and quality characteristics in a structured manner that is required for compatibility between the models. It is achieved by developing a meta-model that explicitly represents the quality characteristics that can be used to decide whether digital twin models can be validly composed. Validation of the approach is illustrated by examples showing how our approach identifies the issues that are otherwise hidden compatibility issues. This paper also provides an algorithm to provide reasoning logic for requirements assessment by making implicit assumptions and contextual factors explicit and enabling the composition of digital twin models to be more effective.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100961"},"PeriodicalIF":10.4,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220641","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}

{"title":"An optimized learning approach for enhancing the security of digital twin-enabled industrial systems from distributed denial-of-service attacks","authors":"Debendra Muduli,&nbsp;Rahul Kumar Gupta,&nbsp;Samir Kumar Majhi,&nbsp;Binayak Ojha,&nbsp;Banshidhar Majhi","doi":"10.1016/j.jii.2025.100960","DOIUrl":"10.1016/j.jii.2025.100960","url":null,"abstract":"<div><div>During the revolution of Industry 4.0, digital twin technology is transforming industrial operations by creating digital models of physical assets, processes, and systems. This innovation enables real-time monitoring, predictive maintenance, and enhanced decision-making capabilities. However, as digital twins become integral to industrial environments, they also introduce new cybersecurity challenges, particularly in the form of distributed denial-of-service (DDoS) attacks, which can disrupt operations and compromise data integrity. This study investigates the resilience of digital twin-based industrial organizations in cyberattack scenarios, specifically focusing on the impacts of DDoS attacks on functional and financial performance. In this paper, a hybrid DDoS attack detection model is introduced, integrating multiple techniques for data preprocessing, feature selection, dimensionality reduction, and classification . To address the class imbalance issue,Synthetic Minority Over-sampling Technique (SMOTE) is applied during preprocessing. Feature selection is performed using filter-based methods, including Information Gain, Gain Ratio, ANOVA F-statistic, Pearson Correlation, and the technique for order preference by similarity to ideal solution (TOPSIS), a multi-criteria decision-making method. To enhance computational efficiency, principal component analysis (PCA) is used for dimensionality reduction, preserving critical information while reducing redundancy. For classification, an extreme learning machine (ELM) is optimized using the particle swarm optimization (PSO) algorithm, improving generalization, preventing overfitting, and ensuring faster convergence. The experiment is conducted using the publicly available CICDDoS2019 dataset in both standalone and cloud-based environments with configurations of vCPU-4, vCPU-8, and vCPU-16. Additionally, a 5-fold stratified cross-validation approach is employed to enhance the model’s generalization performance and ensure robustness across different data distributions. The experimental results indicate that the proposed model achieves a 99.97% detection accuracy and an AUC score of 0.99 in the cloud environment with vCPU-16 and 64GB RAM, outperforming traditional algorithms in DDoS detection. The experimental study finds that increased computational resources improve performance, indicating the model’s adaptability. As digital twins rely on seamless physical-virtual communication, DDoS attacks threaten synchronization, latency, and reliability. The proposed detection approach enhances resilience, minimizes downtime, and preserves process integrity, contributing to secure and robust digital twin architectures in Industry 4.0.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100960"},"PeriodicalIF":10.4,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158368","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}

{"title":"Multi-objective heterogeneous interactive human-robot collaborative disassembly line balancing with partial destructive mode in type-2 fuzzy environment","authors":"Xuesong Zhang ,&nbsp;Qiang Zhao ,&nbsp;Guangdong Tian ,&nbsp;Amir M. Fathollahi-Fard ,&nbsp;Zaher Mundher Yaseen ,&nbsp;Duc Truong Pham","doi":"10.1016/j.jii.2025.100963","DOIUrl":"10.1016/j.jii.2025.100963","url":null,"abstract":"<div><div>Interactive human-robot collaborative disassembly lines represent a prominent application of the Industry 5.0 paradigm, brought to life by the synergy of industrial information integration and advanced automation. This advanced production model enhances recycling efficiency for End-of- Life products by integrating the complementary strengths of humans and robots, enabling their cooperative or parallel execution of disassembly tasks. However, balancing such lines remains challenging. Existing research has limited consideration of the uncertainties in the disassembly process and the heterogeneity of operators, typically adopting non-destructive disassembly modes that are not always feasible in practice, thereby limiting their applicability. Therefore, this study proposes a fuzzy partial destructive heterogeneous interactive human-robot collaborative disassembly line balancing problem. The problem supports flexible configurations of diverse operators, employs interval type-2 triangular fuzzy sets to represent uncertainties during disassembly more precisely, and introduces a partial destructive disassembly mode. It aims to minimise the number of active workstations, differences in operator idle time, and disassembly energy consumption. Given the NP-hard nature of this problem, a multi-objective discrete bees algorithm is developed. This algorithm adopts a two-parameter bees algorithm search framework, incorporating four scout bee initialisation rules, five neighbourhood search operators, an adaptive operator selection strategy, and two enhanced search operators. The proposed model and algorithm are applied to two disassembly scenarios involving a retired power battery and a transmission. Further comparative analysis shows that the proposed method improves disassembly performance. To further understand the behaviour and performance of the model, we also conduct comprehensive sensitivity analyses. Finally, comparative experiments are performed against the GUROBI solver and five other advanced algorithms, validating that the proposed algorithm exhibits superior performance.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100963"},"PeriodicalIF":10.4,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267315","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}

{"title":"Semantic and modular orchestration of AI-driven digital twins for industrial interoperability and optimization","authors":"Maria Gabriela Juarez Juarez,&nbsp;Adriana Giret,&nbsp;Vicente Botti","doi":"10.1016/j.jii.2025.100959","DOIUrl":"10.1016/j.jii.2025.100959","url":null,"abstract":"<div><div>Digital Twins (DTs) are foundational in smart manufacturing, supporting data-driven monitoring and optimization. Yet, many implementations remain monolithic, limiting interoperability and reusability. This paper introduces a semantic and modular architecture for orchestrating AI-driven DTs, designed to enable scalable integration and standardized coordination across industrial systems. The system employs a semantic API aligned with NGSI-LD, to expose industrial entities such as processes, anomalies, assets, and contextual KPIs (e.g., energy usage, <span><math><msub><mrow><mi>CO</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span> emissions, tool wear, product quality). AI techniques ranging from threshold adjustment to symbolic learning are encapsulated as modular agents, each performing targeted optimization tasks. These agents operate over the semantic API, which ensures consistent, interpretable interactions across modules. A Manager and a Recommender agent are defined to coordinate execution; while not yet deployed at runtime, their logic is implemented through semantic interfaces that support traceable, modular activation. The system is validated using synthetic data simulating machining, assembly, and inspection tasks. Results show measurable improvements in sustainability-related KPIs following each module’s activation. More importantly, the semantic orchestration layer enables modularity, interoperability, and AI reuse. This work contributes a standards-compliant foundation for next-generation DTs, supporting integration with ecosystems such as FIWARE, Catena-X, and IDS, and aligned with the principles of Industry 4.0 and 5.0.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100959"},"PeriodicalIF":10.4,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158367","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}

{"title":"Multi-grained teacher–student joint representation learning for surface defect classification","authors":"Chunlei Meng ,&nbsp;Jiacheng Yang ,&nbsp;Wei Lin ,&nbsp;Linqiang Hu ,&nbsp;Bowen Liu ,&nbsp;Zhuo Zou ,&nbsp;LiDa Xu ,&nbsp;Zhongxue Gan ,&nbsp;Chun Ouyang","doi":"10.1016/j.jii.2025.100958","DOIUrl":"10.1016/j.jii.2025.100958","url":null,"abstract":"<div><div>Surface defect classification (SDC) plays a critical role in ensuring product quality within industrial systems. Surface defects are characterized by complex noise backgrounds, diverse defect types, and multi-scale defect shapes. Existing methods often struggle to effectively learn multi-grained defect information in such complex environments. This study introduces a Multi-Grained Teacher-Student Joint Representation Learning (MGJR) framework, which integrates both coarse-grained and fine-grained representation learning in a unified architecture. A ViT-based teacher network first learns holistic global features from defect-rich backgrounds. These features guide a student network enhanced with an Integrated Efficient Multi-Attention (IEMA) module and a Global-Local Attention (GL-Attention) mechanism, enabling the extraction and fusion of multi-scale features to preserve context while emphasizing local anomalies. Additionally, the anchor-guided training strategy (AGTS) serves as a consistency constraint, enhancing robustness by aligning the teacher’s stable coarse-grained signal with the student model’s fine-grained response under noisy inputs. The entire framework is optimized end-to-end using a unified loss that combines coarse-level guidance with task-specific supervision. Extensive experiments demonstrate that MGJR achieves 99.98% accuracy on the NEU-CLS dataset and consistently outperforms previous methods across multiple industrial benchmarks. The model remains lightweight, with 21.14 million parameters and 2.86 billion FLOPs. MGJR shows good performance in noisy conditions and other classification tasks. To demonstrate its practical effectiveness, this study built a wood surface defect dataset with 7 defect types and 2,654 images from real industrial settings. MGJR achieved top performance on this dataset, verifying its applicability in real-world.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100958"},"PeriodicalIF":10.4,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121271","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}

{"title":"Hub-blockchain: Unveiling the transparency paradigm in physical supply chains","authors":"Ardavan Babaei ,&nbsp;Erfan Babaee Tirkolaee ,&nbsp;Mosè Gallo ,&nbsp;Mohammad Reza Akbari Jokar ,&nbsp;Majid Khedmati","doi":"10.1016/j.jii.2025.100964","DOIUrl":"10.1016/j.jii.2025.100964","url":null,"abstract":"<div><div>This work presents a groundbreaking innovation in Supply Chain Management (SCM) by introducing the concept of \"Hub-Blockchain\". While the adoption of blockchain technology already brings transparency to supply chains, its full implementation can be financially burdensome for companies. In order to address this challenge, this research proposes a cost-effective solution where hubs are established to facilitate the storage, exchange, analysis, separation, sending, and distribution of data. These hubs also ensure equal access to data for all supply chain members. By implementing blockchain on these hubs, the approach significantly enhances supply chain transparency and optimizes its operations without necessitating any modifications to the existing data structure. To do so, we develop an optimization model that simultaneously designs the physical supply chain and its blockchain counterpart. The model aims to minimize total costs while maximizing transparency resulting from the adoption of blockchain technology. To effectively and optimally solve this multi-objective model, Fuzzy Goal Programming (FGP) approach is applied. Results demonstrate the cost-effectiveness of hub-blockchains in supply chain planning and in mitigating data traffic congestion. However, the results reveal that the hub-blockchain strategy faces a law of diminishing returns, that is, increasing the number of hub-blockchains may diminish the corresponding advantage.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100964"},"PeriodicalIF":10.4,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158369","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}

{"title":"LAECIPS: Large vision model assisted adaptive edge–cloud collaboration for IoT-based embodied intelligence system","authors":"Shijing Hu ,&nbsp;Zhihui Lu ,&nbsp;Xin Xu ,&nbsp;Ruijun Deng ,&nbsp;Xin Du ,&nbsp;Qiang Duan","doi":"10.1016/j.jii.2025.100955","DOIUrl":"10.1016/j.jii.2025.100955","url":null,"abstract":"<div><div>Embodied intelligence (EI) enables manufacturing systems to flexibly perceive, reason, adapt, and operate within dynamic shop floor environments. In smart manufacturing, a representative EI scenario is <strong>robotic visual inspection</strong>, where industrial robots must accurately inspect components on rapidly changing, heterogeneous production lines. This task requires both high inference accuracy — especially for uncommon defects — and low latency to match production speeds, despite evolving lighting, part geometries, and surface conditions. To meet these needs, we propose <strong>LAECIPS</strong>, a large vision model-assisted adaptive edge–cloud collaboration framework for IoT-based embodied intelligence systems. LAECIPS decouples large vision models in the cloud from lightweight models on the edge, enabling flexible model deployment and continual learning (automated model updates). Through identifying complex inspection cases, LAECIPS routes complex and uncertain inspection cases to the cloud while handling routine tasks at the edge, achieving both high accuracy and low latency. Experiments conducted on a real-world robotic semantic segmentation system for visual inspection demonstrate significant improvements in accuracy, processing latency, and communication overhead compared to state-of-the-art methods. From an industrial information integration perspective, LAECiPS operationalizes a complete edge–cloud information loop for smart manufacturing: integrating multi-source perception data at the edge, adaptively routing information to the cloud for large model assistance, and feeding back distilled knowledge to the edge for continual adaptation. This layered integration improves both accuracy and task-time-constrained latency, aligning with the focus on interoperable and scalable industrial information integration.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"48 ","pages":"Article 100955"},"PeriodicalIF":10.4,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145121270","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}