Pub Date : 2026-01-08DOI: 10.1016/j.jii.2026.101064
Bruno T. Guedes , Diego C. Fettermann , Karl A. Hribernik , Klaus-Dieter Thoben
Digital transformation is driving significant changes across industries by integrating information, computing, communication, and connectivity technologies. Product development stands as a crucial process that provides organizations with a competitive edge by translating ideas or business opportunities into profitable outcomes. Considering both of these aspects, this research addresses the main question: how can digital transformation contribute to improving the execution of activities established in the product development process? This research addresses how digital transformation can improve the execution of activities established in the product development process. It aims to identify and complement the contributions of digital transformation to product development activities, offering a holistic perspective supported by practical evidence. The study employs a mixed-method approach and outlines future research directions for different thematic clusters related to digital transformation. The digital transformation was classified into seven thematic clusters: Mass Customization, Digital Twin, Smart Manufacturing, Digital Manufacturing, Digital Innovation, Digital Servitization, and Circular Economy. The findings emphasize the significance of integrating digital transformation technologies into product development and offer insights for practitioners and researchers.
{"title":"How digital transformation is changing product development: A comprehensive analysis","authors":"Bruno T. Guedes , Diego C. Fettermann , Karl A. Hribernik , Klaus-Dieter Thoben","doi":"10.1016/j.jii.2026.101064","DOIUrl":"10.1016/j.jii.2026.101064","url":null,"abstract":"<div><div>Digital transformation is driving significant changes across industries by integrating information, computing, communication, and connectivity technologies. Product development stands as a crucial process that provides organizations with a competitive edge by translating ideas or business opportunities into profitable outcomes. Considering both of these aspects, this research addresses the main question: how can digital transformation contribute to improving the execution of activities established in the product development process? This research addresses how digital transformation can improve the execution of activities established in the product development process. It aims to identify and complement the contributions of digital transformation to product development activities, offering a holistic perspective supported by practical evidence. The study employs a mixed-method approach and outlines future research directions for different thematic clusters related to digital transformation. The digital transformation was classified into seven thematic clusters: Mass Customization, Digital Twin, Smart Manufacturing, Digital Manufacturing, Digital Innovation, Digital Servitization, and Circular Economy. The findings emphasize the significance of integrating digital transformation technologies into product development and offer insights for practitioners and researchers.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101064"},"PeriodicalIF":10.4,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145957297","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}
Pub Date : 2026-01-08DOI: 10.1016/j.jii.2026.101062
Min-Ho Han , Keun-Woo Lim , Young-Bae Ko
Industrial digital twins (DTs) must fuse data from operational technology (OT) and information technology (IT) platforms in real time. However, the high-frequency ultra-wideband (UWB) sampling needed for real-time fidelity can rapidly drain battery-powered tags, increasing battery-replacement and maintenance burden in large-scale deployments and jeopardizing service-level accuracy. To address this energy-accuracy trade-off, this paper defines mobility-entropy, a three-dimensional metric that quantifies the dynamic characteristics of a mobile entity. A lightweight on-device machine learning scheduler uses this metric to adjust the UWB sampling rate in real time across the end-to-end pipeline from sensor to DT renderer. Evaluated on a seven-anchor indoor testbed mirrored in real time on the MuJoCo DT platform, the proposed approach extends the average tag sleep time by 65.6% compared to a fixed-rate baseline while achieving a Digital Twin Projection Error (DTPE) as low as 3.15 cm across various mobility environments. The result is longer battery life and reduced telemetry data volume without sacrificing geometric accuracy, improving deployment practicality by lowering maintenance overhead and wireless traffic in industrial settings. We also explain how edge decisions are propagated through the integration layer to DT applications, positioning adaptive sensing within the operational technology (OT) to information technology (IT) to digital twin (DT) data flow. These results highlight the framework’s potential for real-world industrial digital twin applications, including worker and asset tracking as well as safety monitoring, by enabling energy-efficient operation with reduced maintenance and communication overhead.
{"title":"Mobility-entropy–aware adaptive UWB sampling for energy-efficient real-time industrial digital twins","authors":"Min-Ho Han , Keun-Woo Lim , Young-Bae Ko","doi":"10.1016/j.jii.2026.101062","DOIUrl":"10.1016/j.jii.2026.101062","url":null,"abstract":"<div><div>Industrial digital twins (DTs) must fuse data from operational technology (OT) and information technology (IT) platforms in real time. However, the high-frequency ultra-wideband (UWB) sampling needed for real-time fidelity can rapidly drain battery-powered tags, increasing battery-replacement and maintenance burden in large-scale deployments and jeopardizing service-level accuracy. To address this energy-accuracy trade-off, this paper defines <em>mobility-entropy</em>, a three-dimensional metric that quantifies the dynamic characteristics of a mobile entity. A lightweight on-device machine learning scheduler uses this metric to adjust the UWB sampling rate in real time across the end-to-end pipeline from sensor to DT renderer. Evaluated on a seven-anchor indoor testbed mirrored in real time on the MuJoCo DT platform, the proposed approach extends the average tag sleep time by <strong>65.6%</strong> compared to a fixed-rate baseline while achieving a Digital Twin Projection Error (DTPE) as low as <strong>3.15<!--> <!-->cm</strong> across various mobility environments. The result is longer battery life and reduced telemetry data volume without sacrificing geometric accuracy, improving deployment practicality by lowering maintenance overhead and wireless traffic in industrial settings. We also explain how edge decisions are propagated through the integration layer to DT applications, positioning adaptive sensing within the operational technology (OT) to information technology (IT) to digital twin (DT) data flow. These results highlight the framework’s potential for real-world industrial digital twin applications, including worker and asset tracking as well as safety monitoring, by enabling energy-efficient operation with reduced maintenance and communication overhead.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101062"},"PeriodicalIF":10.4,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145957298","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}
A Digital Twin requires a user interface to deliver information relevant to its users, hence a model is required to represent the information required by the interface. The objective of this research is to develop a transdisciplinary human factors approach to information gathering and modelling to design Digital Twin information interfaces. Existing approaches to interface modelling either do not consider human factors or those that do provide a high-level view of information insufficient to capture the complexities required for an information interface for a Digital Twin. The approach presented here consists of capturing the information interface requirements using Cognitive Work Analysis to analyse the human-information interaction and structuring this information via Unified Modelling Language (UML) models. To understand human information requirements when interacting with a Digital Twin interface, personas are used to guide the CWA. To illustrate this approach a Digital Twin of an Industrial Gearbox Product-Service is considered. Validation was conducted through a case study with a research and technology organisation. The approach was found to be clear and able to provide information customised to user needs and the level of detail required. The research described creates a more effective approach to creating a Digital Twin information interface model through reducing the number of iterations required to gather information. By specifically considering human interactions the transdisciplinary approach advanced here will augment the development of software systems.
{"title":"A human factors approach to design an information interface model for a digital twin","authors":"Claire Palmer , Ella-Mae Hubbard , Rebecca Grant , Yee Mey Goh","doi":"10.1016/j.jii.2026.101063","DOIUrl":"10.1016/j.jii.2026.101063","url":null,"abstract":"<div><div>A Digital Twin requires a user interface to deliver information relevant to its users, hence a model is required to represent the information required by the interface. The objective of this research is to develop a transdisciplinary human factors approach to information gathering and modelling to design Digital Twin information interfaces. Existing approaches to interface modelling either do not consider human factors or those that do provide a high-level view of information insufficient to capture the complexities required for an information interface for a Digital Twin. The approach presented here consists of capturing the information interface requirements using Cognitive Work Analysis to analyse the human-information interaction and structuring this information via Unified Modelling Language (UML) models. To understand human information requirements when interacting with a Digital Twin interface, personas are used to guide the CWA. To illustrate this approach a Digital Twin of an Industrial Gearbox Product-Service is considered. Validation was conducted through a case study with a research and technology organisation. The approach was found to be clear and able to provide information customised to user needs and the level of detail required. The research described creates a more effective approach to creating a Digital Twin information interface model through reducing the number of iterations required to gather information. By specifically considering human interactions the transdisciplinary approach advanced here will augment the development of software systems.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101063"},"PeriodicalIF":10.4,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145957299","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}
Pub Date : 2026-01-01DOI: 10.1016/j.jii.2025.101038
Didem Gurdur Broo
The rapid advancement of digital artificial intelligence has created unrealistic expectations about its transfer to physical industrial systems. This commentary critically examines the fundamental misalignment between digital AI capabilities and the complex requirements of industrial cyber-physical systems. While digital AI excels in pattern recognition and virtual environments, physical intelligence demands understanding of mechanics, materials, energy, and real-world constraints that current AI paradigms inadequately address. The commentary argues that achieving genuine physical intelligence in industrial settings requires a fundamental reorientation toward information integration as the enabling foundation, rather than pursuing ever-larger foundation models. Industrial information integration frameworks must bridge cyber-physical boundaries, handle temporal characteristics properly, represent uncertainty explicitly, and enable human-AI collaboration. This perspective aims to redirect research efforts toward the critical challenges of industrial information integration that will ultimately enable meaningful progress in physical AI for cyber-physical systems.
{"title":"Physical AI in cyber-physical systems: from digital to embodied industrial agents","authors":"Didem Gurdur Broo","doi":"10.1016/j.jii.2025.101038","DOIUrl":"10.1016/j.jii.2025.101038","url":null,"abstract":"<div><div>The rapid advancement of digital artificial intelligence has created unrealistic expectations about its transfer to physical industrial systems. This commentary critically examines the fundamental misalignment between digital AI capabilities and the complex requirements of industrial cyber-physical systems. While digital AI excels in pattern recognition and virtual environments, physical intelligence demands understanding of mechanics, materials, energy, and real-world constraints that current AI paradigms inadequately address. The commentary argues that achieving genuine physical intelligence in industrial settings requires a fundamental reorientation toward information integration as the enabling foundation, rather than pursuing ever-larger foundation models. Industrial information integration frameworks must bridge cyber-physical boundaries, handle temporal characteristics properly, represent uncertainty explicitly, and enable human-AI collaboration. This perspective aims to redirect research efforts toward the critical challenges of industrial information integration that will ultimately enable meaningful progress in physical AI for cyber-physical systems.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"49 ","pages":"Article 101038"},"PeriodicalIF":10.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145785017","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}
Pub Date : 2025-12-31DOI: 10.1016/j.jii.2025.101043
Ada Bagozi, Devis Bianchini, Massimiliano Garda, Michele Melchiori, Anisa Rula
In modern smart factories, supply chains are no longer isolated; instead, they are evolving into interconnected and dynamic networks, where intertwined supply chains enable real-time collaboration and data sharing for adaptive decision-making across multiple stakeholders. By harnessing data from sensors and connected devices, data-driven decisions can be made to optimize the entire supply chain, and to provide novel and customer-friendly products and services. Cyber-Physical Systems form the foundation of Cyber-Physical Production Systems (CPPS) by enabling real-time data exchange and intelligent automation at the factory level, while horizontal integration connects CPPS across different production facilities to enhance supply chain coordination, thus forming the so-called Cyber-Physical Production Networks (CPPN). In CPPN, the Internet of Services (IoS) paradigm, in combination with the Internet of Things (IoT), plays a crucial role in facilitating horizontal integration and seamless collaboration between intertwined supply chains. Since the IoS paradigm has to enable data sharing and processing within individual smart factories and across factory borders, there is a need to design service-oriented architectures specifically tailored to data governance in both CPPS and CPPN. However, existing service-oriented approaches for CPPS primarily focus on deployment layers (e.g., fog/edge computing or IT/production levels) while neglecting data-oriented aspects, limiting modularity and effective data service design across CPPS and CPPN levels. To bridge this gap, in this paper, we propose a multi-layered service-oriented model for CPPN focused on data services, which includes atomic services for data collection and processing, and composite services for governing the data flow within smart factories and throughout the supply chains they participate in. One of the significant advantages of the multi-layered approach is a clear separation of concerns in service design, with the ability to address issues of modularity, scalability, data sovereignty and data access, by distinguishing between CPPS and CPPN levels. In the paper, we critically evaluate different strategies for the management of a service ecosystem that is compliant with the proposed model.
{"title":"A multi-layered data service model for Cyber-Physical Production Networks","authors":"Ada Bagozi, Devis Bianchini, Massimiliano Garda, Michele Melchiori, Anisa Rula","doi":"10.1016/j.jii.2025.101043","DOIUrl":"10.1016/j.jii.2025.101043","url":null,"abstract":"<div><div>In modern smart factories, supply chains are no longer isolated; instead, they are evolving into interconnected and dynamic networks, where intertwined supply chains enable real-time collaboration and data sharing for adaptive decision-making across multiple stakeholders. By harnessing data from sensors and connected devices, data-driven decisions can be made to optimize the entire supply chain, and to provide novel and customer-friendly products and services. Cyber-Physical Systems form the foundation of Cyber-Physical Production Systems (CPPS) by enabling real-time data exchange and intelligent automation at the factory level, while horizontal integration connects CPPS across different production facilities to enhance supply chain coordination, thus forming the so-called Cyber-Physical Production Networks (CPPN). In CPPN, the Internet of Services (IoS) paradigm, in combination with the Internet of Things (IoT), plays a crucial role in facilitating horizontal integration and seamless collaboration between intertwined supply chains. Since the IoS paradigm has to enable data sharing and processing within individual smart factories and across factory borders, there is a need to design service-oriented architectures specifically tailored to data governance in both CPPS and CPPN. However, existing service-oriented approaches for CPPS primarily focus on deployment layers (e.g., fog/edge computing or IT/production levels) while neglecting data-oriented aspects, limiting modularity and effective data service design across CPPS and CPPN levels. To bridge this gap, in this paper, we propose a multi-layered service-oriented model for CPPN focused on data services, which includes atomic services for data collection and processing, and composite services for governing the data flow within smart factories and throughout the supply chains they participate in. One of the significant advantages of the multi-layered approach is a clear separation of concerns in service design, with the ability to address issues of modularity, scalability, data sovereignty and data access, by distinguishing between CPPS and CPPN levels. In the paper, we critically evaluate different strategies for the management of a service ecosystem that is compliant with the proposed model.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101043"},"PeriodicalIF":10.4,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883956","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}
Pub Date : 2025-12-30DOI: 10.1016/j.jii.2025.101047
Fei Gao
Unmanned aerial vehicles (UAVs) have garnered increasing attention due to their efficiency, cost-effectiveness, and performance, leading to various efforts to implement UAVs in agriculture, especially with the rapid development of low-altitude economy. However, successful UAV application in agriculture is not always achieved, and understanding the barriers and potential solutions is crucial for effective implementation. To this end, this study employs intuitionistic fuzzy sets, the modified Delphi method, the fuzzy weight with zero consistency (FWZIC) method, and the weighted aggregated sum product assessment (WASPAS) method to identify and prioritize barriers and solutions for UAV application in agriculture. Firstly, 32 barriers are identified and categorized into five main categories. The intuitionistic fuzzy FWZIC method is then utilized to calculate weights for prioritizing the barriers. Subsequently, the intuitionistic fuzzy WASPAS method is applied to assess and rank solutions for these barriers. The results indicate that “risk of failures” is the most significant sub-barrier hindering UAV application in agriculture. Additionally, “design and prompt more reliable UAV technologies” is the most effective solution for mitigating these barriers. This study provides a systematic framework to address barriers to UAV application in agriculture, and the findings can assist practitioners by guiding their efforts toward overcoming the most significant barriers and facilitating successful UAV application in agriculture.
{"title":"Prioritizing and overcoming barriers to unmanned aerial vehicles adoption in agriculture using an integrated intuitionistic fuzzy decision-making approach","authors":"Fei Gao","doi":"10.1016/j.jii.2025.101047","DOIUrl":"10.1016/j.jii.2025.101047","url":null,"abstract":"<div><div>Unmanned aerial vehicles (UAVs) have garnered increasing attention due to their efficiency, cost-effectiveness, and performance, leading to various efforts to implement UAVs in agriculture, especially with the rapid development of low-altitude economy. However, successful UAV application in agriculture is not always achieved, and understanding the barriers and potential solutions is crucial for effective implementation. To this end, this study employs intuitionistic fuzzy sets, the modified Delphi method, the fuzzy weight with zero consistency (FWZIC) method, and the weighted aggregated sum product assessment (WASPAS) method to identify and prioritize barriers and solutions for UAV application in agriculture. Firstly, 32 barriers are identified and categorized into five main categories. The intuitionistic fuzzy FWZIC method is then utilized to calculate weights for prioritizing the barriers. Subsequently, the intuitionistic fuzzy WASPAS method is applied to assess and rank solutions for these barriers. The results indicate that “risk of failures” is the most significant sub-barrier hindering UAV application in agriculture. Additionally, “design and prompt more reliable UAV technologies” is the most effective solution for mitigating these barriers. This study provides a systematic framework to address barriers to UAV application in agriculture, and the findings can assist practitioners by guiding their efforts toward overcoming the most significant barriers and facilitating successful UAV application in agriculture.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101047"},"PeriodicalIF":10.4,"publicationDate":"2025-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145883955","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}
Pub Date : 2025-12-27DOI: 10.1016/j.jii.2025.101052
Daniel M. Jimenez-Gutierrez, Mehrdad Hassanzadeh, Aris Anagnostopoulos, Ioannis Chatzigiannakis, Andrea Vitaletti
Federated learning (FL) allows collaborative machine learning (ML) model training among decentralized clients’ information, ensuring data privacy. The decentralized nature of FL deals with non-independent and identically distributed (non-IID) data. This open problem has notable consequences, such as decreased model performance and longer convergence times. Despite its importance, experimental studies systematically addressing all types of data heterogeneity (a.k.a. non-IIDness) remain scarce. This paper aims to fill this gap by assessing and quantifying the non-IID effect through an empirical analysis. We use the Hellinger Distance (HD) to measure differences in distribution among clients. Our study benchmarks five state-of-the-art strategies for handling non-IID data, including label, feature, quantity, and spatiotemporal skews, under realistic and controlled conditions. This is the first comprehensive analysis of the spatiotemporal skew effect in FL. Our findings highlight the significant impact of label and spatiotemporal skew non-IID types on FL model performance, with notable performance drops occurring at specific HD thresholds. The FL performance is also heavily affected, mainly when the non-IIDness is extreme. Thus, we provide recommendations for FL research to tackle data heterogeneity effectively. Our work represents the most extensive examination of non-IIDness in FL, offering a robust foundation for future research.
{"title":"A thorough assessment of the non-IID data impact in federated learning","authors":"Daniel M. Jimenez-Gutierrez, Mehrdad Hassanzadeh, Aris Anagnostopoulos, Ioannis Chatzigiannakis, Andrea Vitaletti","doi":"10.1016/j.jii.2025.101052","DOIUrl":"10.1016/j.jii.2025.101052","url":null,"abstract":"<div><div>Federated learning (FL) allows collaborative machine learning (ML) model training among decentralized clients’ information, ensuring data privacy. The decentralized nature of FL deals with non-independent and identically distributed (non-IID) data. This open problem has notable consequences, such as decreased model performance and longer convergence times. Despite its importance, experimental studies systematically addressing all types of data heterogeneity (a.k.a. non-IIDness) remain scarce. This paper aims to fill this gap by assessing and quantifying the non-IID effect through an empirical analysis. We use the Hellinger Distance (<span>HD</span>) to measure differences in distribution among clients. Our study benchmarks five state-of-the-art strategies for handling non-IID data, including label, feature, quantity, and spatiotemporal skews, under realistic and controlled conditions. This is the first comprehensive analysis of the spatiotemporal skew effect in FL. Our findings highlight the significant impact of label and spatiotemporal skew non-IID types on FL model performance, with notable performance drops occurring at specific <span>HD</span> thresholds. The FL performance is also heavily affected, mainly when the non-IIDness is extreme. Thus, we provide recommendations for FL research to tackle data heterogeneity effectively. Our work represents the most extensive examination of non-IIDness in FL, offering a robust foundation for future research.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101052"},"PeriodicalIF":10.4,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145845120","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}
Pub Date : 2025-12-27DOI: 10.1016/j.jii.2025.101050
Jing Long , Jiahao Zeng , Zhifei Yan , Min Shi , Kun Xie , Meng Shen , Naixue Xiong
In smart manufacturing systems, interconnected systems composed of equipment nodes such as intelligent machine tools and sensors can be abstracted as attributed networks. However, such networks are vulnerable to security risks like cyber-attacks and equipment failures, which directly threaten the stable operation of smart manufacturing systems. In the unsupervised setting, existing anomaly detection models intrinsically lean towards fitting the overwhelming majority of normal patterns during training. However, they cannot escape being influenced by anomalous characteristics, which degrades the detection of anomaly patterns in smart manufacturing environments. To address this challenge, this paper proposes a novel anomaly detection method called ARENA for attributed networks in smart manufacturing, which adopts two-stage dynamic reconstruction bias learning. Firstly, a graph autoencoder uncovers latent data patterns in smart manufacturing scenarios by minimizing reconstruction error. Then, the dynamic reconstruction biased learning module adjusts the training process in two stages to filter out pseudo-normal nodes and pseudo-anomalous nodes, enabling the model to adaptively fine-tune, mitigating the impact of anomalous data during training. Finally, the classification module further amplifies the anomaly score, making abnormal patterns more pronounced and easier to detect. The overall anomaly score is calculated by combining the results of the graph reconstruction and classification modules. Experimental results show that the ARENA method significantly improves performance, with an increase of 3.73% in AUC and 21.1% in AUPRC, including the success of the case study, providing strong support for the intelligent operation and maintenance of equipment in industrial manufacturing systems.
{"title":"Two-stage dynamic reconstruction biased learning for anomaly detection in attributed networks of smart manufacturing","authors":"Jing Long , Jiahao Zeng , Zhifei Yan , Min Shi , Kun Xie , Meng Shen , Naixue Xiong","doi":"10.1016/j.jii.2025.101050","DOIUrl":"10.1016/j.jii.2025.101050","url":null,"abstract":"<div><div>In smart manufacturing systems, interconnected systems composed of equipment nodes such as intelligent machine tools and sensors can be abstracted as attributed networks. However, such networks are vulnerable to security risks like cyber-attacks and equipment failures, which directly threaten the stable operation of smart manufacturing systems. In the unsupervised setting, existing anomaly detection models intrinsically lean towards fitting the overwhelming majority of normal patterns during training. However, they cannot escape being influenced by anomalous characteristics, which degrades the detection of anomaly patterns in smart manufacturing environments. To address this challenge, this paper proposes a novel anomaly detection method called ARENA for attributed networks in smart manufacturing, which adopts two-stage dynamic reconstruction bias learning. Firstly, a graph autoencoder uncovers latent data patterns in smart manufacturing scenarios by minimizing reconstruction error. Then, the dynamic reconstruction biased learning module adjusts the training process in two stages to filter out pseudo-normal nodes and pseudo-anomalous nodes, enabling the model to adaptively fine-tune, mitigating the impact of anomalous data during training. Finally, the classification module further amplifies the anomaly score, making abnormal patterns more pronounced and easier to detect. The overall anomaly score is calculated by combining the results of the graph reconstruction and classification modules. Experimental results show that the ARENA method significantly improves performance, with an increase of 3.73% in AUC and 21.1% in AUPRC, including the success of the case study, providing strong support for the intelligent operation and maintenance of equipment in industrial manufacturing systems.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101050"},"PeriodicalIF":10.4,"publicationDate":"2025-12-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145845506","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}
Pub Date : 2025-12-25DOI: 10.1016/j.jii.2025.101048
Zili Wang , Xinlei Hu , Shuyou Zhang , Lemiao Qiu , Yaochen Lin , Liangyou Li , Yongzhe Xiang , Jie Li
During the metal tube bending (MTB) process, high-fidelity reconstruction of full cross-section (FCS) deformation is critical to the robustness of closed-loop control in tube-bending manufacturing systems. However, the distributed nature of industrial data, the spatiotemporal discontinuity of physical sensing, and the heterogeneity of multimodal physical–virtual data hinder effective integration of distributed sources and precise reconstruction of the transient deformation of tube surfaces. To address these challenges, we propose a Federated Split-Learning–Driven Multimodal Physical–Virtual Integration (FSLD-MPVI) framework. Leveraging a hybrid distributed–centralized architecture with cross-level collaborative fusion, FSLD-MPVI enables efficient integration and knowledge sharing of local high-fidelity visual data, global low-fidelity finite-element (FE) simulation data, and static process parameters that are dispersed across manufacturing nodes. Within the split learning (SL) distributed architecture, three cascaded, heterogeneous subnetworks are deployed, each dedicated to fusing a specific class of hybrid modality inputs, thereby providing the infrastructure needed to integrate modalities originating from different workshops. In the federated learning (FL) layer, a centralized server aggregates the parameters of each subnetwork respectively, mitigating cross-node data isolation while preserving data locality. Experiments demonstrate that FSLD-MPVI achieves high-accuracy global reconstruction (R² = 0.9973); in the 90° bending case, the shape deviation remains within 0.2 mm. These results verify that multimodal physical–virtual integration strongly supports precise global reconstruction of FCS deformation fields and establishes a new paradigm for intelligent process monitoring in advanced manufacturing systems.
{"title":"Federated split learning-driven multimodal physical-virtual integration framework: high-fidelity full-cross-section deformation field reconstruction in precise metal tube bending manufacturing","authors":"Zili Wang , Xinlei Hu , Shuyou Zhang , Lemiao Qiu , Yaochen Lin , Liangyou Li , Yongzhe Xiang , Jie Li","doi":"10.1016/j.jii.2025.101048","DOIUrl":"10.1016/j.jii.2025.101048","url":null,"abstract":"<div><div>During the metal tube bending (MTB) process, high-fidelity reconstruction of full cross-section (FCS) deformation is critical to the robustness of closed-loop control in tube-bending manufacturing systems. However, the distributed nature of industrial data, the spatiotemporal discontinuity of physical sensing, and the heterogeneity of multimodal physical–virtual data hinder effective integration of distributed sources and precise reconstruction of the transient deformation of tube surfaces. To address these challenges, we propose a Federated Split-Learning–Driven Multimodal Physical–Virtual Integration (FSLD-MPVI) framework. Leveraging a hybrid distributed–centralized architecture with cross-level collaborative fusion, FSLD-MPVI enables efficient integration and knowledge sharing of local high-fidelity visual data, global low-fidelity finite-element (FE) simulation data, and static process parameters that are dispersed across manufacturing nodes. Within the split learning (SL) distributed architecture, three cascaded, heterogeneous subnetworks are deployed, each dedicated to fusing a specific class of hybrid modality inputs, thereby providing the infrastructure needed to integrate modalities originating from different workshops. In the federated learning (FL) layer, a centralized server aggregates the parameters of each subnetwork respectively, mitigating cross-node data isolation while preserving data locality. Experiments demonstrate that FSLD-MPVI achieves high-accuracy global reconstruction (R² = 0.9973); in the 90° bending case, the shape deviation remains within 0.2 mm. These results verify that multimodal physical–virtual integration strongly supports precise global reconstruction of FCS deformation fields and establishes a new paradigm for intelligent process monitoring in advanced manufacturing systems.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101048"},"PeriodicalIF":10.4,"publicationDate":"2025-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145845121","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}
The air-ground cooperative plant protection unmanned formation can effectively deal with the complex terrain challenges of mountain orchards and ensure the uniformity of plant protection operation coverage. The core of this system lies in the principles of Industrial Information Integration Engineering (IIIE). Through dynamic scheduling optimization, it can alleviate the problems of large energy consumption and long non-operation paths. Aiming at the dynamic scheduling planning problem, this study proposes an energy-saving hybrid target scheduling optimization method based on an improved Australian wild dog hunting strategy. A novel mountain orchard path coding technology is designed, and an energy consumption model based on the principle of unmanned formation dynamics is established, which provides a scientific basis for formulating efficient energy-saving strategies. The improved Australian wild dog hunting strategy combines the motion constraints of unmanned formation and the requirements of plant protection tasks, and realizes the efficient optimization of the scheduling scheme. Numerical experiments demonstrated the effectiveness of the proposed method, which reduced the objective function to 65.63% of the initial solution in simulations, outperforming the genetic algorithm. This performance was further validated in a real-world scenario, where the value was reduced to 57.34%. This efficient dynamic scheduling optimization serves as a key enabler for agricultural industry integration and informatization.
{"title":"Mixed objective scheduling optimization in mountain orchards under energy-saving for carbon neutrality","authors":"Zhentao Xue , Zhigang Ren , Jian Chen , Xiqing Wang , Shuaisong Zhang","doi":"10.1016/j.jii.2025.101049","DOIUrl":"10.1016/j.jii.2025.101049","url":null,"abstract":"<div><div>The air-ground cooperative plant protection unmanned formation can effectively deal with the complex terrain challenges of mountain orchards and ensure the uniformity of plant protection operation coverage. The core of this system lies in the principles of Industrial Information Integration Engineering (IIIE). Through dynamic scheduling optimization, it can alleviate the problems of large energy consumption and long non-operation paths. Aiming at the dynamic scheduling planning problem, this study proposes an energy-saving hybrid target scheduling optimization method based on an improved Australian wild dog hunting strategy. A novel mountain orchard path coding technology is designed, and an energy consumption model based on the principle of unmanned formation dynamics is established, which provides a scientific basis for formulating efficient energy-saving strategies. The improved Australian wild dog hunting strategy combines the motion constraints of unmanned formation and the requirements of plant protection tasks, and realizes the efficient optimization of the scheduling scheme. Numerical experiments demonstrated the effectiveness of the proposed method, which reduced the objective function to 65.63% of the initial solution in simulations, outperforming the genetic algorithm. This performance was further validated in a real-world scenario, where the value was reduced to 57.34%. This efficient dynamic scheduling optimization serves as a key enabler for agricultural industry integration and informatization.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"50 ","pages":"Article 101049"},"PeriodicalIF":10.4,"publicationDate":"2025-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145845508","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号