Journal of Industrial Information Integration最新文献

{"title":"Digital twin-driven self-adaptive reconfiguration planning method of smart manufacturing systems using game theory and deep Q-network for industry 5.0","authors":"Sihan Huang ,&nbsp;Guangyu Mo ,&nbsp;Shikai Jing ,&nbsp;Jiewu Leng ,&nbsp;Xingyu Li ,&nbsp;Xi Gu ,&nbsp;Yan Yan ,&nbsp;Guoxin Wang","doi":"10.1016/j.jii.2025.100901","DOIUrl":"10.1016/j.jii.2025.100901","url":null,"abstract":"<div><div>In the Industry 5.0 era, as market demand shifts to personalization, smart manufacturing systems (SMS) with the rapid, accurate, responsive and resilient are becoming increasingly critical. To address the reconfiguration problem of SMS due to the dynamic production tasks, a digital twin-driven self-adaptive reconfiguration planning method of SMS is proposed by integrating game theory and deep reinforcement learning (DRL). Firstly, digital twin- driven self-adaptive framework for SMS is proposed to perceive production task changes for dynamically optimizing reconfiguration processes of SMS efficiently. Secondly, game theory is adopted to model the dynamic reconfiguration processes of SMS composed of multi-level reconfiguration, including system level, cell level, and machine level, where virtual manufacturing cells (VMC) as game entities will play games to reach Nash equilibrium by selecting appropriate reconfigurable machine tools (RMT) according to the proposed game strategy and utility function. Thirdly, due to the complexity of the game processes, a DRL algorithm named as deep Q-network (DQN) is used to execute the reconfiguration game for finding the optimal reconfiguration scheme to enhance the resilience of SMS. Finally, a case study is presented to demonstrate the effectiveness and adaptability of the proposed method.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100901"},"PeriodicalIF":10.4,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144566623","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":"Humanitarian relief supply chain performance measurement: A framework and validation","authors":"Humaira Nafisa Ahmed ,&nbsp;Zannatul Maoua ,&nbsp;Sayem Ahmed ,&nbsp;Syed Mithun Ali","doi":"10.1016/j.jii.2025.100898","DOIUrl":"10.1016/j.jii.2025.100898","url":null,"abstract":"<div><div>The concept of humanitarian relief supply chain management has gained a lot of interest among academics and practitioners since the number of natural or human-made disasters has increased drastically. Humanitarian organizations assist in disaster relief operations by planning, sourcing, procuring, transporting, and distributing essential goods and services during emergency operations, known as the humanitarian relief supply chain. This research develops a Bayesian belief network-based framework for predicting the performance of the humanitarian relief supply chain in case of catastrophic events, such as natural disasters and man-made crises. The study begins with identifying performance metrics through factor analysis that directly or indirectly affect the overall performance of a humanitarian organization. Then, with the aid of a Bayesian belief network, a probabilistic graphical model capable of predicting any organization's relief supply chain based on performance metrics was developed. The model demonstrates the interdependencies among the performance metrics within a network setting. The network is constructed through mediating variables by establishing causal relationships among performance metrics and mediating variables. The model has been validated through numerical examples, extreme condition testing, scenario analysis, sensitivity analysis, and diagnostics analysis. Extreme condition tests, diagnostic, and scenario analysis validate the model as reliable and stable. The sensitivity analysis result shows financial performance and monetary support as crucial factors in measuring the performance of the humanitarian relief supply chain. The performance measurement model will assist organizations' decision-makers and policymakers in controlling, monitoring, and enhancing their relief supply chain.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100898"},"PeriodicalIF":10.4,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144503973","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":"Assessment and recognition of driver situation awareness in conditional autonomous driving: Integrating cognitive psychology and machine learning","authors":"Jing Huang ,&nbsp;Tingnan Liu ,&nbsp;Yezi Hu ,&nbsp;Zhipeng He ,&nbsp;Lin Hu","doi":"10.1016/j.jii.2025.100894","DOIUrl":"10.1016/j.jii.2025.100894","url":null,"abstract":"<div><div>Driver Situation Awareness (SA) is crucial for the safety of conditional autonomous driving, posing a significant human factors challenge for the automotive industry in its pursuit of intelligent driving systems. To guide future automotive system design and enhance Human–Machine Interaction (HMI), the present paper presents an interdisciplinary solution that achieves the assessment and recognition of driver SA by deeply integrating cognitive psychology theories and machine learning techniques. First, a takeover driving study was designed and conducted during the conditional autonomous driving phase. Various physiological and behavioral data were collected, along with information from measurement questionnaires. Next, a driver quantitative SA model was developed based on ACT-R and the actual allocation of the driver’s visual attention, accounting for both goal-directed and data-directed processing. For a comprehensive SA assessment, the results from this model and the questionnaires were combined and integrated as composite features. Using the K-means clustering algorithm and the silhouette coefficient method, the assessment of driver SA is achieved by clustering driver SA into two levels: low SA and high SA, overcoming the limitations of single information sources. Furthermore, by fusing physiological and behavioral data as well as visual attention levels as multimodal features, machine learning classifiers were employed for real-time recognition of driver SA, achieving the highest Accuracy of 93.29%. Finally, the validity of the experimental design was confirmed through RM-ANOVA analysis of the experimental conditions and SA indicators, and the effectiveness of the quantitative model was validated via correlation analysis between the model’s results and questionnaire outcomes. This research provides an innovative solution for driver SA assessment and recognition, offering valuable insights for developing safer HMI in intelligent vehicles and advancing industrial information integration within the automotive sector.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100894"},"PeriodicalIF":10.4,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144480190","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":"A physics-informed machine learning approach for temperature field prediction in metallic additive manufacturing","authors":"Zhonghao Chen ,&nbsp;Haochen Mu ,&nbsp;Fengyang He ,&nbsp;Lei Yuan ,&nbsp;Hongtao Zhu ,&nbsp;Ninshu Ma ,&nbsp;Zengxi Pan","doi":"10.1016/j.jii.2025.100899","DOIUrl":"10.1016/j.jii.2025.100899","url":null,"abstract":"<div><div>Advancements in Machine Learning (ML) have provided an efficient solution for elucidating the relationships of process-structure-property in the field of metallic Additive Manufacturing (AM). Yet, the reliance on data-driven ML imposes impediments in terms of model interpretability and flexibility. This study addresses these limitations by providing a physics-informed ML strategy for thermal modelling in metallic AM. The models were developed to employ a physics-informed 3D convolutional autoencoder, integrating the physical mechanisms of heat transfer in latent space, to simulate dynamic thermal patterns during cooling and deposition phases of wire arc additive manufacturing. The physics-informed ML model was trained by finite element method thermal simulations and the performance of the developed model was evaluated through an ablation study by comparing it with two additional ML strategies. One thin-wall and three cubic structures with different deposition paths were built experimentally and modelled to test the model’s performance. Results demonstrate that the proposed physics-informed ML model has shown a rapid convergence rate and high accuracy in predicting temperature fields during the metallic AM processes, with a root mean square error within 2 degrees in the cooling phase and 60 degrees in the deposition phase. The introduction of physical constraints into the latent space can significantly enhance feature recognition and flexibility, thereby fostering a more meaningful physical convergence.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100899"},"PeriodicalIF":10.4,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144503972","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":"Sustainable supply chain network design: Integrating risk management, resilient multimodal transportation, and production strategy","authors":"Seyed Mahameddin Tabatabaei","doi":"10.1016/j.jii.2025.100897","DOIUrl":"10.1016/j.jii.2025.100897","url":null,"abstract":"<div><div>This study presents a novel advancement in sustainable supply chain network design (SSCND) by incorporating risk management, resilience, and production strategies within a multi-modal transportation framework. Focusing on the distribution, production, and inventory (DPI) triad, the literature on the SSCND emphasizes the need for strategic alignment to create a resilient supply chain capable of mitigating risks in multimodal transport. To achieve this, we develop a novel multi-objective mixed-integer programming (MOMIP) model customized to the SSCND aimed at maximizing profit, minimizing transportation time, and reducing environmental impacts. The model is solved using a specialized goal programming approach, ensuring that no objective is compromised at the expense of others. A hybrid solution methodology, combining a local search algorithm with machine learning predictive models, is introduced to navigate the complexity of the MOMIP model efficiently. The model’s validity is confirmed through real-world data from the Iranian chemicals industry, and the proposed algorithm’s performance is tested. On average, the algorithm achieves an optimality gap of &lt;3 %, with a gap of 2.67 % for profit maximization, 1.63 % for transportation time reduction, and 0.71 % for minimizing environmental impact, demonstrating its efficiency and reliability. Sensitivity analyses further highlight the significant impact of risks including environmental, policy, and operational on transportation and financial outcomes, showing up to a 12 % decrease in profits due to environmental risks alone. These findings underscore the robustness of the model and its applicability in complex, real-world industrial scenarios, making valuable contributions to the literature on sustainable supply chain management, risk mitigation, and multimodal transportation optimization.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100897"},"PeriodicalIF":10.4,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144503974","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":"Image anomaly detection with a unified transformer model guided by dual-feature","authors":"Yuanbo Wang,&nbsp;Junfeng Jing,&nbsp;Xin Zhang","doi":"10.1016/j.jii.2025.100892","DOIUrl":"10.1016/j.jii.2025.100892","url":null,"abstract":"<div><div>Unsupervised industrial image anomaly detection has been extensively investigated by reconstruction-based frameworks. While demonstrating promising performance benchmarks, existing reconstruction networks frequently degenerate into identity mapping behavior: they are prone to directly copying inputs as outputs, which fails to truly learn the deep structural information and statistical distribution of the data, and often require training distinct models for individual object categories (one-class-one-model). In this paper, we propose a powerful multi-class unified model based on the Dual-Feature Guided Reconstruction Network (DFGR) for multi-class anomaly detection. One of DFGR strengthens the low-level feature to realize the guidance function of the model to reconstruct important normal features and significantly reducing the reliance on prior knowledge, and the other multi-layer fusion feature provides rich semantic features of the image. We utilize the main structural features to guide the reconstruction, realizes the interaction between the global information of the main structural features and the local information of the reconstructed feature map. Our method better balances the contribution of the low-level spatial structure information to the overall reconstruction process, and also effectively reduces the sharp response of the reconstructed network to small background noises. Experimental results on the MVTec dataset demonstrate an image-level area under the receiver operating characteristic (AUROC) of 98.0% and a pixel-level AUROC of 97.1%, and further validations on the DAGM2007, Rollei, and red and blue datasets confirm the feasibility of the dual-feature structure. The code: <span><span>https://github.com/wyanb/DFGR</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100892"},"PeriodicalIF":10.4,"publicationDate":"2025-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144504029","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":"A Bi-objective robust optimization and heuristic framework for designing resilient and responsive global supply chains with multimodal transportation","authors":"Bahman Manafi ,&nbsp;Hakan Sayan","doi":"10.1016/j.jii.2025.100895","DOIUrl":"10.1016/j.jii.2025.100895","url":null,"abstract":"<div><div>The design of resilient and responsive supply chain networks has become increasingly critical amid demand uncertainty and operational disruptions, especially in complex sectors such as home appliance manufacturing. This study presents a bi-objective mixed-integer linear programming (BOMILP) model that integrates multimodal transportation planning, strategic facility location, inventory control, and capacity management. The model aims to simultaneously (1) minimize total operational costs, including transportation, inventory, and shortage costs, and (2) maximize customer responsiveness, evaluated through service levels and fulfillment lead times. To address uncertainty, a hybrid approach is developed by combining machine learning-based forecasting and scenario-based robust optimization. Long Short-Term Memory (LSTM) networks forecast demand fluctuations using historical and external data, while the robust model ensures effective resource allocation across multiple demand scenarios. To solve the complex BOMILP model efficiently, a hybrid solution methodology is proposed, integrating the Hybrid Augmented ε-Constraint Method (HA-ε) with a Greedy Randomized Adaptive Search Procedure and Adaptive Variable Neighborhood Search (GRASP-AVNS) heuristic. Resilience is embedded through strategies such as redundant capacities, logistics flexibility, and robust optimization. Resilience performance is assessed using indicators like cost stability, service reliability, and responsiveness under uncertainty. A real-world case study involving a multinational home appliance manufacturer in Turkey demonstrates the model’s effectiveness. Results indicate that the proposed framework reduces total costs by up to 15% and enhances responsiveness by over 20% under disruption scenarios. The hybrid GRASP-AVNS heuristic combined with the augmented ε-constraint method demonstrates strong performance in solving large-scale BOMILP problems under uncertainty. This research provides a scalable, data-driven decision-support tool for manufacturers aiming to balance cost-efficiency, responsiveness, and resilience in uncertain global supply chain environments.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100895"},"PeriodicalIF":10.4,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502506","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":"Graph neural networks to model and optimize the operation of Water Distribution Networks: A review","authors":"Giacomo Vittori ,&nbsp;Yelizaveta Falkouskaya ,&nbsp;Daniel M. Jimenez-Gutierrez ,&nbsp;Tiziana Cattai ,&nbsp;Ioannis Chatzigiannakis","doi":"10.1016/j.jii.2025.100880","DOIUrl":"10.1016/j.jii.2025.100880","url":null,"abstract":"<div><div>Water Distribution Networks (WDNs) have become increasingly complex and interconnected, and the need for advanced modeling and optimization techniques has become fundamental to ensure an efficient and reliable clean water supply. Representing WDNs as graphs naturally models the underlying interacting physical structure and enables the usage of Graph Neural Networks (GNN) that combine the physical structure with abstract notions to capture local and global relationships. GNNs offer significant advantages in contrast to generic Deep Learning (DL) techniques and stand out as a promising solution to model intricate dependencies and enable the investigation of key challenges such as leak detection, water quality monitoring, and demand forecasting. This review presents the physics and hydraulics involved in WDN and the prevalent graph-based models used in the literature. The theoretical foundations of GNNs are shown, highlighting their capabilities in capturing complex spatial relationships and dependencies inherent in the network topology. The most promising GNN-based solutions that can address some of the most critical challenges of WDNs are discussed in detail. We outline the open challenges and potential directions for future developments in this field. By combining multidisciplinary and real-world aspects, this critical review highlights the role of GNNs in modeling and optimizing WDNs.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100880"},"PeriodicalIF":10.4,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144482429","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":"A smart industrial information system using a business process model, discrete events simulation, optimization, and machine learning algorithms","authors":"Seyed Alireza Mansoori Al-yasin ,&nbsp;Mohammad Gheibi ,&nbsp;Hassan Montazeri ,&nbsp;Reza Yeganeh Khaksar ,&nbsp;Mehran Akrami ,&nbsp;Amir M. Fathollahi-Fard ,&nbsp;Kuan Yew Wong","doi":"10.1016/j.jii.2025.100896","DOIUrl":"10.1016/j.jii.2025.100896","url":null,"abstract":"<div><div>In industrial systems, managers face the critical challenge of efficiently managing resources to reduce production costs and time while maximizing profits. To address these challenges, production managers require advanced industrial information systems that optimize production time, costs, and profits. This paper presents a smart industrial information system that integrates Business Process Model and Notation (BPMN), AnyLogic simulation software for Discrete Event (DE) modeling, Response Surface Methodology (RSM), and Machine Learning (ML) algorithms. The system’s effectiveness is demonstrated through its application in an industrial steel skeleton production facility in Iran. To enhance revenue, we optimize key factors of the production process through simulation. Various ML algorithms, including Random Forest (RF), Random Tree (RT), and Bagging, were employed to improve system performance, with the Bagging model yielding the best results. The findings indicate that small hardener chamfer and welder for spare parts, with P-values of 0.0002 and &gt;0.0001 respectively, are the most significant parameters impacting total costs and profits. Ultimately, the proposed industrial information system provides a cost-effective simulation approach that improves process-driven business operations, aligning with BPMN standards and economic criteria.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100896"},"PeriodicalIF":10.4,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144335614","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 inversion-based group decision-making method for evaluating industrial information platforms","authors":"Chuan Yue","doi":"10.1016/j.jii.2025.100881","DOIUrl":"10.1016/j.jii.2025.100881","url":null,"abstract":"<div><div>Quality evaluation of industrial information platforms represents a typical multi-dimensional decision-making problem that requires comprehensive integration of multi-stakeholder perspectives. This paper proposes a novel group decision-making evaluation framework with two key innovations: (1) The introduction of the inversion number concept from linear algebra to quantify evaluators’ data quality, combined with median statistics to establish a dynamic weight allocation mechanism for decision-makers; (2) Building upon the traditional VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methods’ group utility measure, this work innovatively incorporates group regret and group satisfaction matrices, constructing a tripartite “utility-regret-satisfaction” evaluation system through a normalized projection technology, thereby forming an extended VIKOR decision architecture. The proposed method’s feasibility and practicality are validated through a case study on industrial information platform assessment. Experiments demonstrate that: (i) Different data centers can lead to distinct decision outcomes; (ii) Different measures can lead to different decision outcomes; (iii) The inversion-based data quality metric outperforms entropy-based alternatives (with 10% accuracy improvement); (iv) Alternative rankings maintain 70%–100% stability intervals. This research provides a quantifiable, highly robust theoretical tool for multi-attributes decision-making in complex industrial systems.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100881"},"PeriodicalIF":10.4,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307047","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}