Journal of Manufacturing Systems最新文献

{"title":"Mixed-model product disassembly sequence optimization based on cognitive digital twin","authors":"Lei Qi ,&nbsp;Wenjun Xu ,&nbsp;Kaipu Wang ,&nbsp;Jiayi Liu ,&nbsp;Xun Ye ,&nbsp;Hang Yang ,&nbsp;Yi Zhong","doi":"10.1016/j.jmsy.2025.07.004","DOIUrl":"10.1016/j.jmsy.2025.07.004","url":null,"abstract":"<div><div>As a core step in remanufacturing, the disassembly process for multiple product structures in mixed-model products can improve disassembly efficiency and reduce costs. There are structure uncertainties in the mixed-model product disassembly process, which must be considered and used to optimize the disassembly strategy and improve the disassembly efficiency. This paper proposes a framework of a cognitive digital twin for mixed-model product disassembly sequence optimization. The cognitive model can reason, predict, and complete missing disassembly information due to uncertainty in the mixed-model product structure. Its cognitive capability is achieved by a knowledge graph and a TransD-based method. To provide a basis for semantic inference and relate different knowledge types, an ontology is designed based on the digital twin, and a knowledge graph is developed. Finally, a cognitive digital twin model is built. Upon that, the Soft Actor-Critic algorithm is utilized to optimize the mixed-model product sequence. The proposed model and algorithm are applied to transmissions disassembly case. The results show that the proposed method is effective in optimizing the disassembly sequences of three different products that make up the mixed-model products. It not only realizes the disassembly sequence optimization under the uncertain product structures, but also reduces the disassembly time of individual products and all products.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 497-508"},"PeriodicalIF":12.2,"publicationDate":"2025-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144614232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quick dimensional inspection for continuous welding and assembly using machine learning-powered smart jig","authors":"Seobin Park ,&nbsp;Taekyeong Kim ,&nbsp;Kyeong Min Kim ,&nbsp;Junyoung Seo ,&nbsp;Jongwon Chung ,&nbsp;Jeong Ho Choi ,&nbsp;Wooseok Ji ,&nbsp;Im Doo Jung","doi":"10.1016/j.jmsy.2025.07.001","DOIUrl":"10.1016/j.jmsy.2025.07.001","url":null,"abstract":"<div><div>In the mass production of metal-based products such as automobiles, continuous welding and assembly processes are essential. The final product is created through multiple stages of welding, and the cumulative misalignment at each stage can lead to excessive residual stresses or dimensional defects in the product. To compensate for these issues, design modifications or significant post-processing costs have been required. Traditional dimensional inspection methods, whether manual or automated, are limited in their ability to keep pace with the speed required for mass production, as they focus on point-by-point measurements. While 3D vision-based methods offer a solution, they are often costly and primarily suited for macro-scale inspections. Here, we propose a machine learning-powered smart jig that enables precise, micro-level dimensional quality monitoring during production, without interrupting the continuous manufacturing process. This method, designed for direct integration into continuous assembly welding lines, reduces inspection time from 12 min to 2.79 s, enabling the detection of dimensional errors at the 500 μm level. Demonstrations conducted on the production line at a commercial automobile manufacturer confirm the feasibility of this approach for comprehensive subassembly inspections during mass production. This system is expected to be highly adaptable for various manufacturing domains utilizing assembly jigs, offering transformative potential in quality inspection processes.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 478-496"},"PeriodicalIF":12.2,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144587458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive evaluations of condition monitoring-based technologies in industrial maintenance: A systematic review","authors":"Mehdi Dadfarnia ,&nbsp;Michael E. Sharp ,&nbsp;Jeffrey W. Herrmann","doi":"10.1016/j.jmsy.2025.06.015","DOIUrl":"10.1016/j.jmsy.2025.06.015","url":null,"abstract":"<div><div>Condition monitoring involves detecting, diagnosing, or predicting faults or failures in industrial equipment. Given advances in the underlying artificial intelligence solutions and internet of things-based technologies, condition monitoring has the potential to improve industrial maintenance processes rapidly. Adopting condition monitoring-based technologies requires evaluating their engineering and financial benefits to determine whether the investment is justified. An increasing number of studies describe procedures to evaluate condition monitoring-based maintenance, but the literature lacks a review of these evaluation studies to identify research opportunities and best practices. This systematic review aims to report and analyze the evaluation methods for using condition monitoring-based technologies in industrial maintenance. This review identified 465 relevant peer-reviewed studies between 2001 and 2023, from which 42 articles met the eligibility criteria. For each article, this paper analyzed facets of the evaluation process related to the study’s characterizations of the industrial application, condition monitoring, maintenance deployment, evaluation techniques, performance measures, and economic analysis. Collectively, these results yield several insights. Few condition monitoring evaluation studies exist for manufacturing systems, unlike the domains of energy systems and transportation modes. Also, many studies lack details about condition monitoring and maintenance models. Additionally, the evaluation techniques across most studies can improve with combinations of analytical frameworks, simulation, and expanded sensitivity analysis. Lastly, the reviewed studies are difficult to directly compare due to heterogeneity in economic analysis, performance measures, and uncertainty analysis — indicating an opportunity for future research to structure comprehensive reporting items to enhance the comparability of domain-specific condition monitoring-based maintenance evaluations. Based on the literature review and analyses, this review suggests specific recommendations for future condition monitoring evaluation and opportunities for further research.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 449-477"},"PeriodicalIF":12.2,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144579429","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Industrial Foundation Models (IFMs) for intelligent manufacturing: A systematic review","authors":"Shuxuan Zhao ,&nbsp;Sichao Liu ,&nbsp;Yishuo Jiang ,&nbsp;Bo Zhao ,&nbsp;Youlong Lv ,&nbsp;Jie Zhang ,&nbsp;Lihui Wang ,&nbsp;Ray Y. Zhong","doi":"10.1016/j.jmsy.2025.06.011","DOIUrl":"10.1016/j.jmsy.2025.06.011","url":null,"abstract":"<div><div>The remarkable success of Large Foundation Models (LFMs) has demonstrated their tremendous potential for manufacturing and sparked significant interest in the exploration of Industrial Foundation Models (IFMs). This study provides a comprehensive review of the current state of IFMs and their applications in intelligent manufacturing. It conducts an in-depth analysis from three perspectives, including data level, model level, and application level. The definition and framework of IFMs are discussed with a comparison to LFMs across these three perspectives. In addition, this paper provides a brief overview of the advancements in IFMs development across different countries, institutions, and regions. It explores the current application of IFMs, including Industrial Domain Models and Industrial Task Models, which are specifically designed for various industrial domains and tasks. Furthermore, key technologies critical to the training of IFMs are explored, such as data pre-processing, model fine-tuning, prompt engineering, and retrieval-augmented generation. This paper also highlights the essential capabilities of IFMs and their typical applications throughout the manufacturing lifecycle. Finally, it discusses the current challenges and outlines potential future research directions. This study aims to inspire new ideas for advancing IFMs and accelerating the evolution of intelligent manufacturing.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 420-448"},"PeriodicalIF":12.2,"publicationDate":"2025-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144579428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Flexible Manufacturing Systems intralogistics: Dynamic optimization of AGVs and tool sharing using Colored-Timed Petri Nets and actor–critic RL with actions masking","authors":"Sofiene Lassoued ,&nbsp;Laxmikant Shrikant Baheti ,&nbsp;Nathalie Weiß-Borkowski ,&nbsp;Stefan Lier ,&nbsp;Andreas Schwung","doi":"10.1016/j.jmsy.2025.06.017","DOIUrl":"10.1016/j.jmsy.2025.06.017","url":null,"abstract":"<div><div>Flexible Manufacturing Systems (FMS) are pivotal in optimizing production processes in today’s rapidly evolving manufacturing landscape. This paper advances the traditional job shop scheduling problem by incorporating additional complexities through the simultaneous integration of automated guided vehicles (AGVs) and tool-sharing systems. We propose a novel approach that combines Colored-Timed Petri Nets (CTPNs) with actor–critic model-based reinforcement learning (MBRL), effectively addressing the multifaceted challenges associated with FMS. CTPNs provide a formal modeling structure and dynamic action masking, significantly reducing the action search space, while MBRL ensures adaptability to changing environments through the learned policy. Leveraging the advantages of MBRL, we incorporate a lookahead strategy for optimal positioning of AGVs, improving operational efficiency. Our approach was evaluated on small-sized public benchmarks and a newly developed large-scale benchmark inspired by the Taillard benchmark. The results show that our approach matches traditional methods on smaller instances and outperforms them on larger ones in terms of makespan while achieving a tenfold reduction in computation time. To ensure reproducibility, we propose a gym-compatible environment and an instance generator. Additionally, an ablation study evaluates the contribution of each framework component to its overall performance.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 405-419"},"PeriodicalIF":12.2,"publicationDate":"2025-07-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144570334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An order postponement strategy for multi-stage production-logistics synchronization towards zero-warehousing smart manufacturing","authors":"Mingxing Li ,&nbsp;Xiaoyu Qian ,&nbsp;Ming Li ,&nbsp;Ting Qu ,&nbsp;Zhen He","doi":"10.1016/j.jmsy.2025.06.024","DOIUrl":"10.1016/j.jmsy.2025.06.024","url":null,"abstract":"<div><div>Nowadays, driven by market dynamics and evolving consumer demands, customized production has emerged as a prevalent trend. This paradigm shift from mass production to customization introduces challenges in production-logistics management, compelling manufacturers to pursue efficient strategies. Zero-warehousing smart manufacturing (ZWSM), an advanced form of Lean manufacturing (LM) and Just-In-Time production (JIT), presents a potential solution to these challenges. ZWSM leverages Industry 4.0 technologies to facilitate seamless production-logistics for eliminating warehouses and minimizing inventory in workshop. Despite the encouraging visions, field study reveals that ZWSM requires highly coordinated material supply, production, and delivery operations, misalignment among these stages frequently results in operational inefficiencies and resource waste, especially when confronted with volatile and diversified customer demand. It is defined as Multi-Stage Production-Logistics Synchronization (MS-PLSync) problem. This study proposes a novel order postponement strategy for MS-PLSync towards ZWSM, a generalizable MS-PLSync model under assemble-to-order (ATO) is formulated using mixed-integer linear programming for production-logistics operations under intricate spatiotemporal constraints. Considering dynamic order arrivals, a postponement strategy is designed and integrated into MS-PLSync model to enhance overall operational efficiency through postponed real-time decision-making, achieving a balance between rapid response to demand fluctuations and economy of scale in customized production-logistics. Numerical analysis validates the effectiveness of the proposed postponement strategy in addressing MS-PLSync problem. Notably, a little postponement can yield substantial operational benefits, while excessive postponement only generates minimal marginal benefits. Furthermore, sensitivity analysis reveals that the proposed postponement strategy performs particularly well in mass customization scenarios characterized by large-scale orders, diverse product portfolios, and extensive distribution networks.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 389-404"},"PeriodicalIF":12.2,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144518900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Skeleton-based action recognition for manufacturing assembly task through graph convolution network","authors":"Maryam Soleymani ,&nbsp;Mahdi Bonyani ,&nbsp;Chao Wang","doi":"10.1016/j.jmsy.2025.06.019","DOIUrl":"10.1016/j.jmsy.2025.06.019","url":null,"abstract":"<div><div>In modern manufacturing, human participation in assembly processes is essential, despite advancements in automation. However, accurately recognizing human actions in these environments presents challenges due to complex spatial–temporal dependencies and dynamic joint relationships. Graph Convolution Networks (GCNs) are utilized widely for action recognition, but they have poor accuracy for modeling long-range node correlations. Also, current GCNs have limitations in extracting various features due to utilizing the same pattern extraction for all frames. To overcome these issues, this study presents a novel approach to skeleton-based action recognition for manufacturing tasks using a Dual-Attention Graph Convolution Network (DAGCN). The proposed model integrates a Parallel Attention-Graph Mixer (PAGM) and Temporal–Spatial Attention Integrator (TSAI), enhancing the capture of both global and local joint relations and addressing the dynamic nature of skeletal joint relationships. Extensive evaluations on benchmark datasets, including HA4M that specifically designed for assembly tasks, NTU RGB+D, Northwestern-UCLA, and NTU RGB+D120, reveal the superior performance of DAGCN over state-of-the-art methods in terms of accuracy and computational efficiency. Experimental results demonstrate that DAGCN outperforms state-of-the-art methods, achieving a Top-1 accuracy of 89.0% on the HA4M dataset. The results validate DAGCN’s effectiveness in recognizing fine-grained human actions in industrial settings, contributing to improved efficiency and safety in human–robot collaboration. The proposed model offers a scalable and computationally efficient solution for intelligent assembly monitoring and automation in smart manufacturing systems.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 362-375"},"PeriodicalIF":12.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144513978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing sustainability of human-robot collaboration in industry 5.0: Context- and interaction-aware human motion prediction for proactive robot control","authors":"Guoyi Xia ,&nbsp;Zied Ghrairi ,&nbsp;Aaron Heuermann ,&nbsp;Klaus-Dieter Thoben","doi":"10.1016/j.jmsy.2025.06.022","DOIUrl":"10.1016/j.jmsy.2025.06.022","url":null,"abstract":"<div><div>Industry 5.0 (I5.0) marks a shift towards human-centric, sustainable, and resilient production systems, with Human-Robot Collaboration (HRC) contributing to these goals. Achieving sustainability of HRC, encompassing economic, environmental, and social dimensions, remains challenging to ensure safety, efficiency, and adaptability. Human Motion Prediction (HMP) can address these challenges by enabling robots to anticipate human actions and respond proactively. However, existing HMP studies often neglect to incorporate contextual and interaction-based information. The practical applicability of HMP in industrial settings requires further demonstration. Therefore, this study aims to apply context- and interaction-aware HMP to enhance sustainability of HRC in I5.0. A motion capture system collects human motion data, while a camera tracks object position as contextual information. Human-Object Interaction (HOI) is identified for HMP. A transformer model is applied for HMP based on integrated context and interaction data. Additionally, the applicability of HMP in industrial settings is demonstrated by a power transformer assembly. Two additional cases are applied for validation. Results show that object recognition achieved 98 % accuracy. The identified interaction periods are effective in enhancing HMP performance. HMP with context and interaction data achieves an Average Displacement Error (ADE) of 0.07 m and a Final Displacement Error (FDE) of 0.10 m. The demonstration results suggest that the HMP enabled proactive robot control, contributing to safer, more efficient, and adaptive production. The findings of this research contribute to enhancing the sustainability of HRC in I5.0, with potential benefits for environmental efficiency, worker safety, and productivity in industrial settings.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 376-388"},"PeriodicalIF":12.2,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144513979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-robot collaborative manufacturing driven by digital twins: Advancements, challenges, and future directions","authors":"Gang Wang ,&nbsp;Cheng Zhang ,&nbsp;Sichao Liu ,&nbsp;Yongxuan Zhao ,&nbsp;Yingfeng Zhang ,&nbsp;Lihui Wang","doi":"10.1016/j.jmsy.2025.06.014","DOIUrl":"10.1016/j.jmsy.2025.06.014","url":null,"abstract":"<div><div>Multi-robot systems envisioned for future factories will promote advancements and capabilities of handling complex tasks and realising optimal robotic operations. However, existing multi-robot systems face challenges such as integration complexity, difficult coordination and control, low scalability, and flexibility, and thus are far from realising adaptive and efficient multi-robot collaborative manufacturing (MRCM). Digital twin technology improves visualisation, consistency, and spatial–temporal collaboration in MRCM through real-time interaction and iterative optimisation in physical and virtual spaces. Despite these improvements, barriers such as undeveloped modelling capabilities, indeterminate collaborative strategies, and limited applicability impede widespread integration of MRCM. In response to these needs, this study provides a comprehensive review of the foundational concepts, systematic architecture, and enabling technologies of digital twin-driven MRCM, serving as a prospective vision for future work in collaborative intelligent manufacturing. With the development of sensors and computational capabilities, robot intelligence is evolving towards multi-robot collaboration, including perceptual, cognitive, and behavioural collaboration. Digital twins play a critical supporting role in multi-robot collaboration, and the architecture, methodologies, and applications are elaborated across diverse stages of MRCM processes. This paper also identifies current challenges and future research directions. It encourages academic and industrial stakeholders to integrate state-of-the-art AI technologies more thoroughly into multi-robot digital twin systems for enhanced efficiency and reliability in production.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 333-361"},"PeriodicalIF":12.2,"publicationDate":"2025-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144481048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From cost to capability: Technology Multiplier in EV manufacturing strategy","authors":"Roman Girke ,&nbsp;Louis Schäfer ,&nbsp;Tanja Maier ,&nbsp;Florian Stamer ,&nbsp;Shun Yang ,&nbsp;Jung-Hoon Chun ,&nbsp;Gisela Lanza","doi":"10.1016/j.jmsy.2025.06.008","DOIUrl":"10.1016/j.jmsy.2025.06.008","url":null,"abstract":"<div><div>In an increasingly dynamic and uncertain global environment, companies face significant challenges in aligning production strategies with long-term competitiveness and innovation. Many firms focus on short-term cost savings through outsourcing and relocation, which can negatively impact internal knowledge, capabilities, and long-term growth. This study explores the strategic decision-making process of production segment allocation, emphasizing the need for a systematic and holistic approach. Using the Analytical Hierarchy Process (AHP), key decision factors are introduced, including cost, innovation capability, and social impact. The qualitative study is applied to the case of electric vehicle (EV) battery production, where in-house production, outsourcing, and joint ventures are evaluated. The results reveal key drivers such as design capabilities and customer centricity, highlighting the strategic importance of maintaining core technological capabilities in-house. The study provides a foundation for the concept of a ”Technology Multiplier” to quantify the long-term value of in-house production in terms of innovation, adaptability, and competitive advantage, offering a new framework for strategic manufacturing decisions.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"82 ","pages":"Pages 319-332"},"PeriodicalIF":12.2,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144471582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}