Pub Date : 2026-02-01Epub Date: 2025-12-12DOI: 10.1016/j.jmsy.2025.11.022
Yuqi Cheng , Yunkang Cao , Haiming Yao , Wei Luo , Cheng Jiang , Hui Zhang , Weiming Shen
Industrial surface defect detection is vital for upholding product quality across contemporary manufacturing systems. As the expectations for precision, automation, and scalability intensify, conventional inspection approaches are increasingly found wanting in addressing real-world demands. Notable progress in computer vision and deep learning has substantially bolstered defect detection capabilities across both 2D and 3D modalities. A significant development has been the pivot from closed-set to open-set defect detection frameworks, which diminishes the necessity for extensive defect annotations and facilitates the recognition of novel anomalies. Despite such strides, a cohesive and contemporary understanding of industrial defect detection remains elusive. Consequently, this survey delivers an in-depth analysis of both closed-set and open-set defect detection strategies within 2D and 3D modalities, charting their evolution in recent years and underscoring the rising prominence of open-set techniques. We distill critical challenges inherent in practical detection environments and illuminate emerging trends, thereby providing a current and comprehensive vista of this swiftly progressing field.
{"title":"A comprehensive survey for real-world industrial surface defect detection: Challenges, approaches, and prospects","authors":"Yuqi Cheng , Yunkang Cao , Haiming Yao , Wei Luo , Cheng Jiang , Hui Zhang , Weiming Shen","doi":"10.1016/j.jmsy.2025.11.022","DOIUrl":"10.1016/j.jmsy.2025.11.022","url":null,"abstract":"<div><div>Industrial surface defect detection is vital for upholding product quality across contemporary manufacturing systems. As the expectations for precision, automation, and scalability intensify, conventional inspection approaches are increasingly found wanting in addressing real-world demands. Notable progress in computer vision and deep learning has substantially bolstered defect detection capabilities across both 2D and 3D modalities. A significant development has been the pivot from closed-set to open-set defect detection frameworks, which diminishes the necessity for extensive defect annotations and facilitates the recognition of novel anomalies. Despite such strides, a cohesive and contemporary understanding of industrial defect detection remains elusive. Consequently, this survey delivers an in-depth analysis of both closed-set and open-set defect detection strategies within 2D and 3D modalities, charting their evolution in recent years and underscoring the rising prominence of open-set techniques. We distill critical challenges inherent in practical detection environments and illuminate emerging trends, thereby providing a current and comprehensive vista of this swiftly progressing field.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 152-172"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786611","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}
Pub Date : 2026-02-01Epub Date: 2025-12-15DOI: 10.1016/j.jmsy.2025.12.008
Xinyu Liu , Yunlei Zan , Honghui Wang , Xu Han , Guijie Liu
Against the background of low-carbon and efficient sustainable manufacturing, production scheduling must not only focus on efficiency and delivery times but also balance energy consumption and carbon emission constraints. However, modern manufacturing workshops commonly adopt a model where batch production and trial production coexist. Production orders and equipment load evolve over time, with frequent occurrences of order insertion, maintenance. At the same time, workshop data exhibits multisource, heterogeneous, and time-varying characteristics. It leads to underutilized production information and delayed scheduling decisions, resulting in efficiency losses and redundant energy consumption. To address the above issues, a deep reinforcement learning approach driven by temporal knowledge graph is proposed in this paper. Firstly, a knowledge graph-based workshop scheduling framework is established to perform unified semantic modeling of production factors and the scheduling relationships. It constructs a multidimensional information matrix with temporal rule constraints. Then, a temporal knowledge-driven LSTM-TD3 algorithm is proposed by replacing the original policy network with an LSTM and employing dual Q-networks with policy delay updates to improve training convergence in continuous action spaces. On this basis, an adaptive weighting mechanism for reward functions targeting different production events is defined to achieve a reasonable balance and dynamic decision-making among multiple objectives. Finally, a case study is conducted with a boiler screen tube production line, and a prototype system is built. The results indicate that compared to historical production data, the average energy consumption of production line equipment decreased by 4.10 %, and the overall efficiency of the workshop increased by 14.29 %, validating the effectiveness of this approach.
{"title":"A deep reinforcement learning approach driven by temporal knowledge graph for dynamic scheduling in discrete manufacturing workshops","authors":"Xinyu Liu , Yunlei Zan , Honghui Wang , Xu Han , Guijie Liu","doi":"10.1016/j.jmsy.2025.12.008","DOIUrl":"10.1016/j.jmsy.2025.12.008","url":null,"abstract":"<div><div>Against the background of low-carbon and efficient sustainable manufacturing, production scheduling must not only focus on efficiency and delivery times but also balance energy consumption and carbon emission constraints. However, modern manufacturing workshops commonly adopt a model where batch production and trial production coexist. Production orders and equipment load evolve over time, with frequent occurrences of order insertion, maintenance. At the same time, workshop data exhibits multisource, heterogeneous, and time-varying characteristics. It leads to underutilized production information and delayed scheduling decisions, resulting in efficiency losses and redundant energy consumption. To address the above issues, a deep reinforcement learning approach driven by temporal knowledge graph is proposed in this paper. Firstly, a knowledge graph-based workshop scheduling framework is established to perform unified semantic modeling of production factors and the scheduling relationships. It constructs a multidimensional information matrix with temporal rule constraints. Then, a temporal knowledge-driven LSTM-TD3 algorithm is proposed by replacing the original policy network with an LSTM and employing dual Q-networks with policy delay updates to improve training convergence in continuous action spaces. On this basis, an adaptive weighting mechanism for reward functions targeting different production events is defined to achieve a reasonable balance and dynamic decision-making among multiple objectives. Finally, a case study is conducted with a boiler screen tube production line, and a prototype system is built. The results indicate that compared to historical production data, the average energy consumption of production line equipment decreased by 4.10 %, and the overall efficiency of the workshop increased by 14.29 %, validating the effectiveness of this approach.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 242-258"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786666","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}
Pub Date : 2026-02-01Epub Date: 2025-12-26DOI: 10.1016/j.jmsy.2025.12.019
Jintao Xue, Xiao Li, Nianmin Zhang
Human–robot collaborative manufacturing, a core aspect of Industry 5.0, emphasizes ergonomics to enhance worker well-being. This paper addresses the dynamic human–robot task planning and allocation (HRTPA) problem, which involves determining when to perform tasks and who should execute them to maximize efficiency while ensuring workers’ physical fatigue remains within safe limits. The inclusion of fatigue constraints, combined with production dynamics, significantly increases the complexity of the HRTPA problem. Traditional fatigue–recovery models in HRTPA often rely on static, predefined hyperparameters. However, in practice, human fatigue sensitivity varies daily due to factors such as changed work conditions and insufficient sleep. To better capture this uncertainty, we treat fatigue-related parameters as inaccurate and estimate them online based on observed fatigue progression during production. To address these challenges, we propose PF-CD3Q, a safe reinforcement learning (safe RL) approach that integrates the particle filter with constrained dueling double deep Q-learning for real-time fatigue-predictive HRTPA. Specifically, we first develop PF-based estimators to track human fatigue and update fatigue model parameters in real-time. These estimators are then integrated into CD3Q by making task-level fatigue predictions during decision-making and excluding tasks that exceed fatigue limits, thereby constraining the action space and formulating the problem as a constrained Markov decision process (CMDP). Experimental results demonstrate that our PF-based estimators achieve high prediction accuracy and strong noise robustness, and that PF-CD3Q outperforms other algorithms across multiple performance metrics, significantly reducing the occurrence of overwork and adapting to unseen fatigue constraints after training. These findings validate the effectiveness of our approach under complex and dynamic production conditions, supporting both human well-being and the development of a more sustainable and efficient manufacturing paradigm.
{"title":"Safe reinforcement learning with online filtering for fatigue-predictive human–robot task planning and allocation in production","authors":"Jintao Xue, Xiao Li, Nianmin Zhang","doi":"10.1016/j.jmsy.2025.12.019","DOIUrl":"10.1016/j.jmsy.2025.12.019","url":null,"abstract":"<div><div>Human–robot collaborative manufacturing, a core aspect of Industry 5.0, emphasizes ergonomics to enhance worker well-being. This paper addresses the dynamic human–robot task planning and allocation (HRTPA) problem, which involves determining when to perform tasks and who should execute them to maximize efficiency while ensuring workers’ physical fatigue remains within safe limits. The inclusion of fatigue constraints, combined with production dynamics, significantly increases the complexity of the HRTPA problem. Traditional fatigue–recovery models in HRTPA often rely on static, predefined hyperparameters. However, in practice, human fatigue sensitivity varies daily due to factors such as changed work conditions and insufficient sleep. To better capture this uncertainty, we treat fatigue-related parameters as inaccurate and estimate them online based on observed fatigue progression during production. To address these challenges, we propose PF-CD3Q, a safe reinforcement learning (safe RL) approach that integrates the <strong>p</strong>article <strong>f</strong>ilter with <strong>c</strong>onstrained <strong>d</strong>ueling <strong>d</strong>ouble <strong>d</strong>eep <strong>Q</strong>-learning for real-time fatigue-predictive HRTPA. Specifically, we first develop PF-based estimators to track human fatigue and update fatigue model parameters in real-time. These estimators are then integrated into CD3Q by making task-level fatigue predictions during decision-making and excluding tasks that exceed fatigue limits, thereby constraining the action space and formulating the problem as a constrained Markov decision process (CMDP). Experimental results demonstrate that our PF-based estimators achieve high prediction accuracy and strong noise robustness, and that PF-CD3Q outperforms other algorithms across multiple performance metrics, significantly reducing the occurrence of overwork and adapting to unseen fatigue constraints after training. These findings validate the effectiveness of our approach under complex and dynamic production conditions, supporting both human well-being and the development of a more sustainable and efficient manufacturing paradigm.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 561-583"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145836451","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}
Pub Date : 2026-02-01Epub Date: 2025-12-18DOI: 10.1016/j.jmsy.2025.12.010
Yan Xu , Li Li , Guanghui Lang , Yang Luo , Junhua Zhao , Congbo Li
Tool wear is a critical factor influencing the cost of machining products, especially in the milling of complex Ti-6Al-4V components, where tool wear is significantly accelerated. Therefore, it is essential to develop accurate methods for diagnosing tool wear. This paper proposes an ensemble learning model that integrates domain knowledge and physical information through regularized fusion for the diagnosis of tool wear conditions. Initially, a physics-based model was developed to relate milling forces to the width of tool flank wear (VB) by analyzing the characteristics of milling forces under conditions of progressive tool wear. Following this, a stacking ensemble framework was utilized to combine the selected base models. Domain-specific knowledge, encompassing different stages of tool wear, as well as physical information, including the force-VB relationship, were integrated through the formulation of the loss function. Moreover, a two-stage feature reduction methodology integrating Spearman's rank correlation analysis (Spearman's ρ) with Principal Component Analysis (PCA) was introduced to improve the relevance and compactness of the features. Subsequently, milling experiments were performed on a machining center to assess the effectiveness and practical applicability of the proposed approach.
{"title":"Tool wear condition diagnosis using ensemble learning with regularized fusion of domain knowledge and physical information for Ti-6Al-4V milling","authors":"Yan Xu , Li Li , Guanghui Lang , Yang Luo , Junhua Zhao , Congbo Li","doi":"10.1016/j.jmsy.2025.12.010","DOIUrl":"10.1016/j.jmsy.2025.12.010","url":null,"abstract":"<div><div>Tool wear is a critical factor influencing the cost of machining products, especially in the milling of complex Ti-6Al-4V components, where tool wear is significantly accelerated. Therefore, it is essential to develop accurate methods for diagnosing tool wear. This paper proposes an ensemble learning model that integrates domain knowledge and physical information through regularized fusion for the diagnosis of tool wear conditions. Initially, a physics-based model was developed to relate milling forces to the width of tool flank wear (<em>VB</em>) by analyzing the characteristics of milling forces under conditions of progressive tool wear. Following this, a stacking ensemble framework was utilized to combine the selected base models. Domain-specific knowledge, encompassing different stages of tool wear, as well as physical information, including the force-<em>VB</em> relationship, were integrated through the formulation of the loss function. Moreover, a two-stage feature reduction methodology integrating Spearman's rank correlation analysis (Spearman's ρ) with Principal Component Analysis (PCA) was introduced to improve the relevance and compactness of the features. Subsequently, milling experiments were performed on a machining center to assess the effectiveness and practical applicability of the proposed approach.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 383-402"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786609","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}
Pub Date : 2026-02-01Epub Date: 2025-12-15DOI: 10.1016/j.jmsy.2025.11.021
I. Rodriguez , P.J. Arrazola , M. Mori , G. Ortiz-de-Zarate , A. Madariaga , M. Cuesta , F. Pušavec
Machining operations influence sustainability not only through immediate energy and resource use but also through their effects on tool longevity and part durability. Conventional life cycle assessments (LCAs) of machining typically adopt gate-to-gate boundaries, overlooking how machining parameters affect surface integrity, fatigue life, and the frequency of cutting-tool and machined component replacement. This study develops a novel cradle-to-cradle LCA framework that incorporates machining-induced tool wear and part durability into system-level environmental impacts. The approach is demonstrated through a real case study of drilling the CFRP/Ti6Al4V stacks used for the Boeing 787 fuselage, comparing dry drilling with liquid carbon dioxide combined with minimum quantity lubrication (LCO₂+MQL). Experimental tests quantified energy demand, resource consumption, tool life, and hole quality, while part durability was evaluated via fatigue testing and finite element simulations. Results show that the cutting tool production was the dominant contributor to the overall environmental impact. Despite requiring additional auxiliary inputs, LCO₂+MQL drilling reduced overall environmental impacts by 60–70 % relative to dry drilling, due to extended tool life and improved component service life. These findings demonstrate that coolant-assisted machining, when durability and tool-life effects are considered, yields a net environmental benefit. The proposed framework provides a transferable method to expand LCAs beyond gate-to-gate boundaries by integrating the influence of machined part quality on durability and in-service repairs, enabling cradle-to-cradle assessments that better capture the system-level implications of machining decisions.
{"title":"A cradle-to-cradle life cycle assessment framework linking machining parameters, tool life and part durability","authors":"I. Rodriguez , P.J. Arrazola , M. Mori , G. Ortiz-de-Zarate , A. Madariaga , M. Cuesta , F. Pušavec","doi":"10.1016/j.jmsy.2025.11.021","DOIUrl":"10.1016/j.jmsy.2025.11.021","url":null,"abstract":"<div><div>Machining operations influence sustainability not only through immediate energy and resource use but also through their effects on tool longevity and part durability. Conventional life cycle assessments (LCAs) of machining typically adopt gate-to-gate boundaries, overlooking how machining parameters affect surface integrity, fatigue life, and the frequency of cutting-tool and machined component replacement. This study develops a novel cradle-to-cradle LCA framework that incorporates machining-induced tool wear and part durability into system-level environmental impacts. The approach is demonstrated through a real case study of drilling the CFRP/Ti6Al4V stacks used for the Boeing 787 fuselage, comparing dry drilling with liquid carbon dioxide combined with minimum quantity lubrication (LCO₂+MQL). Experimental tests quantified energy demand, resource consumption, tool life, and hole quality, while part durability was evaluated via fatigue testing and finite element simulations. Results show that the cutting tool production was the dominant contributor to the overall environmental impact. Despite requiring additional auxiliary inputs, LCO₂+MQL drilling reduced overall environmental impacts by 60–70 % relative to dry drilling, due to extended tool life and improved component service life. These findings demonstrate that coolant-assisted machining, when durability and tool-life effects are considered, yields a net environmental benefit. The proposed framework provides a transferable method to expand LCAs beyond gate-to-gate boundaries by integrating the influence of machined part quality on durability and in-service repairs, enabling cradle-to-cradle assessments that better capture the system-level implications of machining decisions.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 207-222"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786668","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}
Pub Date : 2026-02-01Epub Date: 2025-12-09DOI: 10.1016/j.jmsy.2025.11.020
Wang Cong, Wu Tao, Bao Jinsong
Intelligent operation and maintenance of energy equipment represents a critical component in ensuring the stable performance of new-generation power systems. Faced with complex operational conditions and nonlinear fault characteristics, conventional manual maintenance suffers from delayed perception and ambiguous causality. However, while digital twin technology can establish a virtual-real interaction space, its static modeling approach exhibits prediction failure in dynamic scenarios. To address these challenges, this study proposes an intelligent maintenance methodology based on cognitive agents and virtual-real co-evolution: constructing a dynamic environment representation model to achieve spatiotemporal feature correlation of equipment states and operational condition migration; designing a memory-planning-decision architecture to enhance causal reasoning capabilities for equipment faults and integrating with digital twin models for virtual-real interaction. The methodology is validated through an 18-month case study of a gas-steam boiler in a combined heat and power plant, utilizing 5.2 million historical operational records. Experimental results demonstrate that this approach achieves a 97.3 % accuracy rate in diagnosing non-stationary faults for gas-steam boiler equipment, realizes a 20-fold improvement in knowledge update time (from 48 to 2.3 h), and attains significant performance enhancements including 31.2 % cost efficiency improvement, 3-fold early warning lead time extension (from 24 to 72 h), and 16.2 % overall collaborative performance improvement (from 82.2 % to 95.5 %). The research validates the engineering value of dynamic cognitive paradigms in intelligent maintenance of power equipment, providing a feasible solution for autonomous decision-making in high-real-time scenarios.
{"title":"Agentic digital twin-embedded maintenance methodology for energy equipment: A self-evolving operational paradigm","authors":"Wang Cong, Wu Tao, Bao Jinsong","doi":"10.1016/j.jmsy.2025.11.020","DOIUrl":"10.1016/j.jmsy.2025.11.020","url":null,"abstract":"<div><div>Intelligent operation and maintenance of energy equipment represents a critical component in ensuring the stable performance of new-generation power systems. Faced with complex operational conditions and nonlinear fault characteristics, conventional manual maintenance suffers from delayed perception and ambiguous causality. However, while digital twin technology can establish a virtual-real interaction space, its static modeling approach exhibits prediction failure in dynamic scenarios. To address these challenges, this study proposes an intelligent maintenance methodology based on cognitive agents and virtual-real co-evolution: constructing a dynamic environment representation model to achieve spatiotemporal feature correlation of equipment states and operational condition migration; designing a memory-planning-decision architecture to enhance causal reasoning capabilities for equipment faults and integrating with digital twin models for virtual-real interaction. The methodology is validated through an 18-month case study of a gas-steam boiler in a combined heat and power plant, utilizing 5.2 million historical operational records. Experimental results demonstrate that this approach achieves a 97.3 % accuracy rate in diagnosing non-stationary faults for gas-steam boiler equipment, realizes a 20-fold improvement in knowledge update time (from 48 to 2.3 h), and attains significant performance enhancements including 31.2 % cost efficiency improvement, 3-fold early warning lead time extension (from 24 to 72 h), and 16.2 % overall collaborative performance improvement (from 82.2 % to 95.5 %). The research validates the engineering value of dynamic cognitive paradigms in intelligent maintenance of power equipment, providing a feasible solution for autonomous decision-making in high-real-time scenarios.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 100-116"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786761","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}
Pub Date : 2026-02-01Epub Date: 2026-01-02DOI: 10.1016/j.jmsy.2025.12.025
Seohyun Choi , Young Ha Joo , Hoonseok Park , Younkook Kang , Ri Choe , Jae-Yoon Jung
Reliable transfer time prediction is crucial for productivity in automated material handling systems within modern manufacturing environments. However, the complexity and dynamic behavior of manufacturing and logistics systems make accurate transfer time estimation highly challenging. This study proposes a transfer time prediction approach for automated material handling systems operating across production lines. To forecast inter-building and inter-floor transfer times, this study proposes a hybrid method, called time-series residual regression, that integrates linear time-series analysis with nonlinear machine learning. The framework further employs three data pooling strategies to effectively capture device heterogeneity and improve forecasting robustness. The hybrid method was validated using transfer records from inter-building stockers and inter-floor lifters in a Korean semiconductor fab. The experimental results show that the proposed model delivers superior performance, achieving R-squared values of 64.01 % for inter-building transfers and 72.00 % for inter-floor transfers.
{"title":"Predicting transfer times across production lines using data pooling","authors":"Seohyun Choi , Young Ha Joo , Hoonseok Park , Younkook Kang , Ri Choe , Jae-Yoon Jung","doi":"10.1016/j.jmsy.2025.12.025","DOIUrl":"10.1016/j.jmsy.2025.12.025","url":null,"abstract":"<div><div>Reliable transfer time prediction is crucial for productivity in automated material handling systems within modern manufacturing environments. However, the complexity and dynamic behavior of manufacturing and logistics systems make accurate transfer time estimation highly challenging. This study proposes a transfer time prediction approach for automated material handling systems operating across production lines. To forecast inter-building and inter-floor transfer times, this study proposes a hybrid method, called time-series residual regression, that integrates linear time-series analysis with nonlinear machine learning. The framework further employs three data pooling strategies to effectively capture device heterogeneity and improve forecasting robustness. The hybrid method was validated using transfer records from inter-building stockers and inter-floor lifters in a Korean semiconductor fab. The experimental results show that the proposed model delivers superior performance, achieving R-squared values of 64.01 % for inter-building transfers and 72.00 % for inter-floor transfers.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 629-645"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145880098","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}
Pub Date : 2026-02-01Epub Date: 2025-11-26DOI: 10.1016/j.jmsy.2025.11.016
Yunfei Ma , Shuai Zheng , Zheng Yang , Pai Zheng , Jiewu Leng , Jun Hong
In commercial aircraft manufacturing, process planning serves as a crucial bridge between design and production, ensuring the accurate realization of design concepts and significantly improving manufacturing efficiency and product quality. With the development of knowledge graph technologies, significant progress has been made in using historical process documentation for commercial aircraft manufacturing process planning. However, traditional deep learning-based methods for constructing knowledge graph heavily rely on manual object selection and label assignment, making the process highly time-consuming. Additionally, the methods often face challenges in the field of process planning, including low domain-specific terminology recognition rates and incomplete entity extraction. To tackle these challenges, this paper introduces a hybrid approach that integrates large and small language models to construct an aircraft process planning knowledge graph. Initially, clustering-based multi-agent approach is employed to pre-annotate the process planning dataset, with domain experts re-annotate the defect data to create a high-quality process planning dataset. Subsequently, a knowledge extraction framework for aircraft process planning, KE-LSM, was constructed using the small language model trained on this dataset, together with the LLM. Experimental results show that KE-LSM outperforms existing named entity recognition models. Finally, KE-LSM is applied in a commercial aircraft manufacturing company, accompanied by the development of a prototype system designed to facilitate intelligent process planning. It is hoped that the research can provide valuable insights and support for the application of LLM-based solutions in the field of aircraft manufacturing.
{"title":"Aircraft assembly process planning based on knowledge graph constructed by integrating LLMs and SLMs","authors":"Yunfei Ma , Shuai Zheng , Zheng Yang , Pai Zheng , Jiewu Leng , Jun Hong","doi":"10.1016/j.jmsy.2025.11.016","DOIUrl":"10.1016/j.jmsy.2025.11.016","url":null,"abstract":"<div><div>In commercial aircraft manufacturing, process planning serves as a crucial bridge between design and production, ensuring the accurate realization of design concepts and significantly improving manufacturing efficiency and product quality. With the development of knowledge graph technologies, significant progress has been made in using historical process documentation for commercial aircraft manufacturing process planning. However, traditional deep learning-based methods for constructing knowledge graph heavily rely on manual object selection and label assignment, making the process highly time-consuming. Additionally, the methods often face challenges in the field of process planning, including low domain-specific terminology recognition rates and incomplete entity extraction. To tackle these challenges, this paper introduces a hybrid approach that integrates large and small language models to construct an aircraft process planning knowledge graph. Initially, clustering-based multi-agent approach is employed to pre-annotate the process planning dataset, with domain experts re-annotate the defect data to create a high-quality process planning dataset. Subsequently, a knowledge extraction framework for aircraft process planning, KE-LSM, was constructed using the small language model trained on this dataset, together with the LLM. Experimental results show that KE-LSM outperforms existing named entity recognition models. Finally, KE-LSM is applied in a commercial aircraft manufacturing company, accompanied by the development of a prototype system designed to facilitate intelligent process planning. It is hoped that the research can provide valuable insights and support for the application of LLM-based solutions in the field of aircraft manufacturing.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 1-19"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145594700","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}
Pub Date : 2026-02-01Epub Date: 2025-12-17DOI: 10.1016/j.jmsy.2025.12.006
Jiarui Xie , Zhuo Yang , Haw-Ching Yang , Yan Lu , Yaoyao Fiona Zhao
Advanced manufacturing is increasingly supported by machine learning (ML)-based digital twins (DTs) for real-time process monitoring and quality assurance. However, changes in physical domain configurations (e.g., machines, materials, and sensors) often cause domain shifts, limiting the reusability of existing DT components. Rebuilding DTs from scratch for each new configuration is costly and time-consuming. To address this challenge, we define DT reusability through three key criteria: FAIRness (findability, accessibility, interoperability, and reusability), transferability, and transfer efficiency. We propose a framework to systematically support the reuse of ML-based process modeling components in DTs, consisting of three phases: FAIR compliance, transferability analysis, and domain adaptation. To enhance transfer efficiency, we introduce the domain-adversarial and decision distribution alignment (DADDA) network, which enables class-conditional alignment and mitigates overfitting through competing domain alignment objectives. A case study on vision-based process monitoring in additive manufacturing was conducted to validate the proposed framework. A FAIR-compliant database of existing DT components was developed, and the most suitable source domain for the designated target domain was identified through an evaluation of semantic and statistical similarity. Leveraging the selected source dataset, DADDA achieved 84 % accuracy after unsupervised pre-training and 96.9 % after supervised fine-tuning with only 210 labeled examples. Further validation on acoustic-based monitoring systems demonstrated the applicability of DADDA to various modalities.
{"title":"On the reusability of machine learning-based process monitoring systems for manufacturing digital twins","authors":"Jiarui Xie , Zhuo Yang , Haw-Ching Yang , Yan Lu , Yaoyao Fiona Zhao","doi":"10.1016/j.jmsy.2025.12.006","DOIUrl":"10.1016/j.jmsy.2025.12.006","url":null,"abstract":"<div><div>Advanced manufacturing is increasingly supported by machine learning (ML)-based digital twins (DTs) for real-time process monitoring and quality assurance. However, changes in physical domain configurations (e.g., machines, materials, and sensors) often cause domain shifts, limiting the reusability of existing DT components. Rebuilding DTs from scratch for each new configuration is costly and time-consuming. To address this challenge, we define DT reusability through three key criteria: FAIRness (findability, accessibility, interoperability, and reusability), transferability, and transfer efficiency. We propose a framework to systematically support the reuse of ML-based process modeling components in DTs, consisting of three phases: FAIR compliance, transferability analysis, and domain adaptation. To enhance transfer efficiency, we introduce the domain-adversarial and decision distribution alignment (DADDA) network, which enables class-conditional alignment and mitigates overfitting through competing domain alignment objectives. A case study on vision-based process monitoring in additive manufacturing was conducted to validate the proposed framework. A FAIR-compliant database of existing DT components was developed, and the most suitable source domain for the designated target domain was identified through an evaluation of semantic and statistical similarity. Leveraging the selected source dataset, DADDA achieved 84 % accuracy after unsupervised pre-training and 96.9 % after supervised fine-tuning with only 210 labeled examples. Further validation on acoustic-based monitoring systems demonstrated the applicability of DADDA to various modalities.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 333-356"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786606","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}
Pub Date : 2026-02-01Epub Date: 2025-12-12DOI: 10.1016/j.jmsy.2025.12.004
Suyoung Park, Shreyes N. Melkote
Conventional supplier selection methods for assembled products have primarily relied on qualitative or business-level assessments of supplier capabilities, since manufacturing-related metrics such as product geometry, cost, time, and tolerance are heterogeneous and difficult to integrate into a unified evaluation. This reliance makes the identification of suppliers with adequate manufacturing capability particularly challenging as global supply chains grow increasingly complex. To address this gap, we propose the Deep Unsupervised Assembly Supplier Matcher (DU-ASM), an integrated data-driven framework that jointly embeds geometry, topology, and quantitative manufacturing attributes into a unified latent space for assembly-level supplier selection and ranking. Leveraging a graph autoencoder, DU-ASM reconstructs manufacturing attributes and supports robust supplier selection even with incomplete inputs. Experimental validation across multiple case studies demonstrates that DU-ASM achieves over 95 % supplier selection accuracy under complete requirements and over 90 % with partially masked inputs, while attaining mean normalized Discounted Cumulative Gain scores at top-k positions (nDCG@k) exceeding 0.99 in ranking tasks. By linking geometric, topological, and quantitative data, DU-ASM demonstrates both methodological novelty and strong quantitative performance, providing a scalable foundation for supplier matching at the assembly level and supporting multi-tier decision-making in future manufacturing supply networks.
{"title":"Deep unsupervised learning-based supplier selection and ranking for assembly manufacturing","authors":"Suyoung Park, Shreyes N. Melkote","doi":"10.1016/j.jmsy.2025.12.004","DOIUrl":"10.1016/j.jmsy.2025.12.004","url":null,"abstract":"<div><div>Conventional supplier selection methods for assembled products have primarily relied on qualitative or business-level assessments of supplier capabilities, since manufacturing-related metrics such as product geometry, cost, time, and tolerance are heterogeneous and difficult to integrate into a unified evaluation. This reliance makes the identification of suppliers with adequate manufacturing capability particularly challenging as global supply chains grow increasingly complex. To address this gap, we propose the Deep Unsupervised Assembly Supplier Matcher (DU-ASM), an integrated data-driven framework that jointly embeds geometry, topology, and quantitative manufacturing attributes into a unified latent space for assembly-level supplier selection and ranking. Leveraging a graph autoencoder, DU-ASM reconstructs manufacturing attributes and supports robust supplier selection even with incomplete inputs. Experimental validation across multiple case studies demonstrates that DU-ASM achieves over 95 % supplier selection accuracy under complete requirements and over 90 % with partially masked inputs, while attaining mean normalized Discounted Cumulative Gain scores at top-k positions (nDCG@k) exceeding 0.99 in ranking tasks. By linking geometric, topological, and quantitative data, DU-ASM demonstrates both methodological novelty and strong quantitative performance, providing a scalable foundation for supplier matching at the assembly level and supporting multi-tier decision-making in future manufacturing supply networks.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 173-188"},"PeriodicalIF":14.2,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786758","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}
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