Pub Date : 2025-12-17DOI: 10.1016/j.jmsy.2025.12.013
Yue Zang , Xiazhen Xu , Feiying Lan , Yongquan Zhang , Huayu Duan , Marco Chacin , Farzaneh Goli , Roger Dixon , Amir M. Hajiyavand , Yongjing Wang
Disassembly is important to circular economy, yet it remains challenging to be robotised due to the inherent uncertainty of end-of-life (EoL) products (e.g., corrosion, rust and missing part). A key challenge in robotising disassembly is that the interference information (e.g., spatial relations of components and assembly methods) is usually unavailable or inaccurate. To address this core problem, this paper presents an object-centric disassembly (OCD) framework, allowing robots to adapt dynamically to varying conditions without requiring prior knowledge of component contacts or interferences. In this framework, an OCD model is constructed in which individual disassembly tasks and their associated conditions are represented as modular units that are continuously refined through autonomous exploration. The performance of the framework is evaluated using a robotic platform integrating intelligent perception, planning, and execution modules for autonomous disassembly under uncertain environments. Experimental evaluations provide evidence that the proposed method enhances the flexibility and adaptability of robotic disassembly. Our approach and this new capability allow disassembly robots to handle real-world uncertainties effectively, eliminating the need for pre-defined interference information.
{"title":"Disassembly from scratch: An object-centric approach for robotic autonomous disassembly with zero contact/interference information","authors":"Yue Zang , Xiazhen Xu , Feiying Lan , Yongquan Zhang , Huayu Duan , Marco Chacin , Farzaneh Goli , Roger Dixon , Amir M. Hajiyavand , Yongjing Wang","doi":"10.1016/j.jmsy.2025.12.013","DOIUrl":"10.1016/j.jmsy.2025.12.013","url":null,"abstract":"<div><div>Disassembly is important to circular economy, yet it remains challenging to be robotised due to the inherent uncertainty of end-of-life (EoL) products (e.g., corrosion, rust and missing part). A key challenge in robotising disassembly is that the interference information (e.g., spatial relations of components and assembly methods) is usually unavailable or inaccurate. To address this core problem, this paper presents an object-centric disassembly (OCD) framework, allowing robots to adapt dynamically to varying conditions without requiring prior knowledge of component contacts or interferences. In this framework, an OCD model is constructed in which individual disassembly tasks and their associated conditions are represented as modular units that are continuously refined through autonomous exploration. The performance of the framework is evaluated using a robotic platform integrating intelligent perception, planning, and execution modules for autonomous disassembly under uncertain environments. Experimental evaluations provide evidence that the proposed method enhances the flexibility and adaptability of robotic disassembly. Our approach and this new capability allow disassembly robots to handle real-world uncertainties effectively, eliminating the need for pre-defined interference information.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 314-332"},"PeriodicalIF":14.2,"publicationDate":"2025-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786607","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 : 2025-12-17DOI: 10.1016/j.jmsy.2025.12.016
Changchun Liu , Dunbing Tang , Haihua Zhu , Liping Wang , Qixiang Cai , Qingwei Nie
As the manufacturing industry evolves towards greater intelligence and flexibility, multi-agent manufacturing systems encounter critical challenges, including frequent dynamic disruptions, inefficient inter-agent collaboration, and underutilized manufacturing knowledge. To address these issues, this paper proposes an LLM-enhanced embodied multi-agent system that integrates embodied perception, analysis, and decision-making to establish a novel self-organizing production paradigm. First, an LLM-driven embodied machine agent is developed. Through the precise mapping of physical entities to internal functional modules, these agents are endowed with the capability for context-aware comprehension of manufacturing domain information. Second, an LLM-enhanced multimodal embodied perception mechanism is designed. By deeply integrating continuous data acquisition with implicit domain knowledge, this mechanism equips the system with the sensory capabilities necessary to capture dynamic disturbances in real time. Building on this, an LLM-driven embodied analysis method is developed for dynamic disturbances. This method, which involves systematic data preprocessing, domain knowledge-integrated multimodal data correlation analysis, and predictive model construction, forms the system’s core ability to identify production schedule anomalies and predict failure trends. Finally, an LLM-integrated embodied decision-making framework is established. This framework balances local autonomy with global goal consensus through self-organizing negotiation and dynamic game mechanisms. It further integrates human problem-solving expertise with the efficiency of machine intelligence to generate optimal production strategies, thereby providing the foundational support for high-level autonomous collaboration among embodied agents. Experimental results demonstrate that this self-organizing mode supported by LLM-enhanced embodied agents can achieve a faster response speed than conventional decision-making methods (e.g., Deep Q-Network) and standalone LLMs (e.g., GPT-4). The proposed system effectively overcomes bottlenecks throughout the perception-analysis-decision-making process, successfully establishing a novel self-organizing production paradigm.
{"title":"LLM-enhanced embodied multi-agent manufacturing system: A novel self-organizing production paradigm for embodied perception, embodied analysis and embodied decision","authors":"Changchun Liu , Dunbing Tang , Haihua Zhu , Liping Wang , Qixiang Cai , Qingwei Nie","doi":"10.1016/j.jmsy.2025.12.016","DOIUrl":"10.1016/j.jmsy.2025.12.016","url":null,"abstract":"<div><div>As the manufacturing industry evolves towards greater intelligence and flexibility, multi-agent manufacturing systems encounter critical challenges, including frequent dynamic disruptions, inefficient inter-agent collaboration, and underutilized manufacturing knowledge. To address these issues, this paper proposes an LLM-enhanced embodied multi-agent system that integrates embodied perception, analysis, and decision-making to establish a novel self-organizing production paradigm. First, an LLM-driven embodied machine agent is developed. Through the precise mapping of physical entities to internal functional modules, these agents are endowed with the capability for context-aware comprehension of manufacturing domain information. Second, an LLM-enhanced multimodal embodied perception mechanism is designed. By deeply integrating continuous data acquisition with implicit domain knowledge, this mechanism equips the system with the sensory capabilities necessary to capture dynamic disturbances in real time. Building on this, an LLM-driven embodied analysis method is developed for dynamic disturbances. This method, which involves systematic data preprocessing, domain knowledge-integrated multimodal data correlation analysis, and predictive model construction, forms the system’s core ability to identify production schedule anomalies and predict failure trends. Finally, an LLM-integrated embodied decision-making framework is established. This framework balances local autonomy with global goal consensus through self-organizing negotiation and dynamic game mechanisms. It further integrates human problem-solving expertise with the efficiency of machine intelligence to generate optimal production strategies, thereby providing the foundational support for high-level autonomous collaboration among embodied agents. Experimental results demonstrate that this self-organizing mode supported by LLM-enhanced embodied agents can achieve a faster response speed than conventional decision-making methods (e.g., Deep Q-Network) and standalone LLMs (e.g., GPT-4). The proposed system effectively overcomes bottlenecks throughout the perception-analysis-decision-making process, successfully establishing a novel self-organizing production paradigm.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 357-382"},"PeriodicalIF":14.2,"publicationDate":"2025-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786608","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 : 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":"2025-12-17","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 : 2025-12-16DOI: 10.1016/j.jmsy.2025.12.001
Xun Xu , Tang Ji , Pai Zheng , Lihui Wang
Human-Centric Manufacturing (HCM) stresses inclusion, resilience, and well-being. Recent studies focus on supporting workers on the factory floor, assuming that human presence in production will remain for the foreseeable future. Meanwhile, automation and artificial intelligence (AI) are rapidly transforming manufacturing and redefining human roles. This paper reviews automation trajectories, analyses the evolving roles of humans, and discusses the technological and social factors shaping future manufacturing. Our discussion suggests that human roles will decline and change, but not disappear anytime soon. HCM should evolve from focusing on physical presence to embedding human purpose in advanced and engaging manufacturing systems.
{"title":"Human-centric manufacturing: Re-thinking, Re-justifying, and Re-envisioning","authors":"Xun Xu , Tang Ji , Pai Zheng , Lihui Wang","doi":"10.1016/j.jmsy.2025.12.001","DOIUrl":"10.1016/j.jmsy.2025.12.001","url":null,"abstract":"<div><div>Human-Centric Manufacturing (HCM) stresses inclusion, resilience, and well-being. Recent studies focus on supporting workers on the factory floor, assuming that human presence in production will remain for the foreseeable future. Meanwhile, automation and artificial intelligence (AI) are rapidly transforming manufacturing and redefining human roles. This paper reviews automation trajectories, analyses the evolving roles of humans, and discusses the technological and social factors shaping future manufacturing. Our discussion suggests that human roles will decline and change, but not disappear anytime soon. HCM should evolve from focusing on physical presence to embedding human purpose in advanced and engaging manufacturing systems.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 259-268"},"PeriodicalIF":14.2,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786612","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 : 2025-12-16DOI: 10.1016/j.jmsy.2025.12.002
Junjie Hu , Wei Zhao , Liang Li , Ning He , Muhammad Jamil , Aqib Mashood Khan , Chao Wang , Xiaowei Zheng
The machining processes must achieve sustainability due to growing ecological concerns and energy crises. As an effective tool, sustainability assessment guides the implementation of sustainable strategies in machining processes. However, the complex resource consumption across machining processes and the coupling effects among machining parameters have greatly hindered its implementation. To address this challenge, this study proposes a sustainability assessment method based on progressive analysis of critical sources, enabling the identification and evaluation of key factors influencing sustainability. First, critical sources are identified through quantification of contribution degrees and sensitivity analysis. Progressive analysis is employed to focus resources on in-depth research into the fundamental characteristics and operational mechanisms of critical sources, establishing specialized indicators such as specific embodied energy and specific carbon emissions for cutters. Subsequently, a sustainable soft sensor is developed to enable efficient and cost-effective sustainability assessment. Finally, a milling case study incorporating various tool types and cooling-lubrication strategies demonstrates the method’s effectiveness in comprehensively capturing the coupling effects inherent in machining processes. The results confirm the method’s reliability and clearly validate its capability to evaluate sustainability performance in machining. This study not only provides technical support for sustainability assessments but also delivers actionable insights to facilitate the implementation of sustainable machining strategies.
{"title":"Sustainability assessment for machining processes based on progressive analysis of critical sources","authors":"Junjie Hu , Wei Zhao , Liang Li , Ning He , Muhammad Jamil , Aqib Mashood Khan , Chao Wang , Xiaowei Zheng","doi":"10.1016/j.jmsy.2025.12.002","DOIUrl":"10.1016/j.jmsy.2025.12.002","url":null,"abstract":"<div><div>The machining processes must achieve sustainability due to growing ecological concerns and energy crises. As an effective tool, sustainability assessment guides the implementation of sustainable strategies in machining processes. However, the complex resource consumption across machining processes and the coupling effects among machining parameters have greatly hindered its implementation. To address this challenge, this study proposes a sustainability assessment method based on progressive analysis of critical sources, enabling the identification and evaluation of key factors influencing sustainability. First, critical sources are identified through quantification of contribution degrees and sensitivity analysis. Progressive analysis is employed to focus resources on in-depth research into the fundamental characteristics and operational mechanisms of critical sources, establishing specialized indicators such as specific embodied energy and specific carbon emissions for cutters. Subsequently, a sustainable soft sensor is developed to enable efficient and cost-effective sustainability assessment. Finally, a milling case study incorporating various tool types and cooling-lubrication strategies demonstrates the method’s effectiveness in comprehensively capturing the coupling effects inherent in machining processes. The results confirm the method’s reliability and clearly validate its capability to evaluate sustainability performance in machining. This study not only provides technical support for sustainability assessments but also delivers actionable insights to facilitate the implementation of sustainable machining strategies.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 289-313"},"PeriodicalIF":14.2,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786667","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 : 2025-12-16DOI: 10.1016/j.jmsy.2025.12.009
Xin Guo , Mingyue Yang , Pai Zheng , Jiewu Leng , Chong Chen , Kai Zhang , Jun Li , Zechuan Huang
High-impact disruptions can cause significant performance degradation and even failures in manufacturing systems. Resilient manufacturing systems can absorb such disruptions, adapt to changing environments, and accelerate recovery through strategy scheduling based on real-time performance data. However, the nonlinear nature of degradation processes can lead to deviations from expected recovery outcomes and delays in strategy scheduling, which makes strategy scheduling for repairing manufacturing systems a difficult decision-making problem. Therefore, a resilience-enhancing multi-strategy decision-making for dynamic scheduling model in manufacturing systems is proposed, aiming to determine the optimal strategy and reduce performance anomaly duration. First, a component-based evaluation method is proposed to measure the absorption, adaptation, and recovery capabilities of the system, achieving real-time analysis of resilience levels. Then, a dynamic strategy scheduling method based on Markov chains is proposed to plan strategies and predict trajectories based on the real-time performance status, disruption, and resilience level, which solves the nonlinearity changes of performance state. Finally, a multi-strategy decision-making method based on fuzzy-BWM is proposed to achieve the resilient-oriented multi-objective discrete strategy decision-making, considering cost, recovery time, and recovery degree. The die forging press is used to demonstrate the effectiveness of the proposed model. The results show that the strategy decided by the model enables the system to recover quickly to its expected state with an acceptable cost compared to other strategies.
{"title":"Resilience-enhancing multi-strategy decision-making for dynamic scheduling in manufacturing systems","authors":"Xin Guo , Mingyue Yang , Pai Zheng , Jiewu Leng , Chong Chen , Kai Zhang , Jun Li , Zechuan Huang","doi":"10.1016/j.jmsy.2025.12.009","DOIUrl":"10.1016/j.jmsy.2025.12.009","url":null,"abstract":"<div><div>High-impact disruptions can cause significant performance degradation and even failures in manufacturing systems. Resilient manufacturing systems can absorb such disruptions, adapt to changing environments, and accelerate recovery through strategy scheduling based on real-time performance data. However, the nonlinear nature of degradation processes can lead to deviations from expected recovery outcomes and delays in strategy scheduling, which makes strategy scheduling for repairing manufacturing systems a difficult decision-making problem. Therefore, a resilience-enhancing multi-strategy decision-making for dynamic scheduling model in manufacturing systems is proposed, aiming to determine the optimal strategy and reduce performance anomaly duration. First, a component-based evaluation method is proposed to measure the absorption, adaptation, and recovery capabilities of the system, achieving real-time analysis of resilience levels. Then, a dynamic strategy scheduling method based on Markov chains is proposed to plan strategies and predict trajectories based on the real-time performance status, disruption, and resilience level, which solves the nonlinearity changes of performance state. Finally, a multi-strategy decision-making method based on fuzzy-BWM is proposed to achieve the resilient-oriented multi-objective discrete strategy decision-making, considering cost, recovery time, and recovery degree. The die forging press is used to demonstrate the effectiveness of the proposed model. The results show that the strategy decided by the model enables the system to recover quickly to its expected state with an acceptable cost compared to other strategies.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 269-288"},"PeriodicalIF":14.2,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786665","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 : 2025-12-16DOI: 10.1016/j.jmsy.2025.12.012
Xiaomeng Zhu , Jacob Henningsson , Pär Mårtensson , Lars Hanson , Mårten Björkman , Atsuto Maki
{"title":"Corrigendum to “Designing Synthetic Active Learning for model refinement in manufacturing parts detection [Volume 84, February 2026, Pages 68–84]”","authors":"Xiaomeng Zhu , Jacob Henningsson , Pär Mårtensson , Lars Hanson , Mårten Björkman , Atsuto Maki","doi":"10.1016/j.jmsy.2025.12.012","DOIUrl":"10.1016/j.jmsy.2025.12.012","url":null,"abstract":"","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Page 646"},"PeriodicalIF":14.2,"publicationDate":"2025-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145924722","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 : 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":"2025-12-15","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 : 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":"2025-12-15","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 : 2025-12-15DOI: 10.1016/j.jmsy.2025.11.025
Yezhen Peng , Weimin Kang , Qirui Hu , Fengwen Yu , Wenhong Zhou , Xinhua Yao , Congcong Luan , Songyu Hu , Jianzhong Fu
Tool wear monitoring is crucial for optimizing CNC machining processes in next-generation intelligent manufacturing systems. However, existing methods struggle to capture the dynamic relationship between high-frequency features and wear evolution. Small-sample training and the uneven distribution of labels across the domain exacerbate bias in feature migration, limiting model generalizability and adaptability. To address this, a frequency domain-aware and bionic-aligned collaborative modeling approach for domain shift mitigation is proposed. Firstly, a smoothed wavelet convolution feature extraction method is introduced, enhancing the capture of sensitive frequency bands and stabilizing gradient propagation through a Softplus smoothing mechanism. The method’s ability to suppress domain offset during the initial feature extraction stage is validated by comparing feature activation distributions across two domains. Inspired by bat echolocation, an attention mechanism is proposed that integrates energy guidance, echo alignment, and time-frequency focusing modules to enhance high-frequency signal mapping and mitigate domain shift. The method's effectiveness in high-frequency feature response is validated through enhancement metrics and variance distribution within the attention focus region. Additionally, interpretability of dual-domain feature alignment is improved by calculating working condition similarity, integrating a priori knowledge, and optimizing the MMD loss function. Systematic ablation experiments demonstrate that the proposed method achieves average RMSE, MAE, and R² values of 0.078, 0.063, and 0.817, respectively. It outperforms all ablation models, yielding average reductions of 31.6 % and 32.5 % in RMSE and MAE, and an average improvement of 42.7 % in R². Furthermore, the proposed method outperforms the best-performing method among the four mainstream methods, reducing RMSE and MAE by 13.3 % and 2.5 %, and improving R² by 5.1 %. This method effectively suppresses domain bias in feature extraction, mapping, and training under small sample conditions, providing critical technical support for intelligent manufacturing in complex, variable working environments.
{"title":"Frequency-aware and bionic-aligned collaborative modeling for cross-domain tool wear monitoring under small-sample conditions","authors":"Yezhen Peng , Weimin Kang , Qirui Hu , Fengwen Yu , Wenhong Zhou , Xinhua Yao , Congcong Luan , Songyu Hu , Jianzhong Fu","doi":"10.1016/j.jmsy.2025.11.025","DOIUrl":"10.1016/j.jmsy.2025.11.025","url":null,"abstract":"<div><div>Tool wear monitoring is crucial for optimizing CNC machining processes in next-generation intelligent manufacturing systems. However, existing methods struggle to capture the dynamic relationship between high-frequency features and wear evolution. Small-sample training and the uneven distribution of labels across the domain exacerbate bias in feature migration, limiting model generalizability and adaptability. To address this, a frequency domain-aware and bionic-aligned collaborative modeling approach for domain shift mitigation is proposed. Firstly, a smoothed wavelet convolution feature extraction method is introduced, enhancing the capture of sensitive frequency bands and stabilizing gradient propagation through a Softplus smoothing mechanism. The method’s ability to suppress domain offset during the initial feature extraction stage is validated by comparing feature activation distributions across two domains. Inspired by bat echolocation, an attention mechanism is proposed that integrates energy guidance, echo alignment, and time-frequency focusing modules to enhance high-frequency signal mapping and mitigate domain shift. The method's effectiveness in high-frequency feature response is validated through enhancement metrics and variance distribution within the attention focus region. Additionally, interpretability of dual-domain feature alignment is improved by calculating working condition similarity, integrating a priori knowledge, and optimizing the MMD loss function. Systematic ablation experiments demonstrate that the proposed method achieves average RMSE, MAE, and R² values of 0.078, 0.063, and 0.817, respectively. It outperforms all ablation models, yielding average reductions of 31.6 % and 32.5 % in RMSE and MAE, and an average improvement of 42.7 % in R². Furthermore, the proposed method outperforms the best-performing method among the four mainstream methods, reducing RMSE and MAE by 13.3 % and 2.5 %, and improving R² by 5.1 %. This method effectively suppresses domain bias in feature extraction, mapping, and training under small sample conditions, providing critical technical support for intelligent manufacturing in complex, variable working environments.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"84 ","pages":"Pages 223-241"},"PeriodicalIF":14.2,"publicationDate":"2025-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145786669","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