Journal of Manufacturing Systems最新文献_第5页

Data-driven inpainting for full-part temperature monitoring in additive manufacturing 数据驱动的喷涂，用于增材制造中的全零件温度监测

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-19 DOI: 10.1016/j.jmsy.2024.09.022

Jiangce Chen , Mikhail Khrenov , Jiayi Jin , Sneha Prabha Narra , Christopher McComb

Understanding the temperature history over a part during additive manufacturing (AM) is important for optimizing the process and ensuring product quality, as temperature impacts melt pool geometry, defect formation, and microstructure evolution. While in-process temperature monitoring holds promise for evaluating the part quality, existing thermal sensors used in AM provide only partial measurements of the temperature distribution over the part. In this work, we introduce an innovative approach for reconstructing the complete temperature profile using partial data. We formulate this challenge as an inpainting problem, a canonical task in machine learning which entails recovering missing information across a spatial domain. We present a data-driven model based on graph convolutional neural networks. To train the inpainting model, we employ a finite element simulation to generate a diverse dataset of temperature histories for various part geometries. Cross-validation indicates that the inpainting model accurately reconstructs the spatial distribution of part temperature with strong generalizability across various geometries. Further application to experimental data using infrared camera measurements shows that the model accuracy could be improved by augmenting the training data with simulation data that shares process parameters and geometry with the experimental part. By presenting a solution to the temperature inpainting problem, our approach not only improves the assessment of part quality using partial measurements but also paves the way for the creation of a temperature digital twin of the part using thermal sensors.

在增材制造（AM）过程中，由于温度会影响熔池几何形状、缺陷形成和微观结构演变，因此了解零件的温度历史对于优化工艺和确保产品质量非常重要。虽然过程中的温度监测有望评估零件质量，但现有的热传感器只能部分测量零件的温度分布。在这项工作中，我们介绍了一种利用部分数据重建完整温度曲线的创新方法。我们将这一挑战表述为 "涂色 "问题，这是机器学习中的一项典型任务，需要恢复空间领域中缺失的信息。我们提出了一种基于图卷积神经网络的数据驱动模型。为了训练嵌绘模型，我们采用有限元模拟，为不同的零件几何形状生成不同的温度历史数据集。交叉验证结果表明，涂色模型能够准确地重建零件温度的空间分布，并在各种几何形状中具有很强的通用性。利用红外相机测量实验数据的进一步应用表明，通过使用与实验部件共享工艺参数和几何形状的模拟数据来增强训练数据，可以提高模型的准确性。通过提出温度涂抹问题的解决方案，我们的方法不仅改进了利用部分测量数据对零件质量的评估，还为利用热传感器创建零件的温度数字孪生模型铺平了道路。

{"title":"Data-driven inpainting for full-part temperature monitoring in additive manufacturing","authors":"Jiangce Chen , Mikhail Khrenov , Jiayi Jin , Sneha Prabha Narra , Christopher McComb","doi":"10.1016/j.jmsy.2024.09.022","DOIUrl":"10.1016/j.jmsy.2024.09.022","url":null,"abstract":"<div><div>Understanding the temperature history over a part during additive manufacturing (AM) is important for optimizing the process and ensuring product quality, as temperature impacts melt pool geometry, defect formation, and microstructure evolution. While in-process temperature monitoring holds promise for evaluating the part quality, existing thermal sensors used in AM provide only partial measurements of the temperature distribution over the part. In this work, we introduce an innovative approach for reconstructing the complete temperature profile using partial data. We formulate this challenge as an inpainting problem, a canonical task in machine learning which entails recovering missing information across a spatial domain. We present a data-driven model based on graph convolutional neural networks. To train the inpainting model, we employ a finite element simulation to generate a diverse dataset of temperature histories for various part geometries. Cross-validation indicates that the inpainting model accurately reconstructs the spatial distribution of part temperature with strong generalizability across various geometries. Further application to experimental data using infrared camera measurements shows that the model accuracy could be improved by augmenting the training data with simulation data that shares process parameters and geometry with the experimental part. By presenting a solution to the temperature inpainting problem, our approach not only improves the assessment of part quality using partial measurements but also paves the way for the creation of a temperature digital twin of the part using thermal sensors.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 558-575"},"PeriodicalIF":12.2,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142535597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Study on the application of single-agent and multi-agent reinforcement learning to dynamic scheduling in manufacturing environments with growing complexity: Case study on the synthesis of an industrial IoT Test Bed 研究在复杂性不断增加的制造环境中，将单机和多机强化学习应用于动态调度：工业物联网测试平台综合案例研究

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-19 DOI: 10.1016/j.jmsy.2024.09.019

David Heik, Fouad Bahrpeyma, Dirk Reichelt

Industry 4.0, smart manufacturing and smart products have recently attracted substantial attention and are becoming increasingly prevalent in manufacturing systems. As a result of the successful implementation of these technologies, highly customized products can be manufactured using responsive, autonomous manufacturing processes at a competitive cost. This study was conducted at HTW Dresden’s Industrial Internet of Things Test Bed, which simulates state-of-the-art manufacturing scenarios for educational and research purposes. Apart from the physical production facility itself, the associated operational information systems have been fully interconnected in order to allow fast and efficient information exchange between the various manufacturing stages and systems. The presence of this characteristic provides a strong foundation for dealing appropriately with unexpected or planned environmental changes, as well as prevailing uncertainty, which greatly increases the overall system’s resilience. The main objective of this study is to increase the efficiency of the manufacturing system in order to optimize resource consumption and minimize the overall completion time (makespan). This manuscript discusses our experiments in the area of flexible job-shop scheduling problems (FJSP). As part of our research, different methods of representing the state space were explored, heuristic, meta-heuristic, reinforcement learning (RL), and multi-agent reinforcement learning (MARL) methods were evaluated, and various methods of interaction with the system (designing the action space and filtering in certain situations) were examined. Furthermore, the design of the reward function, which plays an important role in the formulation of the dynamic scheduling problem into an RL problem, has been discussed in depth. Finally, this paper studies the effectiveness of single-agent and multi-agent RL approaches, with a special focus on the Proximal Policy Optimization (PPO) method, on the fully-fledged digital twin of an industrial IoT system at HTW Dresden. As a result of our experiments, in a multi-agent setting involving individual agents for each manufacturing operation, PPO was able to manage the resources in such a way as to improve the manufacturing system’s performance significantly.

工业 4.0、智能制造和智能产品最近引起了广泛关注，并在制造系统中日益普及。由于这些技术的成功实施，高度定制化的产品可以通过反应灵敏、自主的制造流程以具有竞争力的成本生产出来。本研究在 HTW 德累斯顿工业物联网试验台进行，该试验台模拟最先进的制造场景，用于教育和研究目的。除了物理生产设施本身之外，相关的操作信息系统也已完全互联，以便在各个生产阶段和系统之间快速、高效地交换信息。这一特点为妥善处理意外或计划中的环境变化以及普遍存在的不确定性奠定了坚实的基础，从而大大提高了整个系统的应变能力。本研究的主要目标是提高制造系统的效率，以优化资源消耗并最大限度地缩短整体完工时间（makespan）。本手稿讨论了我们在灵活作业调度问题（FJSP）领域的实验。作为研究的一部分，我们探索了表示状态空间的不同方法，评估了启发式、元启发式、强化学习（RL）和多代理强化学习（MARL）方法，并研究了与系统交互的各种方法（设计行动空间和在某些情况下进行过滤）。此外，本文还深入讨论了在将动态调度问题表述为 RL 问题时起重要作用的奖励函数的设计。最后，本文在德累斯顿 HTW 工业物联网系统的成熟数字孪生系统上研究了单机和多机 RL 方法的有效性，并特别关注了近端策略优化 (PPO) 方法。实验结果表明，在涉及每个制造操作的单个代理的多代理环境中，PPO 能够以显著提高制造系统性能的方式管理资源。

{"title":"Study on the application of single-agent and multi-agent reinforcement learning to dynamic scheduling in manufacturing environments with growing complexity: Case study on the synthesis of an industrial IoT Test Bed","authors":"David Heik, Fouad Bahrpeyma, Dirk Reichelt","doi":"10.1016/j.jmsy.2024.09.019","DOIUrl":"10.1016/j.jmsy.2024.09.019","url":null,"abstract":"<div><div>Industry 4.0, smart manufacturing and smart products have recently attracted substantial attention and are becoming increasingly prevalent in manufacturing systems. As a result of the successful implementation of these technologies, highly customized products can be manufactured using responsive, autonomous manufacturing processes at a competitive cost. This study was conducted at HTW Dresden’s Industrial Internet of Things Test Bed, which simulates state-of-the-art manufacturing scenarios for educational and research purposes. Apart from the physical production facility itself, the associated operational information systems have been fully interconnected in order to allow fast and efficient information exchange between the various manufacturing stages and systems. The presence of this characteristic provides a strong foundation for dealing appropriately with unexpected or planned environmental changes, as well as prevailing uncertainty, which greatly increases the overall system’s resilience. The main objective of this study is to increase the efficiency of the manufacturing system in order to optimize resource consumption and minimize the overall completion time (makespan). This manuscript discusses our experiments in the area of flexible job-shop scheduling problems (FJSP). As part of our research, different methods of representing the state space were explored, heuristic, meta-heuristic, reinforcement learning (RL), and multi-agent reinforcement learning (MARL) methods were evaluated, and various methods of interaction with the system (designing the action space and filtering in certain situations) were examined. Furthermore, the design of the reward function, which plays an important role in the formulation of the dynamic scheduling problem into an RL problem, has been discussed in depth. Finally, this paper studies the effectiveness of single-agent and multi-agent RL approaches, with a special focus on the Proximal Policy Optimization (PPO) method, on the fully-fledged digital twin of an industrial IoT system at HTW Dresden. As a result of our experiments, in a multi-agent setting involving individual agents for each manufacturing operation, PPO was able to manage the resources in such a way as to improve the manufacturing system’s performance significantly.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 525-557"},"PeriodicalIF":12.2,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142535592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Industrial data space application framework for semiconductor wafer manufacturing system scheduling 用于半导体晶片制造系统调度的工业数据空间应用框架

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-18 DOI: 10.1016/j.jmsy.2024.09.013

Da Chen , Jie Zhang , Lihui Wu , Peng Zhang , Ming Wang

The complex, large-scale semiconductor wafer manufacturing generates substantial diverse data, creating management hurdles and making efficient use of historical scheduling data difficult. To address these challenges, we propose a four-layer application framework for industrial data space for wafer manufacturing system (IDWFS). Firstly, a multi-level model ontology centred on scheduling tasks is constructed to effectively map the evolution of elemental relationships during wafer processing and adaptively change the data organisation. Then, a system architecture for mining the correlation between dynamic and static element data is proposed to fully explore the spatiotemporal correlation relationship of data elements in the processing process. Finally, a scheduling system architecture of “learning + prediction + scheduling” is proposed to fully utilise the scheduling historical domain knowledge and data correlation relationship in semiconductor wafer manufacturing system during the scheduling process. In addition, through three case studies related to the scheduling of semiconductor wafer manufacturing system, IDWFS is effective in heterogeneous data management, coupling relationship mining of element data, logistics scheduling processing, etc., thereby achieving logistics scheduling control of wafer manufacturing system.

复杂的大规模半导体晶圆制造会产生大量不同的数据，从而造成管理上的障碍，使有效利用历史调度数据变得困难。为应对这些挑战，我们提出了晶圆制造系统工业数据空间（IDWFS）的四层应用框架。首先，我们构建了以调度任务为中心的多层次模型本体，以有效映射晶圆加工过程中元素关系的演变，并自适应地改变数据组织。然后，提出了一种挖掘动态和静态元素数据关联性的系统架构，以充分挖掘加工过程中数据元素的时空关联关系。最后，提出了一种 "学习+预测+调度 "的调度系统架构，以充分利用半导体晶圆制造系统在调度过程中的调度历史领域知识和数据关联关系。此外，通过三个与半导体晶圆制造系统调度相关的案例研究，IDWFS在异构数据管理、要素数据耦合关系挖掘、物流调度处理等方面效果显著，从而实现了晶圆制造系统的物流调度控制。

{"title":"Industrial data space application framework for semiconductor wafer manufacturing system scheduling","authors":"Da Chen , Jie Zhang , Lihui Wu , Peng Zhang , Ming Wang","doi":"10.1016/j.jmsy.2024.09.013","DOIUrl":"10.1016/j.jmsy.2024.09.013","url":null,"abstract":"<div><div>The complex, large-scale semiconductor wafer manufacturing generates substantial diverse data, creating management hurdles and making efficient use of historical scheduling data difficult. To address these challenges, we propose a four-layer application framework for industrial data space for wafer manufacturing system (IDWFS). Firstly, a multi-level model ontology centred on scheduling tasks is constructed to effectively map the evolution of elemental relationships during wafer processing and adaptively change the data organisation. Then, a system architecture for mining the correlation between dynamic and static element data is proposed to fully explore the spatiotemporal correlation relationship of data elements in the processing process. Finally, a scheduling system architecture of “learning + prediction + scheduling” is proposed to fully utilise the scheduling historical domain knowledge and data correlation relationship in semiconductor wafer manufacturing system during the scheduling process. In addition, through three case studies related to the scheduling of semiconductor wafer manufacturing system, IDWFS is effective in heterogeneous data management, coupling relationship mining of element data, logistics scheduling processing, etc., thereby achieving logistics scheduling control of wafer manufacturing system.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 464-482"},"PeriodicalIF":12.2,"publicationDate":"2024-10-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142535590","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}

引用次数: 0

Federated domain generalization for condition monitoring in ultrasonic metal welding 用于超声波金属焊接状态监测的联邦域泛化

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-17 DOI: 10.1016/j.jmsy.2024.09.023

Ahmadreza Eslaminia , Yuquan Meng , Klara Nahrstedt , Chenhui Shao

Ultrasonic metal welding (UMW) is a key joining technology with widespread industrial applications. Condition monitoring (CM) capabilities are critically needed in UMW applications because process anomalies, such as tool degradation and workpiece surface contamination, significantly deteriorate the joining quality. Recently, machine learning models emerged as a promising tool for CM in many manufacturing applications. Yet, many existing models lack the generalizability or adaptability and cannot be directly applied to new manufacturing process configurations (i.e., domains). Although several domain generalization techniques have been proposed, their successful deployment often requires substantial training data, which can be expensive and time-consuming to collect in a single factory. Such issues may be potentially alleviated by pooling data across factories, but data sharing raises critical data privacy concerns that have prohibited data sharing for collaborative model training in the industry. To address these challenges, this paper presents a Federated Domain Generalization for Condition Monitoring (FDG-CM) framework that provides domain generalization capabilities in distributed learning while ensuring data privacy. By effectively learning a unified representation from the feature space, FDG-CM can adapt CM models for new clients (factories) with different process configurations. To demonstrate the effectiveness of FDG-CM, we investigate two distinct UMW CM tasks, including tool condition monitoring and workpiece surface condition classification. Compared with state-of-the-art federated learning algorithms, FDG-CM achieves a 5.35%–8.08% improvement in CM accuracy. FDG-CM is also shown to achieve excellent performance in challenging scenarios involving unbalanced data distributions and limited participating clients. Furthermore, by implementing the FDG-CM method on an edge–cloud architecture, we show that this method is both viable and efficient in practice. The FDG-CM framework is readily extensible to other manufacturing applications.

超声波金属焊接（UMW）是一种关键的连接技术，在工业领域有着广泛的应用。由于工具退化和工件表面污染等工艺异常会严重影响焊接质量，因此 UMW 应用中亟需状态监测 (CM) 功能。最近，机器学习模型作为一种有前途的工具出现在许多制造应用中。然而，许多现有模型缺乏通用性或适应性，无法直接应用于新的制造工艺配置（即领域）。虽然已经提出了几种领域泛化技术，但成功应用这些技术往往需要大量的训练数据，而在单个工厂收集这些数据可能既昂贵又耗时。通过汇集各工厂的数据，这些问题可能会得到缓解，但数据共享会引发关键的数据隐私问题，这就禁止了行业内用于协作模型训练的数据共享。为了应对这些挑战，本文提出了一种用于状态监测的联邦领域泛化（FDG-CM）框架，在分布式学习中提供领域泛化功能，同时确保数据隐私。通过有效学习特征空间的统一表示，FDG-CM 可以针对具有不同流程配置的新客户（工厂）调整 CM 模型。为了证明 FDG-CM 的有效性，我们研究了两个不同的 UMW CM 任务，包括刀具状态监测和工件表面状态分类。与最先进的联合学习算法相比，FDG-CM 的 CM 准确率提高了 5.35%-8.08%。研究还表明，FDG-CM 在数据分布不平衡、参与客户有限等具有挑战性的情况下也能取得优异的性能。此外，通过在边缘云架构上实施 FDG-CM 方法，我们表明该方法在实践中既可行又高效。FDG-CM 框架可随时扩展到其他制造应用。

{"title":"Federated domain generalization for condition monitoring in ultrasonic metal welding","authors":"Ahmadreza Eslaminia , Yuquan Meng , Klara Nahrstedt , Chenhui Shao","doi":"10.1016/j.jmsy.2024.09.023","DOIUrl":"10.1016/j.jmsy.2024.09.023","url":null,"abstract":"<div><div>Ultrasonic metal welding (UMW) is a key joining technology with widespread industrial applications. Condition monitoring (CM) capabilities are critically needed in UMW applications because process anomalies, such as tool degradation and workpiece surface contamination, significantly deteriorate the joining quality. Recently, machine learning models emerged as a promising tool for CM in many manufacturing applications. Yet, many existing models lack the generalizability or adaptability and cannot be directly applied to new manufacturing process configurations (i.e., domains). Although several domain generalization techniques have been proposed, their successful deployment often requires substantial training data, which can be expensive and time-consuming to collect in a single factory. Such issues may be potentially alleviated by pooling data across factories, but data sharing raises critical data privacy concerns that have prohibited data sharing for collaborative model training in the industry. To address these challenges, this paper presents a Federated Domain Generalization for Condition Monitoring (FDG-CM) framework that provides domain generalization capabilities in distributed learning while ensuring data privacy. By effectively learning a unified representation from the feature space, FDG-CM can adapt CM models for new clients (factories) with different process configurations. To demonstrate the effectiveness of FDG-CM, we investigate two distinct UMW CM tasks, including tool condition monitoring and workpiece surface condition classification. Compared with state-of-the-art federated learning algorithms, FDG-CM achieves a 5.35%–8.08% improvement in CM accuracy. FDG-CM is also shown to achieve excellent performance in challenging scenarios involving unbalanced data distributions and limited participating clients. Furthermore, by implementing the FDG-CM method on an edge–cloud architecture, we show that this method is both viable and efficient in practice. The FDG-CM framework is readily extensible to other manufacturing applications.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 1-12"},"PeriodicalIF":12.2,"publicationDate":"2024-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142444772","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}

引用次数: 0

A vision-enabled fatigue-sensitive human digital twin towards human-centric human-robot collaboration 实现以人为本的人机协作的具有视觉功能的疲劳敏感型人类数字孪生系统

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-16 DOI: 10.1016/j.jmsy.2024.10.002

Saahil Chand, Hao Zheng, Yuqian Lu

Within a Human-centric Human-Robot Collaboration (HHRC) system, monitoring, assessing, and optimizing for an operator’s well-being is essential to creating an efficient and comfortable working environment. Currently, monitoring systems are used for independent assessment of human factors. However, the rise of the Human Digital Twin (HDT) has provided the framework for synchronizing multiple operator well-being assessments to create a comprehensive understanding of the operator’s performance and health. Within manufacturing, an operator’s dynamic well-being can be attributed to their physical and cognitive fatigue across the assembly process. As such, we apply non-invasive video understanding techniques to extract relevant assembly process information for automatic physical fatigue assessment. Our novelty involves a video-based fatigue estimation method, in which the boundary-aware dual-stream MS-TCN combined with an LSTM is proposed to detect the operation type, operation repetitions, and the target arm performing each task in an assembly process video. The detected results are then input into our physical fatigue profile to automatically assess the operator’s localized physical fatigue impact. The assembly process of a real-world bookshelf is recorded and tested against, with our algorithm results showing superiority in operation segmentation and target arm detection as opposed to other recent action segmentation models. In addition, we integrate a cognitive fatigue assessment tool that captures operator physiological signals in real-time for body response detection caused by stress. This provides a more robust HDT of the operator for an HHRC system.

在以人为本的人机协作（HHRC）系统中，监测、评估和优化操作员的健康状况对于创造高效舒适的工作环境至关重要。目前，监控系统用于对人为因素进行独立评估。然而，人类数字孪生系统（Human Digital Twin，HDT）的兴起为同步进行多种操作员健康评估提供了框架，以便全面了解操作员的工作表现和健康状况。在制造业中，操作员的动态健康状况可归因于他们在整个装配过程中的身体和认知疲劳。因此，我们应用非侵入式视频理解技术来提取相关装配流程信息，以进行自动身体疲劳评估。我们的新颖之处在于基于视频的疲劳评估方法，其中提出了边界感知双流 MS-TCN 与 LSTM 相结合的方法，用于检测装配过程视频中的操作类型、操作重复次数以及执行每项任务的目标手臂。然后将检测到的结果输入我们的身体疲劳曲线，以自动评估操作员的局部身体疲劳影响。我们录制并测试了真实世界书架的组装过程，结果表明，与其他最新的动作分割模型相比，我们的算法在操作分割和目标手臂检测方面更具优势。此外，我们还集成了认知疲劳评估工具，可实时捕捉操作员的生理信号，以检测压力引起的身体反应。这为人机交互系统提供了更强大的操作员 HDT。

{"title":"A vision-enabled fatigue-sensitive human digital twin towards human-centric human-robot collaboration","authors":"Saahil Chand, Hao Zheng, Yuqian Lu","doi":"10.1016/j.jmsy.2024.10.002","DOIUrl":"10.1016/j.jmsy.2024.10.002","url":null,"abstract":"<div><div>Within a Human-centric Human-Robot Collaboration (HHRC) system, monitoring, assessing, and optimizing for an operator’s well-being is essential to creating an efficient and comfortable working environment. Currently, monitoring systems are used for independent assessment of human factors. However, the rise of the Human Digital Twin (HDT) has provided the framework for synchronizing multiple operator well-being assessments to create a comprehensive understanding of the operator’s performance and health. Within manufacturing, an operator’s dynamic well-being can be attributed to their physical and cognitive fatigue across the assembly process. As such, we apply non-invasive video understanding techniques to extract relevant assembly process information for automatic physical fatigue assessment. Our novelty involves a video-based fatigue estimation method, in which the boundary-aware dual-stream MS-TCN combined with an LSTM is proposed to detect the operation type, operation repetitions, and the target arm performing each task in an assembly process video. The detected results are then input into our physical fatigue profile to automatically assess the operator’s localized physical fatigue impact. The assembly process of a real-world bookshelf is recorded and tested against, with our algorithm results showing superiority in operation segmentation and target arm detection as opposed to other recent action segmentation models. In addition, we integrate a cognitive fatigue assessment tool that captures operator physiological signals in real-time for body response detection caused by stress. This provides a more robust HDT of the operator for an HHRC system.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 432-445"},"PeriodicalIF":12.2,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142440958","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A prediction method of tool wear distribution for ball-end milling under various postures based on WVEM-T 基于 WVEM-T 的各种姿态下球端铣削刀具磨损分布预测方法

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-16 DOI: 10.1016/j.jmsy.2024.09.017

Xudong Wei , Xianli Liu , Changxia Liu , Anshan Zhang , Zhongran Zhang , Zhitao Chen , Zhiming Gou

The contact positions corresponding to various tool location point during ball-end milling are complex, and the actual cutting area of flank face presents uneven wear form, which is closely related to its effective cutting distance, linear velocity of edge line microelement, and instantaneous undeformed chip thickness, etc. It is difficult to accurately predict the actual tool wear distribution by theoretical modeling. Therefore, it is necessary to put forward a prediction method of tool wear distribution to ensure the quality of workpiece and the stable state of tool during machining. In this paper, the effective cutting length of tool edge line microelement is calculated, and the instantaneous undeformed chip thickness under various postures considering edge wear is determined. A weighted voting ensemble multi-Transformer transfer learning (WVEM-T) model is established, motion parameters and the actual wear widths VB per edge line are used as training data. The selective freezing strategy is adopted to update the training parameters of the network, so that the trained multi-layer network can accurately predict the wear distribution of flank face in ball-end milling tool under various machining inclination angles. Finally, the accuracy and effectiveness of the prediction method in this paper are verified by the whole life cycle experiment of milling Ti6Al4V alloy.

球端铣削过程中各种刀具位置点对应的接触位置复杂，侧面实际切削区域呈现不均匀的磨损形式，这与其有效切削距离、刃线微元线速度、瞬时未变形切屑厚度等密切相关。通过理论建模很难准确预测实际刀具磨损分布。因此，有必要提出一种刀具磨损分布的预测方法，以确保工件质量和刀具在加工过程中的稳定状态。本文计算了刀具刃线微元的有效切削长度，并确定了考虑刃口磨损的各种姿态下的瞬时未变形切屑厚度。建立了一个加权投票集合多变换器迁移学习（WVEM-T）模型，将运动参数和每条刃口线的实际磨损宽度 VB 作为训练数据。采用选择性冻结策略更新网络的训练参数，使训练后的多层网络能准确预测球端铣刀在不同加工倾角下的侧面磨损分布。最后，通过对 Ti6Al4V 合金的全寿命周期铣削实验验证了本文预测方法的准确性和有效性。

{"title":"A prediction method of tool wear distribution for ball-end milling under various postures based on WVEM-T","authors":"Xudong Wei , Xianli Liu , Changxia Liu , Anshan Zhang , Zhongran Zhang , Zhitao Chen , Zhiming Gou","doi":"10.1016/j.jmsy.2024.09.017","DOIUrl":"10.1016/j.jmsy.2024.09.017","url":null,"abstract":"<div><div>The contact positions corresponding to various tool location point during ball-end milling are complex, and the actual cutting area of flank face presents uneven wear form, which is closely related to its effective cutting distance, linear velocity of edge line microelement, and instantaneous undeformed chip thickness, etc. It is difficult to accurately predict the actual tool wear distribution by theoretical modeling. Therefore, it is necessary to put forward a prediction method of tool wear distribution to ensure the quality of workpiece and the stable state of tool during machining. In this paper, the effective cutting length of tool edge line microelement is calculated, and the instantaneous undeformed chip thickness under various postures considering edge wear is determined. A weighted voting ensemble multi-Transformer transfer learning (WVEM-T) model is established, motion parameters and the actual wear widths <em>VB</em> per edge line are used as training data. The selective freezing strategy is adopted to update the training parameters of the network, so that the trained multi-layer network can accurately predict the wear distribution of flank face in ball-end milling tool under various machining inclination angles. Finally, the accuracy and effectiveness of the prediction method in this paper are verified by the whole life cycle experiment of milling Ti6Al4V alloy.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 446-463"},"PeriodicalIF":12.2,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142441581","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}

引用次数: 0

NextG manufacturing − New extreme manufacturing paradigm from the temporal perspective 下一代制造--从时间角度看新的极端制造模式

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-15 DOI: 10.1016/j.jmsy.2024.10.008

L. Hu , Y.B. Guo , I. Seskar , Y. Chen , N. Mandayam , W. “Grace” Guo , J. Yi

This paper proposes a new paradigm of extreme manufacturing from the temporal perspective in contrast to the current extreme manufacturing paradigm based on length scales (e.g., from nanometer to close-to-atom). The advent of 5 G and future 6 G (NextG) wireless communication provides unique capabilities of ultra-low end-to-end (E2E) latency (∼1 ms), high speed (up to 20 Gb/s), high reliability (>99.999 %), and high flexibility (wireless) to meet the stringent requirements of future manufacturing. The ultra-low E2E latency enables NextG Manufacturing - a new extreme manufacturing paradigm from the latency perspective. This positioning paper identifies the needs of NextG manufacturing, introduces the characteristics of NextG wireless communication networks, proposes a framework for NextG manufacturing, demonstrates use cases, summarizes current challenges, and provides an outlook for future research directions.

与当前基于长度尺度（如从纳米到接近原子）的极限制造范式相比，本文从时间角度提出了一种新的极限制造范式。5 G 和未来 6 G（NextG）无线通信的出现为满足未来制造的严格要求提供了超低端到端（E2E）延迟（∼1 ms）、高速（高达 20 Gb/s）、高可靠性（99.999 %）和高灵活性（无线）的独特能力。超低的 E2E 延迟实现了 NextG 制造--从延迟角度看，这是一种全新的极端制造模式。本定位论文明确了 NextG 制造业的需求，介绍了 NextG 无线通信网络的特点，提出了 NextG 制造业的框架，演示了使用案例，总结了当前面临的挑战，并展望了未来的研究方向。

引用次数: 0

A review and outlook of airframe digital twins for structural prognostics and health management in the aviation industry 机身数字双胞胎用于航空业结构预报和健康管理的回顾与展望

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-13 DOI: 10.1016/j.jmsy.2024.09.024

Joelle W.Y. Chia, Wim J.C. Verhagen, Jose M. Silva, Ivan S. Cole

The Airframe Digital Twin (ADT) framework was conceived over a decade ago as a revolutionary way to realise condition-based maintenance within the defence aviation field. Since then, this concept has witnessed significant progress not only in terms of its scope and areas of application, but also in the fidelity of the virtual models used to represent physical systems. This paper sheds light on the progress and evolution of the ADT framework and methodologies since 2011 through a systematic literature review. Based on this review, it is understood that the progress in ADT places the aerospace industry on a path towards achieving Structural Prognostics and Health Management (SPHM), nevertheless more work needs to be done. This paper proceeds on evaluating the remaining challenges in the development of the ADT for SPHM, particularly in the context of fatigue and corrosion as the main forms of structural degradation. Modelling of the environmental and operational conditions, multiphysics, and multiscale interactions are highlighted. A further review on the outlook for ADT in the civil aviation industry is presented through comparisons between current industrial regulations and the state-of-the-art in the scientific community, and focus areas for future works in developing the ADT for SPHM are identified.

机身数字孪生系统（ADT）框架是在十多年前提出的，它是在国防航空领域实现基于状态的维护的革命性方法。从那时起，这一概念不仅在应用范围和领域方面取得了重大进展，而且在用于表示物理系统的虚拟模型的保真度方面也取得了重大进展。本文通过系统的文献综述，揭示了 ADT 框架和方法自 2011 年以来的进展和演变。在此基础上，我们了解到，ADT 的进展使航空航天业走上了实现结构诊断和健康管理 (SPHM) 的道路，但仍有更多工作要做。本文着手评估在开发用于 SPHM 的 ADT 过程中仍然面临的挑战，特别是在疲劳和腐蚀作为结构退化的主要形式的背景下。重点介绍了环境和运行条件、多物理场和多尺度相互作用的建模。通过比较当前的工业法规和科学界的先进技术，进一步回顾了民用航空工业 ADT 的前景，并确定了未来开发 SPHM ADT 的重点领域。

{"title":"A review and outlook of airframe digital twins for structural prognostics and health management in the aviation industry","authors":"Joelle W.Y. Chia, Wim J.C. Verhagen, Jose M. Silva, Ivan S. Cole","doi":"10.1016/j.jmsy.2024.09.024","DOIUrl":"10.1016/j.jmsy.2024.09.024","url":null,"abstract":"<div><div>The Airframe Digital Twin (ADT) framework was conceived over a decade ago as a revolutionary way to realise condition-based maintenance within the defence aviation field. Since then, this concept has witnessed significant progress not only in terms of its scope and areas of application, but also in the fidelity of the virtual models used to represent physical systems. This paper sheds light on the progress and evolution of the ADT framework and methodologies since 2011 through a systematic literature review. Based on this review, it is understood that the progress in ADT places the aerospace industry on a path towards achieving Structural Prognostics and Health Management (SPHM), nevertheless more work needs to be done. This paper proceeds on evaluating the remaining challenges in the development of the ADT for SPHM, particularly in the context of fatigue and corrosion as the main forms of structural degradation. Modelling of the environmental and operational conditions, multiphysics, and multiscale interactions are highlighted. A further review on the outlook for ADT in the civil aviation industry is presented through comparisons between current industrial regulations and the state-of-the-art in the scientific community, and focus areas for future works in developing the ADT for SPHM are identified.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 398-417"},"PeriodicalIF":12.2,"publicationDate":"2024-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142433129","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrated system configuration and layout planning for flexible manufacturing systems 柔性制造系统的综合系统配置和布局规划

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-11 DOI: 10.1016/j.jmsy.2024.09.020

Péter Dobrovoczki , András Kovács , Hiroyuki Sakata , Daisuke Tsutsumi

During the (re-)design of manufacturing systems, geometrical limitations on the available floor space may seriously impact the applicable resource configurations, including the selection of machines, robots, as well as auxiliary equipment. In current practice, such cases are managed by arduous manual iterations over the selection of resources and their geometrical arrangement. To overcome this inefficiency of existing approaches, the paper introduces a generic, integrated configuration-and-layout problem where the configuration sub-problem can encode arbitrary application-specific constraints on the selection of items (e.g., CNC machines and robots), while the layout sub-problem ensures geometrical feasibility, via a 2D rectangle packing representation. The generic model is demonstrated on an industrial application that involves the design of a flexible manufacturing system: items corresponding to CNC machines and robots must be selected, assigned to multiple manufacturing cells, and placed in the workshop blueprint to ensure that a given mix of products can be manufactured in the desired volume. For solving the generic configuration-and-layout problem, a logic-based Benders decomposition method is proposed. The efficiency of the approach is ensured by adding lifted cuts, symmetry breaking, and redundant constraints inspired by 2D bin packing lower bounds to the core Benders framework. Thorough computational evaluation is performed on a large set of problem instances, whereas practical applicability is verified in a real industrial case study.

在制造系统的（重新）设计过程中，可用地面空间的几何限制可能会严重影响适用的资源配置，包括机器、机器人和辅助设备的选择。在当前的实践中，这种情况下需要对资源的选择及其几何排列进行艰苦的人工反复处理。为了克服现有方法的低效率问题，本文引入了一个通用的集成配置和布局问题，其中配置子问题可以对项目（如数控机床和机器人）选择的任意特定应用约束进行编码，而布局子问题则通过二维矩形包装表示法确保几何可行性。通用模型在一个涉及柔性制造系统设计的工业应用中进行了演示：必须选择与数控机床和机器人相对应的项目，将其分配给多个制造单元，并将其放置在车间蓝图中，以确保能以所需的数量制造出给定的产品组合。为了解决通用的配置和布局问题，我们提出了一种基于逻辑的本德斯分解法。在本德斯分解法的核心框架中增加了提升切割、对称性破坏和冗余约束，从而确保了该方法的效率。对大量问题实例进行了全面的计算评估，并在实际工业案例研究中验证了该方法的实用性。

{"title":"Integrated system configuration and layout planning for flexible manufacturing systems","authors":"Péter Dobrovoczki , András Kovács , Hiroyuki Sakata , Daisuke Tsutsumi","doi":"10.1016/j.jmsy.2024.09.020","DOIUrl":"10.1016/j.jmsy.2024.09.020","url":null,"abstract":"<div><div>During the (re-)design of manufacturing systems, geometrical limitations on the available floor space may seriously impact the applicable resource configurations, including the selection of machines, robots, as well as auxiliary equipment. In current practice, such cases are managed by arduous manual iterations over the selection of resources and their geometrical arrangement. To overcome this inefficiency of existing approaches, the paper introduces a generic, integrated configuration-and-layout problem where the configuration sub-problem can encode arbitrary application-specific constraints on the selection of items (e.g., CNC machines and robots), while the layout sub-problem ensures geometrical feasibility, via a 2D rectangle packing representation. The generic model is demonstrated on an industrial application that involves the design of a flexible manufacturing system: items corresponding to CNC machines and robots must be selected, assigned to multiple manufacturing cells, and placed in the workshop blueprint to ensure that a given mix of products can be manufactured in the desired volume. For solving the generic configuration-and-layout problem, a logic-based Benders decomposition method is proposed. The efficiency of the approach is ensured by adding lifted cuts, symmetry breaking, and redundant constraints inspired by 2D bin packing lower bounds to the core Benders framework. Thorough computational evaluation is performed on a large set of problem instances, whereas practical applicability is verified in a real industrial case study.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 384-397"},"PeriodicalIF":12.2,"publicationDate":"2024-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142425172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Maintenance plan adaptation based on health ratings of servitised machines through a fleet-wide machine clustering method 通过全机群机器聚类法，根据维修过的机器的健康评级调整维护计划

IF 12.2 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Journal of Manufacturing Systems

Pub Date : 2024-10-09 DOI: 10.1016/j.jmsy.2024.10.001

Alessandro Ruberti , Adalberto Polenghi , Marco Macchi

The increased requests for value-added services to integrate product performance push manufacturing companies to extend their service offerings to meet customers’ needs. In this context, maintenance planning can leverage new possibilities offered by digital technologies for data analytics services. The present research then proposes an approach for maintenance plan adaptation based on a data-driven method applied over a fleet of machines installed in different production sites. The method relies on collaborative prognostics to develop a clustering of machines’ behaviour aimed at providing the health ratings of the machines and the subsequent maintenance plan adaptation due to the deviation from the expected behaviour. The method is adopted from the perspective of an Original Equipment Manufacturer, as part of a transformation path towards an advanced provision of digitalization for maintenance service offerings. The method is validated in the context of two lines at selected customer’s premises. This demonstrates the viability and effectiveness of adapting the maintenance plans thanks to the data analytics in light of the current behaviour of the machines within the lines.

对整合产品性能的增值服务的需求不断增加，促使制造企业扩大服务范围以满足客户需求。在这种情况下，维护计划可以利用数字技术为数据分析服务提供的新可能性。因此，本研究提出了一种基于数据驱动方法的维护计划调整方法，该方法适用于安装在不同生产基地的机群。该方法依靠协作预报技术对机器的行为进行聚类，旨在提供机器的健康评级，并根据与预期行为的偏差对后续维护计划进行调整。该方法从原始设备制造商的角度出发，是向提供先进的数字化维护服务转型的一部分。该方法在选定客户的两条生产线上进行了验证。这证明了根据生产线上机器的当前行为，通过数据分析调整维护计划的可行性和有效性。

{"title":"Maintenance plan adaptation based on health ratings of servitised machines through a fleet-wide machine clustering method","authors":"Alessandro Ruberti , Adalberto Polenghi , Marco Macchi","doi":"10.1016/j.jmsy.2024.10.001","DOIUrl":"10.1016/j.jmsy.2024.10.001","url":null,"abstract":"<div><div>The increased requests for value-added services to integrate product performance push manufacturing companies to extend their service offerings to meet customers’ needs. In this context, maintenance planning can leverage new possibilities offered by digital technologies for data analytics services. The present research then proposes an approach for maintenance plan adaptation based on a data-driven method applied over a fleet of machines installed in different production sites. The method relies on collaborative prognostics to develop a clustering of machines’ behaviour aimed at providing the health ratings of the machines and the subsequent maintenance plan adaptation due to the deviation from the expected behaviour. The method is adopted from the perspective of an Original Equipment Manufacturer, as part of a transformation path towards an advanced provision of digitalization for maintenance service offerings. The method is validated in the context of two lines at selected customer’s premises. This demonstrates the viability and effectiveness of adapting the maintenance plans thanks to the data analytics in light of the current behaviour of the machines within the lines.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"77 ","pages":"Pages 368-383"},"PeriodicalIF":12.2,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142425171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0