Journal of Manufacturing Systems最新文献

{"title":"Selection of manufacturing processes using graph neural networks","authors":"Marco Hussong ,&nbsp;Patrick Ruediger-Flore ,&nbsp;Matthias Klar ,&nbsp;Marius Kloft ,&nbsp;Jan C. Aurich","doi":"10.1016/j.jmsy.2025.02.016","DOIUrl":"10.1016/j.jmsy.2025.02.016","url":null,"abstract":"<div><div>The increasing complexity of modern manufacturing, driven by trends such as product customization and shorter product life cycles, presents significant challenges in process planning. Traditional methods for selecting manufacturing processes in industry rely on expert knowledge and manual intervention, which can be time-consuming and error-prone. Systems that can automate the selection of manufacturing processes become increasingly important. Current approaches for the selection of manufacturing processes focus on deep learning that convert the 3D CAD models to intermediate representations such as voxels, point clouds or dexels. However, this transformation can result in the loss of topological, geometrical, or Product and Manufacturing Information (PMI). To address these challenges, this paper proposes a neural network architecture MaProNet. MaProNet is a graph attention neural network (GAT) designed to capture topological and geometrical information through the analysis of Attributed Adjacency Graphs (AAG) and Mesh structures. MaProNet also incorporates a wide range of PMI information.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 176-193"},"PeriodicalIF":12.2,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563069","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}

{"title":"A design framework for high-fidelity human-centric digital twin of collaborative work cell in Industry 5.0","authors":"Tianyu Wang ,&nbsp;Zhihao Liu ,&nbsp;Lihui Wang ,&nbsp;Mian Li ,&nbsp;Xi Vincent Wang","doi":"10.1016/j.jmsy.2025.02.018","DOIUrl":"10.1016/j.jmsy.2025.02.018","url":null,"abstract":"<div><div>Digital Twin (DT) of a manufacturing system mainly involving materials and machines has been widely explored in the past decades to facilitate the mass customization of modern products. Recently, the new vision of Industry 5.0 has brought human operators back to the core part of work cells. To this end, designing human-centric DT systems is vital for an ergonomic and symbiotic working environment. However, one major challenge is the construction and utilization of high-fidelity digital human models. In the literature, preset universal human avatar models such as skeletons are mostly employed to represent the human operators, which overlooks the individual differences of physical traits. Besides, the fundamental utilization features such as motion tracking and procedure recognition still do not well address the practical issues such as occlusions and incomplete observations. To deal with the challenge, this paper proposes a systematic design framework to quickly and precisely build and utilize the human-centric DT systems. The mesh-based customized human operator models with rendered appearances are first generated within one minute from a short motion video. Then transformer-based deep learning networks are developed to realize the motion-related operator status synchronization in complex conditions. Extensive experiments on multiple real-world human–robot collaborative work cells show the superior performance of the proposed framework over the state-of-the-art.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 140-156"},"PeriodicalIF":12.2,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tool digital twin based on knowledge embedding for precision CNC machine tools: Wear prediction for collaborative multi-tool","authors":"Jiacheng Sun ,&nbsp;Dong Wang ,&nbsp;Zhenyu Liu ,&nbsp;Chan Qiu ,&nbsp;Hui Liu ,&nbsp;Guodong Sa ,&nbsp;Jianrong Tan","doi":"10.1016/j.jmsy.2025.02.021","DOIUrl":"10.1016/j.jmsy.2025.02.021","url":null,"abstract":"<div><div>Tool wear prediction is vital for enhancing machining accuracy and ensuring production safety. However, challenges arise from non-processing data interference and missing tool wear samples, complicating the construction of accurate prediction models. Additionally, the complexity of collaborative multi-tool operations on precision computer numerical control (CNC) machine tools, where varying tool types and complex working conditions exist, further exacerbates the difficulty of achieving precise wear prediction. To address these challenges, this paper introduces a digital twin architecture for tool wear prediction, based on knowledge embedding. The proposed architecture is designed to predict the wear of multiple tools, incorporating modules for processing data screening, missing value completion, wear state classification, and so on. On the basis of obtaining high-quality sensing data and complete tool wear values, the wear state and machining process knowledge are embedded into the prediction process. A tool wear prediction model is then constructed based on a Kolmogorov-Arnold integrated time convolutional network (KA-TCN), so as to achieve accurate prediction of multi-tool wear. The effectiveness of the method is validated using data from two grinding wheel wear test platforms and two milling datasets, PHM2010 and NASA. Experimental results demonstrate that the knowledge embedded KA-TCN model outperforms existing approaches, improving prediction accuracy by over 22.4 % on the milling dataset, and by 76.4 % in grinding wheel wear prediction compared to classical methods.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 157-175"},"PeriodicalIF":12.2,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel intelligent assembly-adjustment method for aero-engine rotor system aimed at improving interface load-bearing performance","authors":"Binbin Zhao ,&nbsp;Xiaokai Mu ,&nbsp;Hao Li ,&nbsp;Wenliao Du ,&nbsp;Qingchao Sun","doi":"10.1016/j.jmsy.2025.03.001","DOIUrl":"10.1016/j.jmsy.2025.03.001","url":null,"abstract":"<div><div>For aero-engine rotor system, the interface load-bearing performance is crucial for its connection stability. In order to effectively ensure the connection performance of aero-engine in strong service environments, this paper proposed an intelligent assembly-adjustment method that integrates key feature measurement, performance prediction and process feedback. Taking the typical multi-bolts connection structure in aero-engine rotor system as the research object, firstly, by using the intelligent fastener, the precise acquisition of preload distribution after assembly process has been achieved. Then, for the issue of interface load-bearing performance weakening represented by non-uniform slip behavior, a <span><math><mi>CON</mi></math></span>-<span><math><mrow><mo>∆</mo><mi>e</mi></mrow></math></span> agent model was constructed for predicting the non-uniformity of interface slip. Finally, based on <span><math><mi>CON</mi></math></span> reduction principle, a local preload feedback-adjustment method for improving the interface load-bearing performance has been proposed. Through a simulated rotor assembly case, the effectiveness of the proposed intelligent assembly-adjustment method in improving interface load-bearing performance was demonstrated. Overall, the research content has a better application prospect in connection stability improvement of mechanical system.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 126-139"},"PeriodicalIF":12.2,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MetaFactory: A cloud-based framework to configure and generate dynamic data structures from the STEP-NC knowledge graph","authors":"Wenlei Xiao ,&nbsp;Tianze Qiu ,&nbsp;Jiurong Guo ,&nbsp;Gang Zhao","doi":"10.1016/j.jmsy.2025.02.012","DOIUrl":"10.1016/j.jmsy.2025.02.012","url":null,"abstract":"<div><div>In our previous studies, twin-oriented manufacturing has been identified as a crucial solution to address the manufacturing crisis. Within this context, the notion of “digital twin as a service” necessitates that various twin services share and communicate with each other in a standardized manner. STEP-NC offers a potentially unified model to facilitate data exchange, providing object-oriented and standardized data models for a comprehensive representation of manufacturing resources in the digital realm. However, the complexity of STEP-NC renders it too cumbersome for implementation in diverse cloud-based services or PC-based software. This complexity is a fundamental reason why STEP-NC has struggled to find application in commercial CNC systems despite years of research. To overcome this technical challenge, this paper introduces a novel concept termed “dynamic STEP-NC data structure”, inspired by the dynamic language philosophy of dynamic programming language (such as Python). This approach allows different services and software packages to maintain their own data definitions while still aligning with the original STEP-NC definition. We have developed a framework called MetaFactory that supports the configuration of streamlined data structures and generates the corresponding program code required by various service developers. On this basis, we implemented automatic modeling for a STEP-NC object-oriented database. Using the data trimming and dimensionality reduction methods provided by MetaFactory, several prototype systems for different application scenarios have been developed.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 89-107"},"PeriodicalIF":12.2,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143549099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An efficient online geometric simulation algorithm for real-time CNC machining process based on look-ahead method","authors":"Tianze Qiu ,&nbsp;Bofang Dai ,&nbsp;Wenlei Xiao ,&nbsp;Chen Zhao ,&nbsp;Gang Zhao","doi":"10.1016/j.jmsy.2025.02.013","DOIUrl":"10.1016/j.jmsy.2025.02.013","url":null,"abstract":"<div><div>Intelligent CNC machining requires advanced online geometric simulation to improve transparency and optimize machining processes. The simulation algorithms need to be efficient enough to keep up with machine tool motions. However, traditional algorithms, which typically discretize the entire blank initially, often result in redundant computations, hindering efficiency in online environments. To strike a balance between efficiency and accuracy, this paper presents an efficient online simulation algorithm with three key innovations. First, the algorithm incorporates the concept of look-ahead into geometric simulation to pinpoint workpiece areas likely to contact the cutting tool. Second, it employs a dynamic voxel partitioning mechanism that adapts to the cutting tool’s movement, reducing data structures and eliminating redundant computations. Third, a hybrid modeling approach integrates voxel model spatial indexing with Tri-dexel model Boolean operations, enabling rapid local positioning and efficient micro-structural representation of the workpiece. Additionally, the algorithm is further optimized in key stages such as online interpolation and surface reconstruction. This algorithm has been integrated into several online simulation software systems and tested and validated on multiple typical 3/5-axis workpieces. Actual machining experiments confirm its efficiency, with over 99% of simulation computation times below 10 ms, meeting the requirements for online environments. The algorithm also demonstrates excellent performance in simulating large-scale aerospace workpieces, providing a solid foundation for real-time synchronization of geometric and physical parameters.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 108-125"},"PeriodicalIF":12.2,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143549100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital twin of dynamics for parallel kinematic machine with distributed force/position interaction","authors":"Fangyan Zheng ,&nbsp;Xinghui Han ,&nbsp;Lin Hua ,&nbsp;Wenjun Xu","doi":"10.1016/j.jmsy.2025.02.019","DOIUrl":"10.1016/j.jmsy.2025.02.019","url":null,"abstract":"<div><div>Extensibility is significant for digital twin (DT) manufacturing systems. However, existing DT models mainly focus on a specific task in manufacturing. The main challenge lies in the specific physical model when addressing different tasks. In fact, the dynamics of machines are the physical basis for most applications, e.g., motion planning, production scheduling, process monitoring, machine maintenance, and so on. Therefore, the Digital Twin of dynamics (DTOD) for machines will be a foundation for a highly integrated and extensible DT system. However, due to the challenges in real-time dynamic modeling and the corresponding data interaction methods, the DTOD for parallel kinematic machines (PKM) has not been realized.</div><div>Facing this challenge, this paper develops a DTOD for PKM with distributed force/position interaction. Firstly, a simplified rigid-flexible coupling dynamic model of PKM, considering link deformations, is established for real-time calculation. Then, a distributed position/force interaction method based on Kalman filter-based data fusion is proposed to realize high-performance data interaction between cyber and physical space. On this basis, a five-dimension digital twin model for DTOD of PKM is established. Further, the DTOD system with an architecture comprising dual central processors and multiple distributed edge executors/sensors is developed and validated by aircraft gear manufacturing, showing 80 % prediction accuracy of dynamic error. Finally, to show the extensibility, integrated error correction for aircraft gear manufacturing is proposed as an extended application of the DTOD system. The gear error is reduced to 218 μm (with error correction) from 503 μm, representing a reduction of about 57 %, showing the high performance of the developed DTOD system and its high application potential.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 70-88"},"PeriodicalIF":12.2,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143549172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Energy-efficient human-robot collaborative U-shaped disassembly line balancing problem considering turn on-off strategy: Uncertain modeling and solution method","authors":"Zhongwei Huang ,&nbsp;Honghao Zhang ,&nbsp;Guangdong Tian ,&nbsp;Mingzhi Yang ,&nbsp;Danqi Wang ,&nbsp;Zhiwu Li","doi":"10.1016/j.jmsy.2025.02.004","DOIUrl":"10.1016/j.jmsy.2025.02.004","url":null,"abstract":"<div><div>The quantity of waste automobile is becoming very large. Waste automobile not only occupies resources, but also easily pollutes the environment. How to realize the efficient and green treatment of recycled automobile is a hot topic in the industrial circular economy today. The disassembly line is the most efficient way to address large-scale waste automobile. Therefore, this paper takes the disassembly experiment of recycled automobile engine as the information orientation to construct energy-efficient human-robot collaborative U-shaped disassembly line balancing (HRU-DLB) framework considering turn on-off strategy. An engine disassembly information modeling method is proposed to address the issue on the actual disassembly space limitation. Establish a based-normal cloud HRU-DLBP mathematical model including disassembly smoothness, disassembly energy consumption (DEC), disassembly cost, disassembly idle time and disassembly carbon emission (DCE). To further reduce the disassembly energy consumption and carbon emission, the well-accepted energy-saving measure, known as the turn on-off strategy, is also integrated. Subsequently, a hybrid multi-objective optimization algorithm called ALNS-NSGA II, which combines the NSGA-II algorithm and adaptive large-scale neighborhood search algorithm is developed to explore the optimal Pareto solution set. Finally, the novel behavioral decision model is proposed to select the optimal HRU-DLB scheme. The comparative analysis shows that the turn on-off strategy can reduce DEC by 26 % and DCE by 3.1 % in a cycle time, respectively. The computational results confirm the feasibility and effectiveness of the proposed ALNS-NSGA II in solving the HRU-DLBP. The comparative analysis and sensitivity analysis demonstrate that the proposed behavioral decision model has better ranking and classification effects.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 38-69"},"PeriodicalIF":12.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143509815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An adaptive genetic algorithm based on Q-learning for energy-efficient e-waste disassembly line balancing and rebalancing considering task failures","authors":"Kaipu Wang ,&nbsp;Xiaoyi Ma ,&nbsp;Yibing Li ,&nbsp;Yabo Luo ,&nbsp;Yingli Li ,&nbsp;Liang Gao","doi":"10.1016/j.jmsy.2025.02.009","DOIUrl":"10.1016/j.jmsy.2025.02.009","url":null,"abstract":"<div><div>The efficient disassembly and recycling of e-waste not only provides economic benefits but also contributes to reducing energy consumption. However, the disassembly process is often influenced by uncertainties, such as damage or deformation of components, which may result in potential task failures. These failures can disrupt the balance of the disassembly line, affecting the efficiency of subsequent tasks. Therefore, it is crucial to develop a decision-making model and optimization method to address disassembly failures. This study presents a predictive disassembly line balancing model with objectives focused on the number of workstations, the smoothness index, and energy consumption. The optimization objective of adjusting the disassembly sequence is introduced, and a rebalancing model is developed to reallocate the remaining tasks in response to various failures. The sequence combination that minimizes comprehensive energy consumption is selected as the optimal disassembly strategy. Considering the complexity and dynamic disturbance of the problem, an adaptive multi-objective genetic algorithm based on Q-learning is proposed. To improve the quality of the disassembly solutions, six evolutionary actions and four population performance states are designed. During the algorithm’s iteration, the search strategy is dynamically adjusted through Q-learning. The effectiveness of the proposed algorithm is verified by solving several classic disassembly cases and comparing the results with those from six advanced algorithms. Finally, in an actual refrigerator disassembly case, 11 disassembly schemes are generated, accounting for task failures. The results indicate that, compared to traditional disassembly methods, the rebalancing approach not only optimizes the station loads but also increases revenue by 11.98 %, demonstrating the effectiveness of the proposed model and method in handling task failures on disassembly lines.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 1-19"},"PeriodicalIF":12.2,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143487380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rescheduling human-robot collaboration tasks under dynamic disassembly scenarios: An MLLM-KG collaboratively enabled approach","authors":"Weigang Yu,&nbsp;Jianhao Lv,&nbsp;Weibin Zhuang,&nbsp;Xinyu Pan,&nbsp;Sijie Wen,&nbsp;Jinsong Bao,&nbsp;Xinyu Li","doi":"10.1016/j.jmsy.2025.02.015","DOIUrl":"10.1016/j.jmsy.2025.02.015","url":null,"abstract":"<div><div>During product recycling, the uncertainty of the degradation level of end-of-life products leads to dynamic conditions such as component corrosion and damage during the disassembly process. Therefore, enhancing the robot's perception of disassembly scenarios and matching historical disassembly experiences is crucial for task rescheduling in human-robot collaborative disassembly (HRCD) under dynamic conditions. To address this, this paper proposes a dynamic task rescheduling method for human-robot collaborative disassembly, empowered by the synergy of Knowledge Graph (KG) and Multimodal Large Language Model (MLLM). Leveraging a Mark-Aware image preprocessing module and prompt-based scene understanding, the physical characteristics and occlusion relationships of disassembly targets are extracted. The concept of affordance is introduced, and an Affordance KG is constructed to recommend disassembly actions based on the physical features of objects in the scene. A task allocation standard for human-robot collaboration is designed, which, combined with depth and human factor information from mixed reality scenarios, enables dynamic task rescheduling and reconstruction of the entire human-robot collaborative disassembly process. The proposed method is validated through a case study on human-robot collaborative disassembly of end-of-life automotive lithium-ion batteries. Experimental results demonstrate that the method exhibits strong robustness and generalizability in dynamic disassembly scenarios, accurately identifying the physical features of components and recommending appropriate disassembly actions under conditions such as component corrosion, damage, and tool unavailability, thus achieving effective task rescheduling.</div></div>","PeriodicalId":16227,"journal":{"name":"Journal of Manufacturing Systems","volume":"80 ","pages":"Pages 20-37"},"PeriodicalIF":12.2,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143487681","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}