Gülcan Aydin, Mehmet Tezcan, Bayram Ozgen, Tuğçe Nur Özkan
{"title":"Digital twin and predictive quality solution for insulated glass line","authors":"Gülcan Aydin, Mehmet Tezcan, Bayram Ozgen, Tuğçe Nur Özkan","doi":"10.1007/s10845-024-02426-y","DOIUrl":"https://doi.org/10.1007/s10845-024-02426-y","url":null,"abstract":"<p>This study is an integral part of an international research and development initiative investigating the application of digital twins and predictive quality solutions to enhance quality control and streamline production processes within the insulating glass manufacturing industry. The critical factor influencing the transformation of insulating glass into a high-quality, energy-efficient product is the gas filling rate. Therefore, this study focuses on the real-time monitoring and analysis of the gas filling process. Concurrently, predictive quality solutions are implemented to improve product quality and reduce defects. Consequently, it is evident that these technologies hold significant potential to advance the quality of insulating glass production and promote sustainable production practices on an international scale.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"1 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141166633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Spyros Theodoropoulos, Dimitrios Dardanis, Georgios Makridis, Patrik Zajec, Jože M. Rožanec, Dimosthenis Kyriazis, Panayiotis Tsanakas
{"title":"Enhancing robustness to novel visual defects through StyleGAN latent space navigation: a manufacturing use case","authors":"Spyros Theodoropoulos, Dimitrios Dardanis, Georgios Makridis, Patrik Zajec, Jože M. Rožanec, Dimosthenis Kyriazis, Panayiotis Tsanakas","doi":"10.1007/s10845-024-02415-1","DOIUrl":"https://doi.org/10.1007/s10845-024-02415-1","url":null,"abstract":"<p>Visual Quality Inspection is an integral part of the manufacturing process that is becoming increasingly automated with the advent of Industry 4.0. While very beneficial, AI-driven Computer Vision Algorithms and Deep Neural Networks face several issues that may impede their adoption in practical real-life settings such as a manufacturing shop floor. One such issue arising during an AI classifier’s continuous operation is the frequent lack of robustness to novel defects appearing for the first time. Such unanticipated inputs can pose a significant risk to cyber-physical applications as a resulting out-of-context decision could compromise the integrity of the production process. While recent Machine Learning methods can theoretically tackle this problem from different angles (e.g., open-set recognition, semi-supervised learning, intelligent data augmentation), applying them to a real-life setting with a small, imbalanced dataset and high inter-class similarity can be challenging. This paper confronts such a use case aiming at the automation of the visual quality inspection of shaver shell brand prints from the electronics industry and characterized by data scarcity and the existence of small local defects. To that end, we introduce a novel data augmentation approach based on the latent space manipulation of StyleGAN, where defect data is intentionally synthesized to simulate novel inputs that can help form a boundary of the model’s knowledge. Our approach shows promising results compared to well-established open-set recognition and semi-supervised methods applied to the same problem, while its consistent performance across classifier embeddings indicates lower coupling to the final classifier.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"33 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141166680","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shenghong Yan, Bo Chen, Caiwang Tan, Xiaoguo Song, Guodong Wang
{"title":"A data-driven time-sequence feature-based composite network of time-distributed CNN-LSTM for detecting pore defects in laser penetration welding","authors":"Shenghong Yan, Bo Chen, Caiwang Tan, Xiaoguo Song, Guodong Wang","doi":"10.1007/s10845-024-02391-6","DOIUrl":"https://doi.org/10.1007/s10845-024-02391-6","url":null,"abstract":"<p>The pore in laser penetration welding significantly deteriorates the mechanical property, and is an important criterion for evaluating the product quality. The intelligent diagnosis of welding can guide the optimization of process parameters to inhibit the pore formation. Considering that the signals in laser welding have time-sequence features and abundant implicitness information may cause high computational effort and information misidentify, an intelligent pore defect diagnosis method based on time–frequency feature extraction and a combined neural network of Convolutional Neural Networks (CNN) and Long short-term memory (LSTM) was proposed. Firstly, the visual signal results of vapor plume demonstrated that the pore formation was accompanied by irregular and continuous variation in vapor plume morphology during the subsequent period. Secondly, this denoising, decomposition, and restructuring of signals were performed by wavelet packet transform, and it was found that the sustaining fluctuation of frequency could localize the pore formation in the corresponding position of weld metal. Therefore, the signal was finely segmented to construct a cube time–frequency spectrogram data with the time-sequence characteristics. Finally, a combined classification model of CNN and LSTM was constructed for recognizing the temporal-spatial information of cube spectrogram data, realizing the online monitoring of pore defect. The results indicated that the proposed method was a promising solution for monitoring pore defect in laser penetration welding and improving product quality.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"48 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141150386","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mochamad Denny Surindra, Gusti Ahmad Fanshuri Alfarisy, Wahyu Caesarendra, Mohamad Iskandar Petra, Totok Prasetyo, Tegoeh Tjahjowidodo, Grzegorz M. Królczyk, Adam Glowacz, Munish Kumar Gupta
{"title":"Use of machine learning models in condition monitoring of abrasive belt in robotic arm grinding process","authors":"Mochamad Denny Surindra, Gusti Ahmad Fanshuri Alfarisy, Wahyu Caesarendra, Mohamad Iskandar Petra, Totok Prasetyo, Tegoeh Tjahjowidodo, Grzegorz M. Królczyk, Adam Glowacz, Munish Kumar Gupta","doi":"10.1007/s10845-024-02410-6","DOIUrl":"https://doi.org/10.1007/s10845-024-02410-6","url":null,"abstract":"<p>Although the aspects that affect the performance and the deterioration of abrasive belt grinding are known, wear prediction of abrasive belts in the robotic arm grinding process is still challenging. Massive wear of coarse grains on the belt surface has a serious impact on the integrity of the tool and it reduces the surface quality of the finished products. Conventional wear status monitoring strategies that use special tools result in the cessation of the manufacturing production process which sometimes takes a long time and is highly dependent on human capabilities. The erratic wear behavior of abrasive belts demands machining processes in the manufacturing industry to be equipped with intelligent decision-making methods. In this study, to maintain a uniform tool movement, an abrasive belt grinding is installed at the end-effector of a robotic arm to grind the surface of a mild steel workpiece. Simultaneously, accelerometers and force sensors are integrated into the system to record its vibration and forces in real-time. The vibration signal responses from the workpiece and the tool reflect the wear level of the grinding belt to monitor the tool’s condition. Intelligent monitoring of abrasive belt grinding conditions using several machine learning algorithms that include K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and Decision Tree (DT) are investigated. The machine learning models with the optimized hyperparameters that produce the highest average test accuracy were found using the DT, Random Forest (RF), and XGBoost. Meanwhile, the lowest latency was obtained by DT and RF. A decision-tree-based classifier could be a promising model to tackle the problem of abrasive belt grinding prediction. The application of various algorithms will be a major focus of our research team in future research activities, investigating how we apply the selected methods in real-world industrial environments.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"53 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141063206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Three-dimensional fabric smoothness evaluation using point cloud data for enhanced quality control","authors":"Zhijie Yuan, Binjie Xin, Jing Zhang, Yingqi Xu","doi":"10.1007/s10845-024-02367-6","DOIUrl":"https://doi.org/10.1007/s10845-024-02367-6","url":null,"abstract":"<p>Assessing the smoothness appearance of fabrics, especially in three-dimensional forms, is vital for quality control. Existing methods often lack objectivity or fail to consider the full 3D structure of the fabric. In this study, we introduce an innovative system that harnesses point cloud data to overcome these limitations. We use a 3D scanning system to capture a multi-directional point cloud representation of the textile surface. The data undergoes stitching and filtering to obtain an optimized point cloud model for feature extraction. We propose the 3D and 2D alpha-shape area ratio as a novel feature parameter for determining surface smoothness. Validation was conducted with 730 point clouds from 146 fabric samples, achieving an impressive 95.81%, recognition accuracy, which aligns with expert subjective evaluations. This research not only presents a dependable method for 3D textile smoothness grading but also indicates its applicability in other industries where surface evaluation is pivotal.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"38 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141063173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Multi-modal background-aware for defect semantic segmentation with limited data","authors":"Dexing Shan, Yunzhou Zhang, Shitong Liu","doi":"10.1007/s10845-024-02373-8","DOIUrl":"https://doi.org/10.1007/s10845-024-02373-8","url":null,"abstract":"<p>Visual defect detection is widely used in intelligent manufacturing to achieve intelligent detection of product quality. Two main challenges remain in industrial applications. One is the scarcity of defect samples and the other is the weak texture variation of industrial defects. The above problems lead to the application of RGB image-based industrial defect segmentation. To this end, we propose a multi-modal background-aware network (MMBA-Net) for few-shot defect (2D+3D) segmentation with limited data, which can segment texture and structural defects in unseen and seen domains (objects). To synthesize the perception capabilities of different imaging conditions, MMBA-Net exploits the point cloud to provide spatial information for the RGB images. Furthermore, we found that background regions are perceptually consistent within an industrial image, which can be leveraged to discriminate between foreground and background regions. To implement this idea, we model correlation learning between multi-modal query samples and multi-modal normal (defect-free) samples as an optimal transport problem, establishing robust multi-modal background correlations between query and normal samples across different modalities. Experiments were conducted on real-world industrial products and food datasets, demonstrating that the proposed method can perform effective base learning and meta-learning on a small number of defective samples (approximately 15–25 defective training samples) to achieve effective segmentation of defects in the seen and unseen domains.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"20 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141063323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Defining a feature-level digital twin process model by extracting machining features from MBD models for intelligent process planning","authors":"Jingjing Li, Guanghui Zhou, Chao Zhang, Junsheng Hu, Fengtian Chang, Andrea Matta","doi":"10.1007/s10845-024-02406-2","DOIUrl":"https://doi.org/10.1007/s10845-024-02406-2","url":null,"abstract":"<p>The booming development of emerging technologies and their integration in process planning provide new opportunities for solving the problems in traditional trial-and-error process planning. Combining digital twin with 3D computer vision, this paper defines a novel feature-level digital twin process model (FL-DTPM) by extracting machining features from model-based definition models. Firstly, a multi-dimensional FL-DTPM framework is defined by fusing on-site data, quality information, and process knowledge, where the synergistic mechanism of its virtual and physical processes is revealed. Then, 3D computer vision-enabled machining features extraction method is embedded into the FL-DTPM framework to support the reuse of process knowledge, which involves the procedures of data pre-processing, semantic segmentation, and instance segmentation. Finally, the effectiveness of the proposed features extraction method is verified and the application of FL-DTPM in machining process is presented. Oriented to the impeller process planning, a prototype of FL-DTPM is constructed to explore the potential application scenarios of the proposed method in intelligent process planning, which could provide insights into the industrial implementation of FL-DTPM for aerospace manufacturing enterprises.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"94 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140928619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 method based on deep learning algorithms for material deformation rate detection","authors":"Selim Özdem, İlhami Muharrem Orak","doi":"10.1007/s10845-024-02409-z","DOIUrl":"https://doi.org/10.1007/s10845-024-02409-z","url":null,"abstract":"<p>Given the significant influence of microstructural characteristics on a material’s mechanical, physical, and chemical properties, this study posits that the deformation rate of structural steel S235-JR can be precisely determined by analyzing changes in its microstructure. Utilizing advanced artificial intelligence techniques, microstructure images of S235-JR were systematically analyzed to establish a correlation with the material’s lifespan. The steel was categorized into five classes and subjected to varying deformation rates through laboratory tensile tests. Post-deformation, the specimens underwent metallographic procedures to obtain microstructure images via an light optical microscope (LOM). A dataset comprising 10000 images was introduced and validated using K-Fold cross-validation. This research utilized deep learning (DL) architectures ResNet50, ResNet101, ResNet152, VGG16, and VGG19 through transfer learning to train and classify images containing deformation information. The effectiveness of these models was meticulously compared using a suite of metrics including Accuracy, F1-score, Recall, and Precision to determine their classification success. The classification accuracy was compared across the test data, with ResNet50 achieving the highest accuracy of 98.45%. This study contributes a five-class dataset of labeled images to the literature, offering a new resource for future research in material science and engineering.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"22 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140928477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"MSOA: A modular service-oriented architecture to integrate mobile manipulators as cyber-physical systems","authors":"Nooshin Ghodsian, Khaled Benfriha, Adel Olabi, Varun Gopinath, Esma Talhi, Lucas Hof, Aurélien Arnou","doi":"10.1007/s10845-024-02404-4","DOIUrl":"https://doi.org/10.1007/s10845-024-02404-4","url":null,"abstract":"<p>In the evolving landscape of the fourth industrial revolution, the integration of cyber-physical systems (CPSs) into industrial manufacturing, particularly focusing on autonomous mobile manipulators (MMs), is examined. A comprehensive framework is proposed for embedding MMs into existing production systems, addressing the burgeoning need for flexibility and adaptability in contemporary manufacturing. At the heart of this framework is the development of a modular service-oriented architecture, characterized by adaptive decentralization. This approach prioritizes real-time interoperability and leverages virtual capabilities, which is crucial for the effective integration of MMs as CPSs. The framework is designed to not only accommodate the operational complexities of MMs but also ensure their seamless alignment with existing production control systems. The practical application of this framework is demonstrated at the Platform 4.0 research production line at Arts et Métiers. An MM named MoMa, developed by OMRON Company, was integrated into the system. This application highlighted the framework’s capacity to significantly enhance the production system's flexibility, autonomy, and efficiency. Managed by the manufacturing execution system (MES), the successful integration of MoMa exemplifies the framework's potential to transform manufacturing processes in alignment with the principles of Industry 4.0.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"123 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140928840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yifan Tang, Mostafa Rahmani Dehaghani, Pouyan Sajadi, G. Gary Wang
{"title":"Selecting subsets of source data for transfer learning with applications in metal additive manufacturing","authors":"Yifan Tang, Mostafa Rahmani Dehaghani, Pouyan Sajadi, G. Gary Wang","doi":"10.1007/s10845-024-02402-6","DOIUrl":"https://doi.org/10.1007/s10845-024-02402-6","url":null,"abstract":"<p>Considering data insufficiency in metal additive manufacturing (AM), transfer learning (TL) has been adopted to extract knowledge from source domains (e.g., completed printings) to improve the modeling performance in target domains (e.g., new printings). Current applications use all accessible source data directly in TL with no regard to the similarity between source and target data. This paper proposes a systematic method to find appropriate subsets of source data based on similarities between the source and limited target datasets. Such similarity is characterized by the spatial and model distance metrics. A Pareto frontier-based source data selection method is developed, where the source data located on the Pareto frontier defined by two similarity distance metrics are selected iteratively. This method is integrated into an instance-based TL method (decision tree regression model) and a model-based TL method (fine-tuned artificial neural network). Both models are then tested on several regression tasks in metal AM. Comparison results demonstrate that (1) the source data selection method is general and supports integration with various TL methods and distance metrics, (2) compared with using all source data, the proposed method can find a subset of source data from the same domain with better TL performance in metal AM regression tasks involving different processes and machines, and (3) when multiple source domains exist, the source data selection method could find the subset from one source domain to obtain comparable or better TL performance than the model constructed using data from all source domains.</p>","PeriodicalId":16193,"journal":{"name":"Journal of Intelligent Manufacturing","volume":"3 1","pages":""},"PeriodicalIF":8.3,"publicationDate":"2024-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140928604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}