Procedia CIRP最新文献_第7页

From Concept to Tool: Formalization and Computational Advances in Quantifier-Based Tolerance Analysis 从概念到工具：基于量词的公差分析的形式化和计算进展

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-08-05 DOI: 10.1016/j.procir.2026.03.138

Jean-Yves Dantan, Alex Ballu, Jawad Qureshi, Antoine Dumas, Edoh Goka, Lazhar Homri

引用次数: 0

Fast Industrial Computed Tomography for Remanufacturing: Evaluating Measurement Time savings with 3D UNETR Sparse-view Segmentation 用于再制造的快速工业计算机断层扫描：评估使用3D UNETR稀疏视图分割节省的测量时间

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-08-05 DOI: 10.1016/j.procir.2026.03.121

Edwin Blum, Tom Hild, Dominik Koch, Martin Benfer, Gisela Lanza

{"title":"Fast Industrial Computed Tomography for Remanufacturing: Evaluating Measurement Time savings with 3D UNETR Sparse-view Segmentation","authors":"Edwin Blum, Tom Hild, Dominik Koch, Martin Benfer, Gisela Lanza","doi":"10.1016/j.procir.2026.03.121","DOIUrl":"10.1016/j.procir.2026.03.121","url":null,"abstract":"<div><div>Industrial computed tomography (CT) provides non-destructive insight into internal structures and full 3D geometry of complex assemblies, but its use in production is limited by long acquisition times. Sparse-view CT can reduce measurement time, yet introduces artefacts and geometric bias that may compromise metrology. We evaluate sparse-view, machine-learning-based component extraction on an assembled angle grinder containing a bevel gear, using five gear variants with wear and damage. A 3D UNETR model is trained for signed distance field (SDF) regression to enable robust surface extraction. Across five leave-one-gear-out runs, we assess segmentation robustness under reduced acquisition time and translate the results into metrology-relevant outcomes via surface deviations and a tolerance-oriented tooth-profile evaluation. Results show that acquisition time can be substantially reduced within the trained regime while maintaining high segmentation quality and largely stable feature-level deviations. For stronger reductions, errors increase non-linearly and become spatially localized, dominating tail statistics and tolerance failures. Beyond demonstrating time savings, the proposed tolerance-centered evaluation links sparse-view segmentation performance to feature-level deviation distributions and pass rates, providing a transferable basis for selecting acquisition-time settings and motivating adaptive inspection strategies that trade measurement time against decision reliability.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"145 ","pages":"Pages 171-176"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148694517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting Sheet-Metal Part Form Deviation Using Discrete Inspection Points 用离散检测点预测钣金件形状偏差

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-08-05 DOI: 10.1016/j.procir.2026.03.127

David Renborg, Peter Edholm, Maria Andrén, Kristina Wärmefjord, Lars Lindkvist, Rikard Söderberg

{"title":"Predicting Sheet-Metal Part Form Deviation Using Discrete Inspection Points","authors":"David Renborg, Peter Edholm, Maria Andrén, Kristina Wärmefjord, Lars Lindkvist, Rikard Söderberg","doi":"10.1016/j.procir.2026.03.127","DOIUrl":"10.1016/j.procir.2026.03.127","url":null,"abstract":"<div><div>In high-volume automotive production there is a need to describe the geometry of a part more accurately in early product-realization phases, when comprehensive measurement data are not yet available at scale, so that large-scale statistical variation analyses can be run quickly with simplified yet credible models. This paper proposes a method to predict part shape from sparse information by optimizing both the number and placement of inspection points subject to a defined accuracy threshold. Two production body-in-white parts, a compact inner d-pillar and a large outer fender, were 3D scanned with a Nikon H120 blue-laser high-precision scanner and compared against finite-element (FE) simulation models. The surface is reconstructed by combining the mesh and stiffness matrix of an FE-based material model implemented in a computer-aided tolerancing (CAT) software, RD&T, and the method is commissioned/validated against the scan. The study is experimental: a high-density scan is used as reference to evaluate how accurately sparse point constraints can reconstruct the full-field deformation. An error-driven greedy algorithm is used to iteratively select additional inspection points at locations where the current reconstruction deviates most from the scanned geometry, until the accuracy criterion is satisfied while minimizing model size and computation time.</div><div>Results indicate a correlation between the 3D-scanned geometry and the simulation-based reconstructions, demonstrating that the approach can reproduce part shape with reasonable precision for early-phase analyses. The method enables rapid simulations on large populations of parts with a reduced model that remains accurate enough to support decision-making in tolerancing and variation analysis. The reduced model is compatible with standard non-rigid Monte Carlo variation simulations in RD&T, enabling fast early-phase tolerance studies on large virtual populations of parts.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"145 ","pages":"Pages 129-134"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148694577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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 study on the generation of skin model shapes considering population characteristics of flatness 考虑平整度群体特征的皮肤模型形状生成研究

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-08-05 DOI: 10.1016/j.procir.2026.03.124

Akimasa Otsuka, Naruse Kougo, Shinya Suzuki, Fusaomi Nagata

{"title":"A study on the generation of skin model shapes considering population characteristics of flatness","authors":"Akimasa Otsuka, Naruse Kougo, Shinya Suzuki, Fusaomi Nagata","doi":"10.1016/j.procir.2026.03.124","DOIUrl":"10.1016/j.procir.2026.03.124","url":null,"abstract":"<div><div>In the production process of mechanical parts, machining errors are inevitable, and tolerances should be designed in advance considering those errors, measurement methods and product performance. To predict product performance in advance, suitable random variable models, especially skin model shapes in tolerance design, should be prepared. In our previous research, we focused on generating each skin model shape more realistic. For practical use, a population of skin model shapes should be closer to a realistic population. In this study, we focus on distribution type as population characteristics. Linear and angular dimensions and geometrical deviations are variables, and it is usually assumed that each follows a normal distribution or sometimes log-normal distribution. However, several geometric tolerances tend to be non-normal due to their mathematical definitions. For example, flatness of a surface is difference between the maximum and the minimum values of measured points of the surface after several filtering processes. Statistically, when the <em>z</em>-values of measured points follow a normal distribution, the distributions of the maximum and minimum values follow the generalized extreme value distribution (GEV). Distribution of difference between those two distributions cannot be expressed as a closed form, but it can be expressed as the convolution of the two GEV distribution. Therefore, a distribution of flatness can be analyzed numerically. Flatness distributions of the actual surface and the skin model shapes are analyzed using statistical tools on MATLAB. The results are reflected in parameters in our generation of skin model shapes, and the distribution of skin model shapes is confirmed to be improved.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"145 ","pages":"Pages 153-157"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148694581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Comparison of computational methods for establishing datum reference frames 建立基准参考系的计算方法比较

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-09-02 DOI: 10.1016/j.procir.2026.08.002

Jaime Berez

{"title":"Comparison of computational methods for establishing datum reference frames","authors":"Jaime Berez","doi":"10.1016/j.procir.2026.08.002","DOIUrl":"10.1016/j.procir.2026.08.002","url":null,"abstract":"<div><div>Datum reference frames (DRFs) are localized coordinate systems defined by geometrical tolerances used to specify position, orientation, or some other relative geometric characteristic. A variety of computational means for realizing DRFs from coordinate data have been considered and studied over the years, with relatively novel methods being presented in the peer-reviewed literature as recently as in the last 5-10 years. Standardization organizations that control normative practices of geometric dimensioning and tolerancing (GD&T) and geometric product specification (GPS), have even more recently started to incorporate these novel methods. While this represents progress towards unified and reasonable approaches for establishing DRFs, there likely remains significant variety in the computational means used industry by the actual commercial instruments and software. Altogether, this means that a divergence of computational methods from the design intent as well as fundamental differences between the available algorithms can result in potentially non-negligable disagreement between measurements. The study presented herein applies a variety of computational coordinate metrology methods to verify a simple geometric specification that serves as a case study - a DRF based on three nominally orthogonal planar datum features. The individual datums that make up the DRF are fitted using three prominent approaches: unconstrained L2, constrained L2, and constrained L1 norms. The effect of on as-measured feature position is emphasized, highlighting how computational metrology techniques can affect measurement results on the order of datum feature form error.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"145 ","pages":"Pages 320-325"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148936349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evaluating feedback mechanisms in the automatic creation of assembly instructions using retrieval-augmented generation 评估使用检索增强生成自动创建汇编指令中的反馈机制

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-09-03 DOI: 10.1016/j.procir.2026.03.242

Julia Becker, Benjamin Adrian, Dominik Green, Sven Hinrichsen

{"title":"Evaluating feedback mechanisms in the automatic creation of assembly instructions using retrieval-augmented generation","authors":"Julia Becker, Benjamin Adrian, Dominik Green, Sven Hinrichsen","doi":"10.1016/j.procir.2026.03.242","DOIUrl":"10.1016/j.procir.2026.03.242","url":null,"abstract":"<div><div>In recent years, Retrieval Augmented Generation (RAG) has emerged as a promising technology for designing application-specific chatbots. RAG combines a Large Language Model (LLM) with application-specific information provided via a database. This opens up the possibility for companies to at least partially automate existing business processes using RAG. One use case relates to work preparation at a mechanical engineering company. Due to the high complexity of the products and the wide variety of variants, creating assembly instructions for individual assemblies and modules involves a considerable amount of administrative work. The use of RAG can help to significantly improve the efficiency of this process. However, a key challenge is to ensure high quality of automatically generated assembly instructions. While existing research has focused on feedback mechanisms for improving LLMs, the effectiveness of different prompt-based feedback types on RAG-generated technical documentation quality has not yet been sufficiently investigated. This article therefore explores how quality of assembly instructions can be improved by integrating feedback mechanisms into a RAG system. For this purpose, two different feedback mechanisms were used in a laboratory experiment. On the one hand, an exclusively quantitative feedback mechanism was used. On the other hand, quantitative and qualitative feedback were combined. The results demonstrate that feedback mechanisms contribute to improving quality of information in assembly instructions. In addition, it was found that exclusively quantitative feedback leads to similarly good results as more complex combined feedback.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"146 ","pages":"Pages 243-248"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148939052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Implementation challenges of artificial intelligence in the circular economy 人工智能在循环经济中的实施挑战

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-09-03 DOI: 10.1016/j.procir.2026.03.261

Julia Kunert, Alexander Wuttke, Hendrik van der Valk

{"title":"Implementation challenges of artificial intelligence in the circular economy","authors":"Julia Kunert, Alexander Wuttke, Hendrik van der Valk","doi":"10.1016/j.procir.2026.03.261","DOIUrl":"10.1016/j.procir.2026.03.261","url":null,"abstract":"<div><div>The circular economy (CE) plays a vital role in achieving long-term sustainability and is significantly supported by digital technologies. Among them, artificial intelligence (AI) is a frequently proposed technology in recent scientific literature due to its wide range of applications in CE, including disassembly and remanufacturing assistance, quality control, and automated waste sorting. However, real-world implementation of AI beyond theory and pilot projects faces several challenges. This paper conducts a structured literature review (SLR) to identify and analyze the key challenges in implementing AI in CE contexts. The reviewed publications are examined based on their focus on specific AI subfields and the identified challenges are categorized into overarching dimensions, including data, algorithm selection, model performance and output, model training, organization, infrastructure, finance, and environment. The SLR reveals that the primary challenge is to identify the most suitable algorithm for specific use cases. Other frequently mentioned challenges mention the lack of data and the low quality of available data. The implications of the findings are discussed, and the paper concludes by outlining future research paths.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"146 ","pages":"Pages 583-588"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148939046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Agentic AI for decentralized, natural-language communication and interaction between agents in self-organizing manufacturing systems 用于自组织制造系统中分散的、自然语言的通信和交互的代理AI

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-09-03 DOI: 10.1016/j.procir.2026.03.192

Jan Hinrich Krüger, Tobias Reichenstein, Benjamin Gutwald, Baris Eray Albayrak, Adrian Müller, Oguz Kedilioglu, Martin Krueger, Jörg Franke, Till Sindel, Jens Fürst

{"title":"Agentic AI for decentralized, natural-language communication and interaction between agents in self-organizing manufacturing systems","authors":"Jan Hinrich Krüger, Tobias Reichenstein, Benjamin Gutwald, Baris Eray Albayrak, Adrian Müller, Oguz Kedilioglu, Martin Krueger, Jörg Franke, Till Sindel, Jens Fürst","doi":"10.1016/j.procir.2026.03.192","DOIUrl":"10.1016/j.procir.2026.03.192","url":null,"abstract":"<div><div>Self-organization in manufacturing systems has been a topic of research for several decades. Due to their inherent advantages, such as increased flexibility, robustness, and adaptability to dynamic environments, self-organizing systems are considered highly relevant for addressing modern manufacturing challenges. Still, only few industrial implementations of self-organization in manufacturing systems have been achieved so far. With recent advances in AI technologies, this paper investigates the potential of agentic AI to enable self-organizing, decentralized, and agent-based manufacturing systems. Specifically, a use case of an agentic AI-driven multi-agent system for manufacturing systems utilizing the tool calling and natural language processing capabilities of large language models is presented and validated. It demonstrates that large language model-based agents can facilitate dynamic task allocation through communication in natural language, thereby increasing the autonomy of single agents and advancing the practical realization of self-organization in manufacturing systems. The main challenge remaining is utilizing emergent behavior, as the current system was too limited in size to display it.</div></div>","PeriodicalId":20535,"journal":{"name":"Procedia CIRP","volume":"146 ","pages":"Pages 90-95"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148939148","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Software-heavy Asset Administration Shells: Classification and Use Cases 重软件资产管理外壳：分类和用例

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-09-03 DOI: 10.1016/j.procir.2026.03.249

Carsten Ellwein, David Dietrich, Jessica Roth, Rozana Cvitkovic, Andreas Wortmann

引用次数: 0

Co-Production by 3D-Printing at Home – A Systematic Literature Review 联合生产的3d打印在家里-系统的文献综述

Procedia CIRP Pub Date : 2026-01-01 Epub Date: 2026-09-03 DOI: 10.1016/j.procir.2026.03.185

Wolfgang Boos (Prof. Dr.-Ing.), Thomas Eberius, Matthias Oly, Steffen Ostwald

引用次数: 0