[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118733-en":3,"doc-seo-118733-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},118733,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Graph Machine Learning for Assembly Modeling - Extended Abstract","Assembly modeling is the engineering process of composing new products from a shared catalog of existing parts, and it naturally corresponds to graph structures. Graph machine learning can support tasks such as part recommendation, clustering/taxonomy creation, and anomaly detection, but the domain also brings challenges including unknown or new parts, ambiguously extracted edges, missing design-sequence information, and interaction with design engineers as users. The work outlines key modeling assumptions and presents a new assembly dataset to enable research in this semi-structural setting.","Graph Machine Learning for Assembly Modeling  \nCarola Lenzen∗  \nInstitute for Software & Systems Engineering, University of Augsburg, Augsburg, Germany  \n[lenzen@isse.de](lenzen@isse.de)  \nAlexander Schiendorfer∗  \nInstitute AImotion Bavaria, Technische Hochschule Ingolstadt, Ingolstadt, Germany  \nAlexander .Schiendorfer@thi .de  \nWolfgang Reif  \nInstitute for Software & Systems Engineering, University of Augsburg, Augsburg, Germany  \n[reif@isse.de](reif@isse.de)  \nAbstract  \nAssembly modeling refers to the design engineering process of composing assemblies (e.g., machines or machine components) from a common catalog of existing parts. There is a natural correspondence of assemblies to graphs which can be exploited for services based on graph machine learning such as part recommendation, clustering/taxonomy creation, or anomaly detection. However, this domain imposes particular challenges such as the treatment of unknown or new parts, ambiguously extracted edges, incomplete information about the design sequence, interaction with design engineers as users, to name a few. Along with open research questions, we present a novel data set.  \n1 Assembly Modeling  \nAssemblies are collections of parts that make up a product (see Figure 1) . In computer-aided design (CAD), assembly modeling refers to designing a new product based on existing parts – think of a cabinet that consists of screws, doors, and hinges; or a bike that consists of a frame, wheels, etc [1] . The connection type (e.g., welding or fastening using bolts) may contain geometric information or constraints that are also part of the assembly model. By its very nature, assembly modeling gives rise to a number of interesting novel applications for graph machine learning. Note that assembly modeling in this paper refers to the act of (iteratively) designing a new product using the same library of existing parts whereas other lines of work emphasize the computer vision perspective of perceiving physical parts (e.g. [2]) or the 3D perspective of constraining pairs of parts according to their position and relative movement (e.g., [3])– also using geometric deep learning. Our goal is to support design engineers, e.g., by suggesting next parts to insert or categorizing the existing parts by their usage.  \nSome challenges that manufacturing companies face are:  \n• Assemblies similar to existing ones frequently need to be designed and adjusted – in accordance to customer specifications (e.g., in special mechanical engineering) .  \n• Knowledge about proven part combinations (e.g., particular hinges and doors, screws and bolts, . . . ) is available to senior design engineers and may follow a desirable part management but not made explicit and enforced in CAD software.  \n• Assembly models are produced in an arbitrary sequence which depends on the designer’s individual preferences (e.g., start working on the front or back wheel of a bicycle is arbitrary); moreover, this insertion ordering is not stored in the final design by common CAD tools.  \n∗Equal contribution.  \nC. Lenzen et al., Graph Machine Learning for Assembly Modeling (Extended Abstract) . Presented at the First Learning on Graphs Conference (LoG 2022), Virtual Event, December 9–12, 2022 .  \nFigure 1: Assembly models (here, a jaw of a gripper) contain the structure of the included parts. Multiple instances of the same part type (here, A, B, C, D) may occur multiple times.  \n• Extracting a useful graph structure from CAD assembly models to begin with is not obvious. Although design engineers can define so-called “mates” relations between parts in a design to, e.g., define the rotation of a hinge, they are sometimes used for convenience in the CAD tool (cf. grouping elements) instead of actually denoting a physical connection or meaningful co-occurrence that could be reused.  \nIn this extended abstract, we highlight opportunities for the graph machine learning community to work on CAD assembly modeling as a novel application a","cbCaikPfxNblTKNV","https://ap.wps.com/l/cbCaikPfxNblTKNV","pdf",504505,1,7,"English","en",105,"# Assembly Modeling\n## Graph representation and assumptions\n# Graph ML Use Cases in Assembly Modeling","[{\"question\":\"What does assembly modeling mean in this work?\",\"answer\":\"It refers to designing a new product by iteratively composing assemblies from a common library of existing parts, as used in computer-aided design (CAD).\"},{\"question\":\"How are assemblies represented for graph machine learning?\",\"answer\":\"Each assembly is modeled as an undirected, unweighted graph whose nodes are heterogeneous part types and whose edges represent connectivity between parts. Geometry is initially ignored, while additional edge or node features may be added later.\"},{\"question\":\"Why is permutation invariance important in this domain?\",\"answer\":\"CAD assemblies have no inherent order, and the insertion sequence of parts is not stored in final designs, so part recommendation must be insensitive to node ordering.\"}]","Graph Machine Learning for Assembly Modeling - Extended Abstract | PDF",1785719975,18,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"graph-machine-learning-for-assembly-modeling-extended-abstract","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/graph-machine-learning-for-assembly-modeling-extended-abstract/118733/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What does assembly modeling mean in this work?","Question",{"text":76,"@type":77},"It refers to designing a new product by iteratively composing assemblies from a common library of existing parts, as used in computer-aided design (CAD).","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How are assemblies represented for graph machine learning?",{"text":81,"@type":77},"Each assembly is modeled as an undirected, unweighted graph whose nodes are heterogeneous part types and whose edges represent connectivity between parts. Geometry is initially ignored, while additional edge or node features may be added later.",{"name":83,"@type":74,"acceptedAnswer":84},"Why is permutation invariance important in this domain?",{"text":85,"@type":77},"CAD assemblies have no inherent order, and the insertion sequence of parts is not stored in final designs, so part recommendation must be insensitive to node ordering.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]