[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-650629-105":59,"doc-detail-650629-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","from-pid-drawings-to-process-graphs-a-multimodal-language-model-approach","From P&ID Drawings to Process Graphs: A Multimodal Language Model Approach","","Piping and instrumentation diagrams (P&IDs) encode the functional structure of process plants but are difficult to digitise for digital twins due to heterogeneous drawing standards and brittleness in symbol recognition and connectivity reconstruction. This work reframes P&ID digitisation as extracting equipment tags and inferring process topology via a two-stage multimodal large language model workflow. Visual extraction and topology reasoning are separated and guided by chemical engineering process knowledge, outperforming end-to-end digitisation on two ANSI-standard case studies as more accurate and structurally consistent representations.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/from-pid-drawings-to-process-graphs-a-multimodal-language-model-approach/650629/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/from-pid-drawings-to-process-graphs-a-multimodal-language-model-approach/650629.png","ImageObject",300,407,{"name":92,"@type":93},"Rowan","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-29","2026-10-24",true,{"@type":102,"interactionType":103,"userInteractionCount":34},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does the paper address in P&ID digitisation?","Question",{"text":112,"@type":113},"Digitising legacy P&IDs is challenging because drawing standards vary and existing methods rely on brittle symbol recognition and rule-based connectivity reconstruction.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What is the proposed approach?",{"text":117,"@type":113},"The method uses a two-stage workflow with multimodal large language models, treating visual extraction of equipment tags and topology reconstruction as separate reasoning stages.",{"name":119,"@type":110,"acceptedAnswer":120},"How is the approach evaluated?",{"text":121,"@type":113},"It is evaluated on two ANSI-standard P&ID case studies of increasing complexity, showing improved accuracy and structural consistency compared with end-to-end digitisation.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},650629,1792994499,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":34,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":39,"language":139,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":67,"update_tm":143,"read_time":46},1099514067415,"https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502","Original Research Article  \nPeer Reviewed Conference Proceeding  \nESCAPE 36-European Symposium on Computer Aided Process Engineering Sheffield, UK. 21-24 June 2026  \nFrom P&ID Drawings to Process Graphs: A Multimodal Language Model Approach  \nBaikai Zhua, Samuel Duonga, Javal Vyasa,*, Mehmet Mercangöza  \na Imperial College London, Department of Chemical Engineering, London, Greater London, United Kingdom  \n* Corresponding Author: [j.vyas24@imperial.ac.uk](j.vyas24@imperial.ac.uk)  \nABSTRACT  \nPiping and instrumentation diagrams (P&IDs) encode the functional structure of process plants and are a critical yet underutilised source of engineering knowledge for digital twins and intelligent decision support. However, digitising legacy P&IDs remains challenging due to heterogeneous drawing standards and the reliance of existing methods on brittle symbol recognition and rulebased connectivity reconstruction. This work reframes P&ID digitization as the extraction of equipment tags and inference of process topology, rather than graphical reproduction. We propose a two-stage workflow based on multimodal large language models, in which visual extraction and topology reconstruction are treated as distinct reasoning stages guided by chemical engineering process knowledge. The approach is evaluated on two ANSI-standard P&ID case studies of increasing complexity. Results show that decomposing visual extraction and topology reasoning yields more accurate and structurally consistent process representations than end-to-end digitization, highlighting the potential of language-model-based, knowledge-guided workflows for scalable and semantically reliable P&ID digitization.  \nKeywords: P&ID digitisation, Multimodal large language models, Graph reconstruction  \nINTRODUCTION  \nPiping and instrumentation diagrams (P&IDs) constitute the structural backbone of a process plant, capturing how equipment, instrumentation, and control elements are interconnected to realise the intended process operation. Beyond providing spatial or locational information, P&IDs encode the functional logic of the plant, describing how units interact, how material and signals propagate, and how control strategies are implemented. As such, they represent an authoritative source of process understanding that spans design intent, operational constraints, and safety considerations.  \nDespite their central role, the information embedded in P&IDs remains largely underexploited in industrial practice. In many plants, P&IDs continue to be treated primarily as static reference documents rather than as structured, machine-interpretable sources of knowledge. This limits their integration into advanced engineering workflows, where explicit representations of process topology could support automated consistency checking, decision support, and interaction with digital twins. Asthe process industries move toward smart factories and  \ndata-driven operation, the ability to interrogate and reason over P&IDs becomes increasingly important.  \nA major barrier to such “smart” use of P&IDs is the legacy nature of existing documentation. A substantial fraction of industrial P&IDs still exists as paper drawings or unstructured raster images, often produced over decades using heterogeneous drafting standards and conventions. Converting these legacy artefacts into structured, machine-readable representations remains labour-intensive and typically requires extensive manual interpretation by domain experts. This manual effort does not scale and represents a significant obstacle to unlocking the full value of P&IDs within modern digital infrastructures, including retrieval-augmented generation systems and interactive engineering tools.  \nFrom a technical standpoint, P&ID digitization is challenging due to variability in symbol rendering, annotation practices, and drawing quality across companies and engineering contractors. While human engineers resolve such variability using contextual and domain knowled","cbCaii4wqMO5tf7y","https://ap.wps.com/l/cbCaii4wqMO5tf7y","pdf",994057,"English","# Abstract\n# Introduction\n# Literature Review","[{\"question\":\"What problem does the paper address in P\\u0026ID digitisation?\",\"answer\":\"Digitising legacy P\\u0026IDs is challenging because drawing standards vary and existing methods rely on brittle symbol recognition and rule-based connectivity reconstruction.\"},{\"question\":\"What is the proposed approach?\",\"answer\":\"The method uses a two-stage workflow with multimodal large language models, treating visual extraction of equipment tags and topology reconstruction as separate reasoning stages.\"},{\"question\":\"How is the approach evaluated?\",\"answer\":\"It is evaluated on two ANSI-standard P\\u0026ID case studies of increasing complexity, showing improved accuracy and structural consistency compared with end-to-end digitisation.\"}]","From P&ID Drawings to Process Graphs: A Multimodal Language Model Approach | PDF",1792835174]