{"id":12257,"date":"2025-08-11T10:00:52","date_gmt":"2025-08-11T08:00:52","guid":{"rendered":"https:\/\/sxe-consulting.com\/?p=12257"},"modified":"2025-07-21T12:34:36","modified_gmt":"2025-07-21T10:34:36","slug":"lia-a-2d-drawing-generation-aid","status":"publish","type":"post","link":"https:\/\/sxe-consulting.com\/en\/lia-une-aide-a-la-generation-des-plans-2d\/","title":{"rendered":"AI as an aid to generating 2D drawings"},"content":{"rendered":"<p data-start=\"553\" data-end=\"841\">At a time when artificial intelligence (AI) is transforming many industrial sectors, one question comes up regularly in workshops and design offices:<br data-start=\"718\" data-end=\"721\" \/><strong data-start=\"721\" data-end=\"841\">\u201cWhy can\u2019t AI directly generate 2D drawings for the manufacture of mechanical parts?\u201d<\/strong><\/p>\n<p data-start=\"843\" data-end=\"1080\">On paper, it seems simple: you would just describe a part or provide a sketch, and a well-trained AI would produce a dimensioned, standardised drawing ready to send to the workshop. In reality, things are far more complex.<\/p>\n<p data-start=\"1082\" data-end=\"1475\">At SXE-Consulting, we support industrial companies in their digital transformation and in optimising their technical processes. This naturally includes the interface between the physical world (workshop, manufacturing, maintenance) and digital tools (CAD, ERP, <a href=\"https:\/\/sxe-consulting.com\/en\/how-manufacturing-execution-systems-are-revolutionizing-industrial-efficiency\/\">MES<\/a>). Today, let us take stock of the current limitations of AI in generating 2D mechanical drawings.<\/p>\n<h3 data-start=\"1482\" data-end=\"1541\">1. A manufacturing drawing is much more than a sketch<\/h3>\n<p data-start=\"1543\" data-end=\"1711\">A 2D drawing intended for mechanical manufacturing is not a simple sketch. It must comply with strict standards (ISO, <a href=\"https:\/\/www.asme.org\/\" target=\"_blank\" rel=\"noopener\">ASME<\/a>\u2026) and contain precise information on:<\/p>\n<ul data-start=\"1712\" data-end=\"1974\">\n<li data-start=\"1712\" data-end=\"1760\">\n<p data-start=\"1714\" data-end=\"1760\">les <strong data-start=\"1718\" data-end=\"1759\">geometric dimensions and tolerances<\/strong>,<\/p>\n<\/li>\n<li data-start=\"1761\" data-end=\"1801\">\n<p data-start=\"1763\" data-end=\"1801\">les <strong data-start=\"1767\" data-end=\"1800\">surface finishes and roughness values<\/strong>,<\/p>\n<\/li>\n<li data-start=\"1802\" data-end=\"1867\">\n<p data-start=\"1804\" data-end=\"1867\">les <strong data-start=\"1808\" data-end=\"1830\">raw materials<\/strong> and any heat treatments,<\/p>\n<\/li>\n<li data-start=\"1868\" data-end=\"1922\">\n<p data-start=\"1870\" data-end=\"1922\">les <strong data-start=\"1874\" data-end=\"1921\">welding, threading and machining symbols<\/strong>,<\/p>\n<\/li>\n<li data-start=\"1923\" data-end=\"1974\">\n<p data-start=\"1925\" data-end=\"1974\">les <strong data-start=\"1929\" data-end=\"1973\">functional and assembly requirements<\/strong>.<\/p>\n<\/li>\n<\/ul>\n<p data-start=\"1976\" data-end=\"2175\">This level of detail cannot be improvised. It stems from <strong data-start=\"2043\" data-end=\"2072\">engineering reasoning<\/strong> that takes into account the mechanical, thermal, environmental and economic constraints of the product.<\/p>\n<p data-start=\"2177\" data-end=\"2370\">Today, generative AI (such as ChatGPT or other specialised systems) <strong data-start=\"2255\" data-end=\"2336\">does not yet spontaneously incorporate all these normative and functional elements<\/strong>, unless it is guided step by step.<\/p>\n<h3 data-start=\"2377\" data-end=\"2437\">2. Lack of technical context and specifications<\/h3>\n<p data-start=\"2439\" data-end=\"2586\">An industrial drawing cannot be produced generically. The same geometric shape may meet very different requirements depending on:<\/p>\n<ul data-start=\"2587\" data-end=\"2852\">\n<li data-start=\"2587\" data-end=\"2637\">\n<p data-start=\"2589\" data-end=\"2637\">the sector (aerospace, automotive, energy\u2026),<\/p>\n<\/li>\n<li data-start=\"2638\" data-end=\"2703\">\n<p data-start=\"2640\" data-end=\"2703\">the function of the part (transmission, sealing, structure\u2026),<\/p>\n<\/li>\n<li data-start=\"2704\" data-end=\"2773\">\n<p data-start=\"2706\" data-end=\"2773\">the operating environment (temperature, pressure, vibrations\u2026),<\/p>\n<\/li>\n<li data-start=\"2774\" data-end=\"2852\">\n<p data-start=\"2776\" data-end=\"2852\">the available manufacturing processes (milling, moulding, 3D printing\u2026).<\/p>\n<\/li>\n<\/ul>\n<p data-start=\"2854\" data-end=\"3081\">AI needs a <strong data-start=\"2876\" data-end=\"2926\">clear, structured and complete set of specifications<\/strong> to propose a technically viable drawing. Yet these specifications are rarely formulated in a language that an AI can interpret autonomously.<\/p>\n<p data-start=\"3083\" data-end=\"3300\">At SXE-Consulting, this is precisely where we help our clients to <strong data-start=\"3140\" data-end=\"3179\">structure their technical data<\/strong>, so that it becomes usable in digital systems, including with automation or AI tools.<\/p>\n<h3 data-start=\"3307\" data-end=\"3376\">3. AI is not (yet) integrated into industrial CAD tools<\/h3>\n<p data-start=\"3378\" data-end=\"3538\">Another important limitation lies in the <strong data-start=\"3422\" data-end=\"3468\">software tools used in industry<\/strong>. CAD (Computer-Aided Design) software such as:<\/p>\n<ul data-start=\"3539\" data-end=\"3622\">\n<li data-start=\"3539\" data-end=\"3555\">\n<p data-start=\"3541\" data-end=\"3555\"><strong data-start=\"3541\" data-end=\"3555\">SolidWorks<\/strong><\/p>\n<\/li>\n<li data-start=\"3556\" data-end=\"3567\">\n<p data-start=\"3558\" data-end=\"3567\"><strong data-start=\"3558\" data-end=\"3567\">CATIA<\/strong><\/p>\n<\/li>\n<li data-start=\"3568\" data-end=\"3582\">\n<p data-start=\"3570\" data-end=\"3582\"><strong data-start=\"3570\" data-end=\"3582\">Inventor<\/strong><\/p>\n<\/li>\n<li data-start=\"3583\" data-end=\"3597\">\n<p data-start=\"3585\" data-end=\"3597\"><strong data-start=\"3585\" data-end=\"3597\">PTC Creo<\/strong><\/p>\n<\/li>\n<li data-start=\"3598\" data-end=\"3622\">\n<p data-start=\"3600\" data-end=\"3622\"><strong data-start=\"3600\" data-end=\"3622\">AutoCAD Mechanical<\/strong><\/p>\n<\/li>\n<\/ul>\n<p data-start=\"3624\" data-end=\"3747\">are complex and highly standardised environments, often coupled with technical data management tools (PDM\/PLM).<\/p>\n<p data-start=\"3749\" data-end=\"3998\">Today, generative AIs are not natively integrated into these tools. They therefore cannot create a directly usable 2D drawing, that is to say <strong data-start=\"3912\" data-end=\"3959\">in the form of a DWG, DXF or dimensioned PDF file<\/strong>, in line with industrial practice.<\/p>\n<p data-start=\"4000\" data-end=\"4217\">Some solutions are emerging in more recent tools such as <strong data-start=\"4063\" data-end=\"4077\">Fusion 360<\/strong> or <strong data-start=\"4081\" data-end=\"4092\">Onshape<\/strong>, which offer intelligent assistance modules. But we are still a long way from complete and robust automation.<\/p>\n<h3 data-start=\"4224\" data-end=\"4279\">4. Engineering is not just geometry<\/h3>\n<p data-start=\"4281\" data-end=\"4454\">An AI can imagine a part geometrically. But mechanical engineering also relies on physical principles, calculations and functional reasoning:<\/p>\n<ul data-start=\"4455\" data-end=\"4653\">\n<li data-start=\"4455\" data-end=\"4501\">\n<p data-start=\"4457\" data-end=\"4501\">Will the part withstand the load?<\/p>\n<\/li>\n<li data-start=\"4502\" data-end=\"4566\">\n<p data-start=\"4504\" data-end=\"4566\">Is it compatible with the other elements of the assembly?<\/p>\n<\/li>\n<li data-start=\"4567\" data-end=\"4620\">\n<p data-start=\"4569\" data-end=\"4620\">Can it be machined with the available resources?<\/p>\n<\/li>\n<li data-start=\"4621\" data-end=\"4653\">\n<p data-start=\"4623\" data-end=\"4653\">What is its cost price?<\/p>\n<\/li>\n<\/ul>\n<p data-start=\"4655\" data-end=\"4873\">These decisions still fall within <strong data-start=\"4688\" data-end=\"4713\">the engineer\u2019s profession<\/strong>. AI can assist, suggest and validate certain hypotheses, but it does not (yet) replace the reasoning of an experienced technician or designer.<\/p>\n<h3 data-start=\"4880\" data-end=\"4938\">5. The risks of faulty manufacturing are too high<\/h3>\n<p data-start=\"4940\" data-end=\"5009\">In an industrial environment, a poorly defined drawing can lead to:<\/p>\n<ul data-start=\"5010\" data-end=\"5158\">\n<li data-start=\"5010\" data-end=\"5033\">\n<p data-start=\"5012\" data-end=\"5033\">un <strong data-start=\"5015\" data-end=\"5032\">costly scrap<\/strong>,<\/p>\n<\/li>\n<li data-start=\"5034\" data-end=\"5073\">\n<p data-start=\"5036\" data-end=\"5073\">des <strong data-start=\"5040\" data-end=\"5072\">extended production lead times<\/strong>,<\/p>\n<\/li>\n<li data-start=\"5074\" data-end=\"5107\">\n<p data-start=\"5076\" data-end=\"5107\">une <strong data-start=\"5080\" data-end=\"5106\">quality non-conformity<\/strong>,<\/p>\n<\/li>\n<li data-start=\"5108\" data-end=\"5158\">\n<p data-start=\"5110\" data-end=\"5158\">or even risks of <strong data-start=\"5131\" data-end=\"5157\">failure in service<\/strong>.<\/p>\n<\/li>\n<\/ul>\n<p data-start=\"5160\" data-end=\"5300\">This is why companies remain cautious and prefer to rely on proven tools and rigorous validation processes.<\/p>\n<h3 data-start=\"5307\" data-end=\"5362\">So what role can AI play today?<\/h3>\n<p data-start=\"5364\" data-end=\"5478\">Even if AI does not yet generate 2D drawings ready for manufacturing, it can already provide many services:<\/p>\n<ul data-start=\"5479\" data-end=\"5807\">\n<li data-start=\"5479\" data-end=\"5556\">\n<p data-start=\"5481\" data-end=\"5556\"><strong data-start=\"5481\" data-end=\"5518\">Generation of concept sketches<\/strong>, to help during the design phase,<\/p>\n<\/li>\n<li data-start=\"5557\" data-end=\"5636\">\n<p data-start=\"5559\" data-end=\"5636\"><strong data-start=\"5559\" data-end=\"5589\">Analysis of existing drawings<\/strong>, to detect inconsistencies or omissions,<\/p>\n<\/li>\n<li data-start=\"5637\" data-end=\"5702\">\n<p data-start=\"5639\" data-end=\"5702\"><strong data-start=\"5639\" data-end=\"5701\">Translation of business needs into technical specifications<\/strong>,<\/p>\n<\/li>\n<li data-start=\"5703\" data-end=\"5748\">\n<p data-start=\"5705\" data-end=\"5748\"><strong data-start=\"5705\" data-end=\"5747\">Assistance with functional dimensioning<\/strong>,<\/p>\n<\/li>\n<li data-start=\"5749\" data-end=\"5807\">\n<p data-start=\"5751\" data-end=\"5807\"><strong data-start=\"5751\" data-end=\"5806\">Searching libraries of standardised components<\/strong>.<\/p>\n<\/li>\n<\/ul>\n<p data-start=\"5809\" data-end=\"5838\">And tomorrow, we can imagine:<\/p>\n<ul data-start=\"5839\" data-end=\"6052\">\n<li data-start=\"5839\" data-end=\"5921\">\n<p data-start=\"5841\" data-end=\"5921\">design assistants capable of automatically incorporating tolerances,<\/p>\n<\/li>\n<li data-start=\"5922\" data-end=\"5968\">\n<p data-start=\"5924\" data-end=\"5968\">AI copilots integrated into CAD software,<\/p>\n<\/li>\n<li data-start=\"5969\" data-end=\"6052\">\n<p data-start=\"5971\" data-end=\"6052\">platforms for the automatic configuration of parts based on a customer catalogue.<\/p>\n<\/li>\n<\/ul>\n<h3 data-start=\"6059\" data-end=\"6073\">Conclusion<\/h3>\n<p data-start=\"6075\" data-end=\"6443\">Artificial intelligence opens up many prospects for the manufacturing industry. But when it comes to <strong data-start=\"6188\" data-end=\"6256\">the automatic generation of 2D drawings for mechanical manufacturing<\/strong>, we are not there yet. It requires a combination of human skills, integrated digital tools and a normative rigour that AI does not yet fully master.<\/p>\n<p data-start=\"6445\" data-end=\"6669\">At SXE-Consulting, we are convinced that the <strong data-start=\"6496\" data-end=\"6553\">digital transformation of industrial companies<\/strong> will come about through the progressive and intelligent integration of these technologies, in support of existing expertise.<\/p>\n<p data-start=\"6671\" data-end=\"6899\">If you would like to discuss the possibilities of automation, MES\/ERP\/CAD integration, or the structuring of your technical data, contact us. We will help you turn your ideas into concrete, effective projects.<\/p>","protected":false},"excerpt":{"rendered":"<p>\u00c0 l\u2019heure o\u00f9 l\u2019intelligence artificielle (IA) transforme de nombreux secteurs industriels, une question revient r\u00e9guli\u00e8rement dans les ateliers et bureaux d\u2019\u00e9tudes :&#8220;Pourquoi l\u2019IA ne peut-elle pas g\u00e9n\u00e9rer directement des plans 2D pour la mise en fabrication de pi\u00e8ces m\u00e9caniques ?&#8221; Sur le papier, cela semble simple : il suffirait de d\u00e9crire une pi\u00e8ce ou de [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":12258,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_feature_clip_id":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_post_was_ever_published":false},"categories":[85],"tags":[],"class_list":["post-12257","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ingenierie"],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"https:\/\/sxe-consulting.com\/wp-content\/uploads\/2025\/07\/C02-1.jpg","_links":{"self":[{"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/posts\/12257","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/comments?post=12257"}],"version-history":[{"count":0,"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/posts\/12257\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/media\/12258"}],"wp:attachment":[{"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/media?parent=12257"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/categories?post=12257"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sxe-consulting.com\/en\/wp-json\/wp\/v2\/tags?post=12257"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}