Artificially Intelligent Engineering (AIE) software firm Monolith AI and Imperial College London have received a £500,000 grant from the UK’s innovation agency Innovate UK to build a new type of AI tool capable of assessing the manufacturability of metal components.
The goal of the project is to revolutionize Computer-Aided Engineering (CAE) within the manufacturing industry through building a new version of ‘explainable AI’ that can provide clear feedback to engineers on why a part may not be manufacturable.
“CAE has done a fantastic job advancing component manufacturing, but there are still many areas where physical simulations still cannot capture the true complexity of components,” said Dr Richard Ahlfeld, CEO and Founder of Monolith AI. “Large engineering companies collect a lot of data when assessing manufacturability and our goal is to make that data work to their advantage.
“This latest funding will allow us to explore this possibility and drive not only the automotive industry, but other sectors, forwards.”
AI-driven 3D printing
Over the past year, substantial progress has been made in the development of machine learning (ML) and AI-driven 3D printing technologies. In fact, AI was named alongside 3D printing, green hydrogen, and autonomous sensors in Lux Research’s top 12 emerging technologies to watch in 2021, and more recently has been projected to continue to play a key role over the next year by additive manufacturing leaders sharing their 2022 3D printing trends.
AI software firm Oqton, now acquired by 3D printer manufacturer 3D Systems, is one of those leading the charge in this field. Its ML and AI-powered software platform is able to automate, accelerate, and optimize the throughput of manufacturing firms by streamlining their production workflows.
Metal and composite 3D printer manufacturer Markforged has also embraced AI with the launch of its AI-based Blacksmith software for use with its X7 3D printer. The software deploys a patented scanning algorithm that measures the precision of parts as they are being printed and collects data to improve the quality of future jobs.
The past year has seen 3D printing data specialist Senvol begin developing an ML software with “additional capabilities” to address the US Department of Defense’s (DoD) production needs, and researchers at Oak Ridge National Laboratory (ORNL) unveil their Peregrine AI-driven real-time monitoring software which they claim could potentially turn systems into ‘self-correcting machines.’
Elsewhere, software developer Autodesk has become a founding partner of the nFrontier Emerging Technology Center, a dynamic lab environment seeking to integrate the “Emerging Eight Technologies”, of which AI is one, under one roof. Fraunhofer ILT has also recently leveraged the benefits of AI to develop a 3D printed sensor system capable of intelligently maintaining train components, called SenseTrAIn.
Assessing part manufacturability
With the Innovate UK grant, Monolith AI and Imperial College London will work with a cluster of industrial partners to build their novel AI tool over the next 18 months. By the end of the project, the partners hope to have developed an AI tool that will be able to assess if metal components are manufacturable and, if they are not, be capable of explaining why this is the case.
Essentially, the project will seek to build on the capabilities of current CAE simulations to help manufacturers not only predict whether a part can be successfully manufactured or not, but to assess the simulation results through ‘explainable AI.’
“Knowing that a door is not manufacturable is not enough,” said project leader Dr Joël Henry. “You need to understand why, and even more importantly how you could change the design and operating conditions to make it manufacturable.”
Once built, it is hoped the AI tool will be able to provide clear feedback to engineers on how it arrived at its conclusions in order to remove what the project partners call the ‘black box’ dilemma. The partners also hope to streamline the manufacturing process by leveraging AI learnings from what could be manufactured in the past to predict what would be best for new components and provide a new competitive advantage to high-volume manufacturers.
Using the AI tool, engineers could potentially build expert simulations based on repetitive tasks and historic data, and run complex manufacturability assessments in seconds rather than weeks. This would in turn free up engineering skills for other important tasks within the manufacturing workflow.
Regarding the AI tool’s potential in relation to 3D printing, Ahlfeld said: “One possible application of this technology for 3D printing is to assess based on a CAD file whether this geometry can be printed by your printing facility or not. Imagine an online prototype website where you can upload your file – this solution could tell you if the geometry can be printed having learnt from previous designs.
“More, it could even highlight the parts of the geometry that could not be printed and explain why.”
Subscribe to the 3D Printing Industry newsletter for the latest news in additive manufacturing. You can also stay connected by following us on Twitter and liking us on Facebook.
Looking for a career in additive manufacturing? Visit 3D Printing Jobs for a selection of roles in the industry.
Subscribe to our YouTube channel for the latest 3D printing video shorts, reviews and webinar replays.
Featured image shows Dr Richard Ahlfeld, CEO and Founder of Monolith AI. Photo via Monolith AI.
