How Data-Powered 3D Printers Will Change Manufacturing

Machine learning can optimize hardware to increase printing speeds and improve resolution

Published: Monday, October 30, 2023 - 11:02

We’re all familiar with photos of Ford’s production lines in 1920. But would we recognize them today? As part of a broader trend referred to as “Industry 4.0,” systems in many factories have modernized considerably in recent years. This digitization of the manufacturing sector aims to apply emerging technologies—such as cloud computing, smart automation, IoT devices, digital inventories, and data analytics—to use data to help engineers make better decisions, improve factory processes, and ultimately require less human oversight.

At the center of the factory’s shift toward digital is additive manufacturing (AM), known colloquially as 3D printing, which enables the core suite of Industry 4.0 technologies to be directly applied to the process of fabrication itself. 3D printers have come a long way from their humble beginnings as rapid prototyping machines and gimmick factories for novelty items. They’ve evolved not just into machines that fabricate high-performance parts for airplanes, but into data collection hubs that gather massive quantities of information about how different parts are built during the fabrication process.

Similar to how autonomous vehicles collect and apply data to continuously improve a car’s ability to drive, connected 3D printers can use collected data for artificial intelligence-powered automation. During each print job, 3D printers produce large quantities of data that are sent to and stored in the cloud. The print job data—ripe for AI, machine learning, and automation-based product features—can then be fed to algorithms, which printers and users can access through the cloud. Among other things, these data help businesses make decisions about what parts to print and how best to print them, while improving the quality of print jobs.

AI steps in

With the current state of manufacturing technology, AI can take a hardware problem and offer a software solution. Machine learning can optimize hardware, automatically enhancing 3D printers through software updates to increase printing speeds and improve resolution. AI can help businesses determine which parts, when produced in-house through additive manufacturing, will have the biggest effect on their bottom line. It can also connect to digital catalogs of existing parts and detect which specific parts are the best candidates to be printed through various AM techniques.

With connectivity to cloud software, 3D printers can use machine learning to automatically generate tooling jigs or fixtures to hold the parts they print. Manufacturers with more mature AM proficiencies are applying AI-based optimizations during the design stage of new parts—simulating how the digital design for a part, once printed, will perform under specific loads. AI is also employed in AM to detect print failures (proactively pausing prints when needed), and to inspect parts as they’re being printed to ensure quality and conformance.

Much like Tesla cars, the same hardware out in the field is consistently learning, improving, and getting smarter with every over-the-air update. As providers advance the quality of information collected during fabrication and build modes of collecting data about how each 3D printed part does its job in the field, manufacturing technology is approaching a fully automated “closed-loop” printing process. It can simply be presented with a real-world manufacturing problem to solve, and then design and build the part using the specific digital fabrication technology that makes the most sense, given the defined time, cost, and performance constraints.

This smart, closed-loop automation of fabrication lifts a massive burden off manufacturers. The ability to automate the fabrication of parts—in a manner that is both quick and reliable—substantially increases manufacturers’ outputs and production speeds. And while AM inherently streamlines the process of building parts, each savvy application of data collected by the printers helps streamline distinct points within the AM process. Engineers can move away from annoying manual tasks that are now automated—such as quality inspection and manually designing tooling or fixtures for parts—and focus on innovating and solving more interesting problems. Machinists are also freed up to focus on fabricating the parts that either contribute the most business value or can only be made through machining.

Improving the supply chain

The biggest area for these data-driven 3D printers to make an impact? The global supply chains of manufacturers—where inability to swiftly address ongoing disruptions in an agile manner often creates massive barriers to success. Adopting just-in-time manufacturing processes created a world of brittle supply chains that were unable to keep pace as the supply chain disruptions of the pandemic washed over the entire globe. The year 2020 exposed this precariousness.

The state of global supply chains hasn’t meaningfully improved in subsequent years, and manufacturing in 2022 is still not for the faint of heart. While the complexity and all-around difficulty of procuring parts and materials continues to result in countless delays, shortages, and soaring prices, it’s a problem that’s avoidable with some proactive action—and the world is catching on. With the introduction of federal initiatives like the U.S. government’s Additive Manufacturing Forward program, global supply chains are increasingly turning to 3D printing for fast lead times and products that can be conveniently delivered at points of need.

However, the supply chain value of smarter 3D printers is still relatively understated. Smarter 3D printers are more precise and fail-proof; their closed-loop learning can add another layer of insurance against untimely holdups and failures. Considering today’s labor shortages, the newer smart automation features that vendors are bringing to market will probably also make 3D printing more attractive for labor-strapped manufacturers, fueling AM adoption and ultimately stabilizing supply.

Furthermore, augmenting human decision-making in the critical problem-solving stages and fully automating many manual processes snips off even more of part lead times, on top of the time saved from fabrication speeds and convenient point-of-need logistics.

Tip of the iceberg

As AI’s presence in 3D printing continues to evolve as we speak, workflows found in factories increasingly resemble the industry’s vision for fully automated manufacturing. Given how smart AM is addressing today’s broken supply chains, imagine the effect of the smartest 3D printers, which are still yet to come. Fully automated part design, fabrication, and quality control for given problems won’t just bring the person-power required to address many acute supply-chain challenges down to near nil and further abridge lead times. It will also uncover part designs that allow manufacturers to get the best performance and business returns from every 3D-printed part, and uncover designs that can solve the most challenging problems comparable to top engineering talent.

Additive-manufacturing adoption continues to expand to address ongoing supply chain challenges. Alongside AM, Industry 4.0 technologies continue to push the boundaries of what 3D printers are capable of and the utility they provide for businesses.

For the manufacturing industry, the view looking forward is clear. Now, the open question should prompt us to look inward. Given the pace of AI adoption in other industries, why has it taken this long to bring effective AI to manufacturing, and what is preventing more manufacturers from adopting these technologies faster and more comprehensively? The answer might lie in the perpetual struggle between innovation and risk reduction—an issue that AI could prove to be a false dichotomy.

About The Author

Julie van der Hoop

Julie van der Hoop is a software product manager at Markforged who partners with customers to help them scale in their adoption of additive manufacturing. After working with enterprise companies to transform their operations, Julie knows the business value of automation and integration. Julie holds a Ph.D. from MIT.