The cloud-based platform is designed to interconnect all of the Markforged systems currently being used around the world, a figure which reportedly stands at over 12,000. The Digital Forge also claims to be the first of its kind to use machine learning, a feature which enables the company’s Eiger print preparation software to constantly learn from the 12,000 systems in the 73 country-wide global fleet. As such, every print on a connected Markforged system should theoretically be more accurate than the last.
A cloud-first, software-first philosophy
The Digital Forge uses what the company calls fleet federated learning, an interconnected form of ML that relies on data from other nodes, or other printers, in an expansive network. By sharing and receiving printer data on an ongoing basis, the platform itself gets ‘smarter’ day-by-day, enabling users to leverage the latest advancements over the cloud. The idea is that the Digital Forge and its printers should become better equipped at correcting print jobs mid-print, reducing the number of global print failures on Markforged systems over time, all while increasing part qualities.
Since the platform is cloud resource-based it takes the load off local systems. As such, the Digital Forge can be used to print both metal and carbon fiber-reinforced parts from just a single browser tab.
Greg Mark, Chairman of Markforged, states: “We started Markforged with a cloud-first, software-first approach that was designed for the modern world, and now we are applying that approach to accomplish things people thought were still decades away from coming to market. Through the Digital Forge, manufacturers can use our powerful software to easily fabricate strong, accurate, and durable metal and composite parts for orders of magnitude cheaper than they’ve traditionally been made — on-demand and directly at the point of need.”
The web of global customers
Looking at the list of 12,000 Markforged customers, we can see some pretty big names, including Siemens, Porsche, and Microsoft. The company is also proud to announce it serves the 10 largest aerospace companies, 12 of the 14 biggest automotive manufacturers, and 5 of the 6 branches in the U.S. Armed Forces, each of which contributes to the Digital Forge platform.
Mark concludes: “Electricity was invented in 1880, but it took 40 years and the pandemic of 1918 to spark the Industrial Revolution that built our modern world. 3D printing has reached a similar tipping point. We are nearing the 40th anniversary of the 3D printer (2026), and I believe the pandemic of 2020 and the supply chain disruption it has caused will usher in the next great Industrial Revolution — the era of Digital Manufacturing — and we are on a mission to put The Digital Forge in every factory on Earth as part of that revolution.”
While the cloud-based approach is a novel one, there have been previous cases of ML-based methods used to improve the quality of 3D printed parts. A team of researchers from Argonne National Laboratory and Texas A&M University recently developed one such method. Using real-time temperature data, together with machine learning algorithms, the scientists were able to make correlative links between thermal history and the formation of subsurface defects during the laser PBF process, enabling intelligent defect detection.
Elsewhere, at the Swinburne University of Technology, a researcher has used machine learning to give insight into the compressive strength of 3D printed construction materials. With the aim of developing a process for classifying different 3D printed geopolymer samples, the researcher targeted specific variables, and optimized the makeup of the 3D printed materials using machine learning models.
Looking for a career in additive manufacturing? Visit 3D Printing Jobs for a selection of roles in the industry.
Featured image shows the web of cloud-based connections on the Digital Forge platform. Image via Markforged.