One of the biggest issues in industrial additive manufacturing (AM) is differences between print jobs, parts in the same build, and on from one machine to the next, even if it’s the same system model. It’s all fine and dandy to promise everyone local, distributed manufacturing of the same parts in different locations, but 3D printing is not exactly very good at this. For this reason, the US Army has selected Senvol to research the ability to obtain the same components from different machines at different locations.
To do this, the company is using Senvol ML, a machine learning-based parameters divination tool. The aim is to actually make 3D printing useful, specifically for ground vehicles. The program, titled “Applying Machine Learning to Ensure Consistency and Verification of Additive Manufacturing (AM) Machine and Part Performance Across Multiple Sites” and taking place over the course of the next two years, is meant to eventually enable a methodology that would work with any 3D printing process, any material and any machine. If successful, this project will greatly enhance the deployability of AM throughout the Army. Another aim is to develop procedures for updating settings and verifying performance when variables such as machines and material are changed.
“For additive manufacturing to be successfully implemented into the Army’s supply chain, it is essential to be able to produce parts of consistent performance even if different machines are used at different locations. Today, that is much easier said than done. During this program, we are pleased to work with Senvol to demonstrate the use of its machine learning technology to aid in achieving what everyone in the additive manufacturing industry strives for – a truly flexible supply chain,” said Aaron LaLonde, Technical Specialist at U.S. Army DEVCOM Ground Vehicle Systems Center.
“Consistency – or a lack thereof – is a problem that nearly everyone in the additive manufacturing industry can relate to. The Army, and DoD in general, has been at the forefront of tackling pressing issues in our industry, and we are pleased to work with them again to demonstrate the use of our machine learning software as a mechanism to ensure consistent part performance across different sites and machines,” stated Senvol President Zach Simkin.
This is a big deal for Senvol and for our industry. The sector is moving slow as moss in deploying additive across large organizations. One of the major speed bumps is that it is difficult to design, qualify, and industrialize 3D printing. The high variability caused by designs, machines, materials, and parameters is causing the loss of thousands of production hours. These hours aren’t lost collectively. There is no joint effort to deal with all of this as a single industry. Instead, everyone is doing this work over and over again. Often different people are using the same machine and material at the same time without sharing anything with each other.
This program by the US Army may change all of that. If Senvol can demonstrate that it can eliminate run-to-run differences at varying sites, then it will find its products and services in high demand. It will have claimed a central role for itself in the future of additive as a time-saving translation engine for all parameters.