Artificial Intelligence Is Playing
An Integral Role in Additive Manufacturing

Additive manufacturing, also known as 3D printing, is the process by which a computer creates three-dimensional objects by adding layer upon layer of material. AM has an incredible number of benefits over traditional manufacturing, namely the reduced tooling costs and production time. Additive manufacturing is quickly becoming accessible to smaller businesses, who are finding the cost of entry much more affordable than traditional manufacturing.

Additive manufacturing is responsible for the production of parts in a wide variety of industries, from the aerospace industry to the automotive industry. Even healthcare is being aided by AM’s ability to help doctors create models of cancerous organs that are specific to each individual patient. This allows surgeons to give more informed pre-operative assessments and guides them during surgery. There’s no limit to the potential of additive manufacturing, and AI is quickly paving the way for its continued growth.

But one of the difficulties of additive manufacturing is the process of dialing in the different variables of your design. When 3D printing, you have to factor in variables such as speed, material, and layer thickness. This can make it difficult to maintain consistency and find a reliable process for producing the part that you need. Most users find their process through repeated trial and error, but this time-consuming method is quickly being replaced through the use of artificial intelligence and machine learning.

AI is being used by businesses such as Markforged to aid the design process and make additive manufacturing more dependable. Markforged has developed a new tool called Blacksmith that uses AI, 3D scan data and design tools to compare the design of the product with the actual 3D-printed product. This tool then makes automatic adjustments to the design in order to improve the next iteration of the product.

By using AI and 3D scanning technology, Blacksmith is able to refine the scope of what the scanner captures. Blacksmith can focus in on precisely what the scanner should look for and capture only the data relevant to your needs. It can even combat process drift by learning to account for the differences between machines. This allows additive manufacturers to quickly dial in parts and decrease production time, which streamlines the manufacturing process .

Smaller companies have begun using AI to aid additive manufacturing as well. The Denver-based machine shop Faustson Tool began exploring AI as a way to remain relevant to their customers in the aerospace and defense sectors. In order to make additive manufacturing tenable for them, they ended up partnering with other businesses, academic and public institutions to create the Alliance for the Development of Additive Processing Technologies (ADAPT) at the Colorado School of Mines.

The ADAPT Center uses AI technology to understand the interior composition of 3D-printed parts. They seek to reduce the amount of trial and error involved in additive manufacturing by finding the best machine parameters and choices with the use of AI. Machine learning in particular is helping them to determine the necessary measurements and additive builds are needed to create a better model. It’s also helping them to determine what the most important findings are in their data so that they can apply these findings on a broader scale.

Senvol is another company that uses AI and machine learning to help with additive manufacturing. Based out of New York, Senvol is a company that’s designed to provide data to help other companies implement additive manufacturing in their business. They’ve created a machine learning system called “Senvol ML” that aims to reduce the amount of testing for 3D metal printing. Because metal 3D printing has taken manufacturing by storm, and because it involves a number of variables that make testing a costly expense, Senvol’s new system is more important than ever.

The way it works is by analyzing four aspects of the data: process parameter data, process signature data, material property data, and mechanical performance data.

Senvol ML then uses this data to create a mathematical model of the scenario, and then gives the user a prediction of the four aspects of the data.

Once the initial tests are done, Senvol ML can then predict different scenarios to help the user find the correct printing parameters.

By using data from a smaller number of tests to predict a wider range of testing, the AI technology can bypass the need for endless tests and find the solution in a much faster way .

Universities have also joined in on the research into AI’s use in additive manufacturing.

A team of researchers at Carnegie Mellon University created an automated method for identifying metal AM powders. While humans were only able to correctly sort powder images with an accuracy of 50%, their method was able to accurately identify 95% of powders. This ability to identify metal AM powders could give manufacturers the ability to qualify materials much more rapidly, as well as look for any changes in between batches.

There are incredibly minute differences between metal AM powders that would take humans an inordinate amount of time to identify manually. With the help of artificial intelligence to do the brunt work, humans can have more time to develop better parts and improve their product. AI takes away the busy work of 3D printing and helps researches find new advances in additive manufacturing’s potential.

Although many companies have yet to take advantage of artificial intelligence in 3D printing, the amount of research being done on the topic is staggering. AI tools are continually being developed to aid businesses that use additive manufacturing. As more research is done, these tools will become even more widely available.