Additive Manufacturing

JAN 2018

ADDITIVE MANUFACTURING is the magazine devoted to industrial applications of 3D printing and digital layering technology. We cover the promise and the challenges of this technology for making functional tooling and end-use production parts.

Issue link:

Contents of this Issue


Page 7 of 43

JANUARY 2018 Additive Manufacturing Something to Add 6 We Will Skip Ahead 100 Years Every part-making process is subject to variables that affect its outcome, but with additive manufacturing, there is a difference of degree. In machining, for example, we know the impact of choices of tooling, cutting fluid, feed rate and so on, and in molding we know the effect of characteristics of the material and mold design. By comparison, when it comes to an additive process such as selective laser melting in metal, there are more variables and more combinations of variables affecting the form and properties of the part than we have yet recognized, let alone mastered. But additive has something going for it. Additive will bene- fit from one other important difference: a difference of timing. Whereas other manufacturing processes came of age well in the past, additive is maturing during a time of advanced computing power. As a result, it appears likely that machine learning—the theme of this month's issue—will prove to be a major factor in the way that AM advances. Specifically, it will enable the speed of AM's advance. The term "machine learning" does not imply a type of calculation humans can't do, but instead a volume of calculations. A billion different relationships among many inputs and outputs can be explored rapidly through computation. Just a fraction of the same exploration might take human beings working on their own something like a century to carry out. In the past, we had no choice but to spend that century. Some past year's new cutter design for machining might have drawn upon research begun many years prior. Our knowl- edge proceeded slowly in this way, even generationally, with researchers exploring process variables exclusively through experimentation and their own inference of what they saw. The early successes in AM came the same way. The new way knowledge and successes will come can be With machine learning, AM will advance faster than if it had been invented in an earlier time. As computation identifies promising connections, human experts will judge which findings are noise and which make sense. Peter Zelinski / EDITOR-IN-CHIEF glimpsed in this issue. Three feature articles this month (pages 22, 28 and 32) describe how machine learning is being used as a tool to more rapidly advance our mastery of AM. And a tool is what machine learning is, says Bryce Meredig, co-founder of Citrine Informatics, a firm applying machine learn- ing as part of the research discussed in the story on page 22. It is a different kind of tool than manufacturers are accustomed to, but one they will come to understand and even make better through their efforts. He uses the term "artificial intelligence" (AI), and to him, "artificial" captures the tool's power and its limitations. "AI is really good at taking a billion possibilities and short-listing them to ten," he says. "But humans have important pieces of un- derstanding that are too subtle to incorporate into computation." Thus, only a human can look at a set of possibilities produced by machine learning and see something like, Options one and two are not feasible, but option three is a promising way to go. Success will lead to success, he says. Progress will accelerate because mathematical relationships will be found that point to physical relationships. Rules will be found. And when a rule is dis- covered—that is, do X in selective laser melting of Inconel 718 and it reliably leads to Y—then that knowledge can be incorporated into the model, allowing subsequent machine learning to accept the rule as known and direct its power elsewhere. Humans will find these rules, he notes. That much will re- main as true as it ever was. As computation identifies promising connections between inputs and outputs, human experts will judge which findings are noise and which make sense. It is simply the speed of discovery that will change, but this change will be profound, and we will see it accelerate. As people continue to perform the crucial last step of recognizing the value in AI's find- ings, he says, that very contribution will allow the pace of those findings to increase.

Articles in this issue

Links on this page

Archives of this issue

view archives of Additive Manufacturing - JAN 2018