How to optimize 3D Printing for aerospace applications

[#1]

According to Allied Market Research, the 3D printing market was valued at $13.2 billion in 2020 and is expected to reach $94 billion by 2030. Additive manufacturing is a rapidly growing industry with many applications in high tech industries, specifically in the aerospace industry. Given its need for complex parts, the aerospace industry can benefit greatly from the design freedom that 3D printing allows.

How can you use 3D printing if you have no experience? There are myriad settings and variables at play, but it can be difficult to understand exactly how each variable affects things such as print time or quality. In the video below, we explore answers to this question. Using design of experiments (DOE) and group orthogonal supersaturated designs (GOSSD), we can discover the effects that these variables have on a product.

[anne_milley]

@brycerjs , this is brilliant!  Whole new appreciation for the complexity in achieving efficient and high-quality 3-D printing results.  Thank you for sharing!

[Vins]

@brycerjs, excellent presentation, always great to see real world applications.

what are your thoughts on the use of Space filling designs vs GoSSDs for computer experiments where you have lots of factors and the computer experiments are not expensive? Would you learn more from a GoSSD or a large space filling design our is it better to start with one and refine proposed design points with the other? I have 12 factors and a single run is ~2mins . 
‘thanks,

vins

[nick_shelton]

@Vins, As you mentioned space filling designs tend to be better suited for computer experiments. The print time results can be obtained through simulation very quickly so a space filling design could have been an option if print time was the only response under experimentation. Since @brycerjs  experiment involved optimizing both print time and print quality a DOE methodology needed to be selected which would allow for the most efficient collection of simulated and physical data. Given the large number of factors and time/resources required to obtain print quality data GOSSD was the best method to determine optimal settings for the two responses under study.

Fundamentally, "pure" computer experimentation is best done with space filling designs. The link below references some advantages:

https://www.jmp.com/support/help/en/17.2/index.shtml#page/jmp/overview-of-spacefilling-designs.shtml