Trevor: One of the core competencies at Zeiss is that we want to be well connected with our customers. They're job number one when it comes to us handling the industry as a whole. So, recently we did a very large study with a huge population of our customers to try to get more information from them about what matters. What trends are they seeing? What are the challenges they're facing? And how can we as a quality company help them meet those demands?

A pretty heavy contingent of that was in the aerospace industry, which happens to be one of my focuses at Zeiss. I lead a team of people focused on our aerospace customers. So, there's a really healthy amount of valuable information in that insights report that we've been able to learn a lot from and go back to our R&D team and make suggestions for on ways that we think we can improve our processes and in turn help their processes in the field.

Michelle: Of course, the other big buzzer people are talking about is AI. So, with artificial intelligence entering manufacturing now, what specific role is AI playing in improving aerospace?

Trevor: That's a good question. There's a number of different ways that we're adding to it right now. Some of them are more throughput based where we're finding ways to generate results faster. Some of that is through image reconstruction in non-destructive testing, specifically our X-ray microscope, our Versa. We can improve scan times by interpreting data as it's coming in using artificial intelligence, where it's able to blend together the different slices that it's taking faster to give you a much faster scan time and a faster time to results.

There's other techniques that we use it for when it comes to pattern recognition or image analysis, both in our microscopes as well as our CT x-ray systems.

On the CT x-ray side, we have a defect detection algorithm that we are able to train based off of known defects to where it can recognize those in the sample as it's finding it. So, they're automated. It's not something that somebody has to sit there and comb through a bunch of images as it's taking these x-ray slices and say, okay, that looks like a void that shouldn't be there. It's finding it in real time without the operator having to intervene at all.