• Benjamin Moses

AMT Tech Trends: Unorthodox Tradition

Updated: Apr 15

Release date: 13 February 2020


Ben and Steve don’t own space, NASA does! Ben and Steve marvel over Christina Koch’s record breaking stint… IN SPACE! Steve got the cobot moving and has some plans for the newest addition to the testbed. Ben talks about little RFID tags to track tooling, workholding, workpieces and other manufacturing stuff. Steve mentions a dude making brass trumpets in a new way. Ben keeps it real with Honeywell before he busts some AI myths.


IMTS Insider www.imts.com/show/newsletter/insider/index.cfm

Benjamin’s Linked In www.linkedin.com/in/benjamin-moses-b13b44a2/

Amateur Machinist Blog swarfysteve.blogspot.com/

Music provided by www.freestockmusic.com



Transcript:

Benjamin Moses:

Hello, everybody, and welcome to the Tech Trends podcast where we discuss the latest manufacturing research and news. I am Benjamin Moses, the director of Manufacturing Technology and I’m here with…


Stephen LaMarca:

Stephen LaMarca, Manufacturing Technology analyst.


Benjamin Moses:

Hey, Steve, how’s it going? Welcome to the second episode of 2020.


Stephen LaMarca:

Thank you, sir. It’s going great.


Benjamin Moses:

Yeah.


Stephen LaMarca:

I think we’re having an awesome year this year. Best wishes to China and all that they’re going through right now.


Benjamin Moses:

That’s a lot of problems there.


Stephen LaMarca:

That is a lot of issues. It’s actually affected the test bed a little bit, but we’ll get into that later.


Benjamin Moses:

I just want to mention that this episode is sponsored by the MT360 Conference. It’s a great conference that we discuss the latest and transformative technology for manufacturing. We have a great series of speakers and some dynamic personalities on stage to talk about use cases and new technologies entering manufacturing. And also it’s a great exhibition space. We’re converting it to a virtual factory. We’ll be able to meet and greet everyone. So, check out MT360conference.com and hopefully, we’ll see you there.


Benjamin Moses:

Before I get into the meat of it, I just want to give a shout out to one of our space NASA astronauts, Christina Koch, her efforts. So, she’s a NASA astronaut that spent 326 consecutive days on this station.


Stephen LaMarca:

Wow.


Benjamin Moses:

Which is a long time.


Stephen LaMarca:

Yeah.


Benjamin Moses:

Just under a year.


Stephen LaMarca:

Her tour included 5,248 orbits.


Benjamin Moses:

That’s a lot of sunrises.


Stephen LaMarca:

Yeah, that is. That’s actually fascinating, now that you think about it, because it’s like 328 days but 5,000-plus orbits. It’s like wow, they’re on a different level. Quite literally.


Benjamin Moses:

She traversed 139 million miles. That’s a lot of frequent flyer miles.


Stephen LaMarca:

She better be getting first-class for life.


Benjamin Moses:

Also, a cool thing that she mentioned is, she spent quite a bit of time outside the station. So, she completed six spacewalks.


Stephen LaMarca:

No way.


Benjamin Moses:

It’s really cool. And she was one of the first, she is the first. They did an all-female spacewalk with a group of other astronauts up there. So, that was really fascinating.


Stephen LaMarca:

That is awesome.


Benjamin Moses:

So, I just wanted to give a shout out to her that her efforts, that the article mentions her contributions will be used for a long duration space flight to the moon. Of course, I think that’s the next thing to start for NASA to look into but also, everyone’s looking at Mars, so really, really long space travel. So, and also, I mean, along the way, they’re looking at manufacturing in space-


Stephen LaMarca:

Manufacturing in space, I was just about to say that. Why focus on the planets then? We’ve pretty much got a space station up there.


Benjamin Moses:

Yeah, that’s right.


Stephen LaMarca:

Why not make a test bed on the ISS?


Benjamin Moses:

I think there is, I think they have-


Stephen LaMarca:

Yeah, I’m sure they do.


Benjamin Moses:

A plastic 3D printer already up there.


Stephen LaMarca:

Oh, wow.


Benjamin Moses:

And I remember about three or four years ago, people were talking about the 3D printing on the surface of Mars. So, if you look up construction 3D printing, that concept was an idea of someone, some university that wanted to do in-situation manufacturing.


Stephen LaMarca:

Right. I think I remember that.


Benjamin Moses:

So, scooping up all the material around you, turning that in aggregate, and being able to print it. And that’s one of the steps. But there’s a lot of interesting work on 3D printing in construction.


Stephen LaMarca:

You said six spacewalks while she was up there.


Benjamin Moses:

Yeah.


Stephen LaMarca:

328 days. And you’ve only been outside six times.


Benjamin Moses:

See, can you imagine that the clothes you’re wearing protects you from utter death?


Stephen LaMarca:

Absolutely. Yeah. I mean, well, when you mention it like that, it’s like wow, I can’t believe she went outside six times. I think somebody… When I was an undergrad studying physics, my professor told me… Okay, so, before we were allowed to start quantum mechanics and really get into a lot of our radiation labs where we’d be messing with Caesium-137 and stuff like that, we had to take a short course on the dangers of radiation poisoning.


Benjamin Moses:

Oh, yeah.


Stephen LaMarca:

And that was really brutal.


Benjamin Moses:

Yep.


Stephen LaMarca:

Like watching… Last year, HBO had the series on Chernobyl, and you saw all of these miserable ways to go out from the radiation poisoning and stuff like that. Yeah, we learned all of that in undergrad. It’s frightening stuff. And I asked my professor, “Is radiation poisoning the worst way to go out?” He pauses, and he thinks a little bit. And he’s like, “It’s probably tied for being in space without a spacesuit.”


Benjamin Moses:

Oh, yeah.


Stephen LaMarca:

And I was like, “Okay, explain that.” Then I kind of ate my words a little bit but he’s like, “So, on one side of you, you have the sun.”


Benjamin Moses:

Right.


Stephen LaMarca:

And because there’s no buffer between you and the sun’s heat, you’re absorbing all that heat. So, one side of you is going to start boiling, while the other side of you that’s not exposed to the sun is going to freeze. And so, you’re experiencing both freezing to death and burning to death at the same time.


Benjamin Moses:

All in one shot.


Stephen LaMarca:

And that’s…


Benjamin Moses:

That’s terrible.


Stephen LaMarca:

Okay. All right. Now, I’m eating my words again. 328 days. Man.


Benjamin Moses:

I wouldn’t go outside either.


Stephen LaMarca:

That’s impressive for having to go outside six times.


Benjamin Moses:

That’s awesome. So, let’s transition to the test bed. What do we got going on this couple of weeks of the test beds?


Stephen LaMarca:

So, the cobot is installed, it is mounted to the workbench, and we have it moving.


Benjamin Moses:

Nice. What difficulties did you run into with your installation or software-wise? Since you got it moving.


Stephen LaMarca:

It’s been really easy. I got to say, compared to the Pocket NC, everything’s been a cakewalk.


Benjamin Moses:

Wow.


Stephen LaMarca:

Now, the Pocket NC’s not bolted down to the test bed. So, I think the hardest thing so far with the robot arm was getting it bolted down onto the workbench.


Benjamin Moses:

Right. That’s also because we wanted the robot arm to reach everything.


Stephen LaMarca:

Yeah.


Benjamin Moses:

So, some of it is the planning, yeah.


Stephen LaMarca:

We want it to reach everything. We want it to be stable.


Benjamin Moses:

Right.


Stephen LaMarca:

And it’s got to be mounted to that metal where it’s seven-gauge steel metal workbench, so it doesn’t flex too much. There’s a little bit of play when you’re at the outer limits of its reach.


Benjamin Moses:

Sure.


Stephen LaMarca:

But it’s been a cakewalk in terms of, the Pocket NC really is an industrial piece of

manufacturing technology.


Benjamin Moses:

Right.


Stephen LaMarca:

There’s no getting started guide. There’s no hand-holding. I mean, yeah, they have it on the Pocket NC website. That is a serious piece of kit. As for this cobot, it has been, I can’t emphasize, it’s been a cakewalk, super easy to get into. There’s been a few times where I’ve almost crashed it into itself, and then it stops and it’s like, “I know, I’m self-aware.”


Benjamin Moses:

Yeah.


Stephen LaMarca:

The cobot’s totally self-aware of itself and knows when one joint will collide with another joint, because it’s a seven-joint arm and it’s just been incredible. Now, it doesn’t know where other stuff is. And if you’ve got the CAD models, which we do, you can put that in. This cobot actually does have a digital twin.


Benjamin Moses:

Okay.


Stephen LaMarca:

So, while you’re operating the arm, you can look on your computer screen and watch the arm move.


Benjamin Moses:

That’s cool.


Stephen LaMarca:

And another cool thing is, when it’s about to crash, a big warning pops up on the computer screen and it says, “This is about to crash. Would you like to simulate a fix?” with the digital twin. Or “would you like to just go right back to controlling it and get it out of whatever risky business it’s in, whatever tight spot it’s in?” And I haven’t tried the simulation yet just because I’m not doing anything super dangerous yet.


Benjamin Moses:

Right.


Stephen LaMarca:

But yeah, and I think you can add in all of your other 3D models of stuff that you have. So, I’ve got to add that next.


Benjamin Moses:

Nice.


Stephen LaMarca:

Once I upload the model I have of our test bed.


Benjamin Moses:

Yep.


Stephen LaMarca:

It theoretically should know where everything is. It should know where the bench is.


Benjamin Moses:

Yeah.


Stephen LaMarca:

And it should know where the Pocket NC is.


Benjamin Moses:

That’s awesome.


Stephen LaMarca:

Haven’t programmed anything else in there yet.


Benjamin Moses:

Excited to see that.


Stephen LaMarca:

But all of that stuff you make me slave over in the past, it might actually be useful.


Benjamin Moses:

It’ll come together.


Stephen LaMarca:

Of course, that stuff’s useful, but it might actually be something that we can use on the software for the robot. So, the only major hiccup was, and it’s not even a major hiccup, the only thing that I had to be careful with was measuring and marking all the spots on the workbench to drill the holes, to bolt it down, which was easy.


Stephen LaMarca:

I had Sean and Jesse from IT help with that. And when it came to actually bolting it down, I was on the underside of the table, threading the nuts onto the bolts with the walk washers and whatnot. And because we had just drilled that steel table, there was a lot of burrs and pretty sharp little metal ships on the underside of the table. So, I had to be careful, wear eye protection, wear gloves to not cut my hands up trying to thread this stuff, just trying to bolt it down. And that’s been the most dangerous, the hardest part of the robot so far. Other than that, it’s really been a breath of fresh air.


Benjamin Moses:

And we still are waiting on the gripper.


Stephen LaMarca:

Waiting on the gripper. So, earlier, before we even got into our banter, I mentioned China and they’re going through the whole coronavirus thing. And I hope that stuff gets cleared up quickly, but probably not quickly, but it will… Hopefully, it gets cleared up soon enough. Reached out to Zoe, who’s been my primary point of contact at Ufactory. This is the nice lady that I was pestering for 14 months straight on “where the hell is my robot?” But I checked in with her. I was like, “I hope everything’s all right. Wanted to let you know, the robot’s awesome. Thank you guys for making such a killer product. This is some of the things we’re doing with it.” Then, just updating her, hoping she’s doing, her and her team are doing well. And then also like, “Where’s my gripper?”


Benjamin Moses:

Just squeeze in that last request.


Stephen LaMarca:

Yeah. So, they’re like, “Oh, we’re so sorry. We’ll try to get it out to you as soon as possible. But nobody’s allowed back at the office right now.”


Benjamin Moses:

Yeah, they’ve completely shut down a lot of industries.


Stephen LaMarca:

Yeah. A lot of this stuff has been totally shut down. And plus, I had to wait so long to reach out to them because of Chinese New Year.


Benjamin Moses:

Oh, yeah.


Stephen LaMarca:

Wasn’t able to reach out to them until February 6.


Benjamin Moses:

Yep.


Stephen LaMarca:

So, there was that. But yeah, “we’ll try to get it out to you by March.” And so, that’s when I think the gripper might be here.


Benjamin Moses:

Yeah. Good.


Stephen LaMarca:

I’m not going to pester them that hard like I did with the initial robot arm, the arm shipment, just because-


Benjamin Moses:

Always the left-


Stephen LaMarca:

They got something, they got some stuff going down right now.


Benjamin Moses:

They have a few problems. But we do have stuff that’ll keep us busy up until-


Stephen LaMarca:

Oh, absolutely.


Benjamin Moses:

So, you’re going to walk us through the next series of projects that we got lined up.


Stephen LaMarca:

I mean, before I even get into… Yesterday, I just started fiddling with the collision detection and the teach sensitivity. I still have no idea how to teach the robot anything yet, which is one of the follow-up projects. And the next step is, I need to learn how to teach it things.


Benjamin Moses:

Yep.


Stephen LaMarca:

Once I figured out how to teach it things, I want to automate some moves for the arm. And once I have it automating some motions, I want to endurance-test those, the automation.


Benjamin Moses:

Yep.


Stephen LaMarca:

So, I want to get it to go through some motions over a half-hour, and then I want to scale it up to over lunch, which is an hour break. And then I want to scale it up to overnight. While we’re all away from the office, see if it didn’t crash at night. Then, if that goes well, we’ll endurance-test some automation over the weekend. And then all the way to the end of while I’m away on travel. Or when I’m just at my desk and there’s people touring our floor, and I want people to safely be able to walk up to it while it’s doing something. It is a cobot after all. It’s supposed to be able to detect collision and not hurt anybody. When Steve Lesnewich in membership is walking a new member around the office, introducing them, when they go by the test bed, I want them, “Oh, and this is our test bed. And look at the robot doing things, this shiny thing, messing with our five-axis mill.” No, he’ll probably say it’s a 3D printer. No, he knows better. I got to give him credit.


Benjamin Moses:

He does. That’s good.


Stephen LaMarca:

But yeah, that’s really what I got going on.


Benjamin Moses:

Yeah, I’m excited.


Stephen LaMarca:

Let me endurance-test and then automating some moves and teaching it things. Yeah. Cool.


Benjamin Moses:

that’s the next steps.


Benjamin Moses:

So, we want to get in some articles. The first one I got, so kind of a light one, but it’s pretty useful. So, RFID tags for tracking tools real-time, so.


Stephen LaMarca:

Awesome.


Benjamin Moses:

It’s a little widget, little guy, that’s only 6 millimeters by 2 millimeters by 2.3 millimeters.


Stephen LaMarca:

Wow.


Benjamin Moses:

Really really small tag that can be used for attaching onto metallic objects. So, the idea would be, you can place it on tooling, on any work-holding equipment, anything semi conceivable. And then you can read the tags. I mean, it’s got 96 bytes of primary storage with another 64 bits of user memory. So, you can hold a fair amount of data. And I’m not sure the distance that you need to be able to read from this, but it is an ultra-high-frequency band range.


Stephen LaMarca:

Wow.


Benjamin Moses:

So, she will have read it pretty far. So.


Stephen LaMarca:

And 6 millimeters by 2 millimeters by something else, whatever you say.


Benjamin Moses:

That’s nothing.


Stephen LaMarca:

That is nothing. I mean, even with the Pocket NC, with that work envelope and how small some of the parts are on that, that would fit on the vice.


Benjamin Moses:

Yeah, absolutely.


Stephen LaMarca:

That would fit in the little tool holders. Everything could be marked or RFID-tagged with that.


Benjamin Moses:

One of the use cases I’ve seen recently is the automation of checking parts in and out of your tooling crib. So, assigning it to a user, assigning it to a machine, instead of having to manually put information on a sheet of paper or a spreadsheet using the RFID tags. And then tracking it once it gets into a cell or a machine, and then knows it’s at a machine. So, that was really interesting, I thought, how useful, how small the RFID tag is. And we’ll probably see that gets smaller and smaller in the next couple of years.


Stephen LaMarca:

Nice.


Benjamin Moses:

What article you’ve got, Steve?


Stephen LaMarca:

Actually, before I get into that, remind me, another thing that I keep forgetting to do on the test bed.


Benjamin Moses:

Sure.


Stephen LaMarca:

We’ve got the new robot. One of the first things I did with the Pocket NC was mark each axis.


Benjamin Moses:

Yeah. Yeah.


Stephen LaMarca:

You reminded me to mark each joint on the robot, because you can control the joints individually.


Benjamin Moses:

Right.


Stephen LaMarca:

I got to get the label maker from IT and mark the joints. I keep forgetting to do that. I got to stop putting that off. It will be so much easier.


Benjamin Moses:

After we did it on the Pocket NC, it became a way of life. It’s so much easier trying to manually move the machine with the labels.


Stephen LaMarca:

It really is. But okay. The article that I have is, I’m going off the rails here from what I typically pull, an article from a Tech Trends that I find really interesting. Today for this podcast, this is a couple of things that are different than normal. Number one, this article isn’t published yet. Number two, it’s not from Tech Trends. It’s from Kathy Webster of exhibitions. She writes a lot of articles for IMTS Insider. And she was out talking to Big Kaiser, a great member of ours, shout out to Chris Kaiser, that she was talking with them. And a question came up and she asked them, “Hey, what are some really interesting customers that you can disclose to me?” Obviously, government contractors don’t want to be disclosed. They don’t want to talk about what they do and stuff. But they were like, “You want to know a really cool customer of ours? A company called…” Let me get it… I think it’s Harrelson. Harrelson Trumpets. This guy-


Benjamin Moses:

It’s not a nickname.


Stephen LaMarca:

Yeah. Harrelson Trumpets, this company. This guy makes trumpets, a brass musical instrument. But with totally nontraditional, unorthodox means.


Benjamin Moses:

Cool.


Stephen LaMarca:

So, a brass, the trumpet has been around since the beginning of time. It’s like one of the, I mean, maybe not one of the first musical instrument, but it’s been around for a long time. It’s old.


Benjamin Moses:

How It’s Made had an episode on it.


Stephen LaMarca:

Yeah. Yeah. And actually, now I want to watch that episode because I’m sure it is totally different than how this guy makes it.


Benjamin Moses:

They do a lot of forming and spinning.


Stephen LaMarca:

Yeah. I would think maybe the most advanced they get is mandrel-bent tubing. But they didn’t have mandrels bending brass tube back in the day, when the trumpet was invented. I have no idea how they were made in the old-timey days. But this guy is using a lot of CNC machines, CNC lathes, and CNC mills and even a metal 3D printer here and there. And he’s just trying this. He said something like, “I know I can make a better trumpet using modern technology and science.”


Benjamin Moses:

That’s cool.


Stephen LaMarca:

And it’s just so wild because I’m thinking, “I did that.” I mean, I didn’t do that exactly but I milled brass. People told me I couldn’t mill brass and I figured it out eventually. And this guy’s doing that. He’s milling brass. And he said, Big Kaiser said he’s bought a handful of spindles from them, a 20,000 RPM spindle and two 50,000 RPM air spindles. And I was like, “That’s exactly what you need.” The Pocket NC spindle tops out at 10,000 RPM and it’s simply not fast enough to cut brass. You really need to be moving to cut brass. It’s a soft metal but you really need to be chunking through it, to cut it cleanly, and high RPM, high torque. So, it was just really cool hearing all of that, being able to relate to it.


Benjamin Moses:

That’s cool.


Stephen LaMarca:

And I want to be able to talk to him and see the company. Hopefully, we can set up a shop tour somehow.


Benjamin Moses:

Yeah, that’d be awesome, to see that factory.


Stephen LaMarca:

But it’s just really, it’s just… And I really like it when there is a traditional product out there, something that’s been around forever, but somebody tries to make it differently. I’m sorry, reinventing the wheel is cool to me. Like when people… I hate it when engineering students, real engineers don’t say this, engineering students say something like, “Oh, you don’t want to reinvent the wheel.” It’s like, “You absolutely want to reinvent the wheel.” Wheels are terribly inefficient in terms of aerodynamics. The wheel’s the worst thing ever. We need something different than a wheel… Well, I’m getting out of hand.


Stephen LaMarca:

But like Rolex, Rolex became such a huge company because they took traditional watchmaking and were like, “Yeah, we’re not going to do that anymore. Our high-end Swiss timepieces, mechanical watches are made on a fully-automated assembly line.” They don’t see hairspring setters or people that do black or white polishing. It’s all done by robots and machines, man.


Benjamin Moses:

Changing the paradigm.


Stephen LaMarca:

And that’s what this guy is trying to do, and I think it’s awesome.


Benjamin Moses:

That is cool.


Stephen LaMarca:

He could be the Rolex of trumpets.


Benjamin Moses:

And it is, I do like the mixed factories where you have combinations of manufacturing processes, so not just a company that does subtractive so they don’t have just lines and lines of mills. They’ve got some forming, they got some assembly.


Stephen LaMarca:

Yes, yes.


Benjamin Moses:

So, seeing all-encompassing manufacturing facilities, kind of a rare thing nowadays. And it’s really cool to see.


Stephen LaMarca:

Yeah, a lot of shops like specialized, right?


Benjamin Moses:

Yeah, yeah, exactly. That’s awesome. That’s really cool. Hopefully, Kathy will give us some contact info. The next article I want to get into is additive manufacturing, so further acceptance. So, we’ve got an article from 3Dprint.com and they talk about how Honeywell is working with VELO3D. They purchased a bunch of Sapphire 3D printers to specifically attack the airspace market, so they are making parts that can be someday used in airplanes. And a quote from Dr. Soeren Wiener. “We intend to qualify this equipment through repeatability testing in our production environment, including build and post-processing, to generate an acceptable set of material property data and qualification of flight hardware.”


Stephen LaMarca:

Wow.


Benjamin Moses:

I think that’s one thing that’s not talked about quite a bit in overall acceptances. The material property data on where… When I was designing parts a bunch of years ago, all the mathematical information or material characterization of material was from mill handbooks and data that’s been tested for a long time. So, not only do yield and ultimate testing of several pieces, so you have a good lot size, but they also do fatigue testing. They do stress creep testing. They do testing at temperature, so you don’t have to guess on what those properties are. They’ve done that testing for you.


Stephen LaMarca:

Wow.


Benjamin Moses:

And it’s good to see that they’re taking a very similar approach on additive. Well, of course, it’s going to take time. It increases your confidence. And that’s the biggest thing that they talk about in the material handbooks is, “What’s your confidence interval?” So now, they’re able to boost that. Then being able to break through what they perceive the performance will be. It will be super valuable to them.


Stephen LaMarca:

So, when you say material handbooks, you mean the machinist handbook?


Benjamin Moses:

No, no, no. This is more high-speed than that. No, these are books that characterize just materials.


Stephen LaMarca:

Okay.


Benjamin Moses:

So, if you look at Inconel 625 properties, there’s 200-some pages on its fatigue strength at temperature, at room temperature, at-


Stephen LaMarca:

Just Inconel 625?


Benjamin Moses:

Yeah.


Stephen LaMarca:

Wow.


Benjamin Moses:

And it gives some characteristics on machine mobility, formability, some narrative on that. Then it gives charts and graphs of all these different properties at different temperatures, different situations.


Stephen LaMarca:

That’s cool.


Benjamin Moses:

It was a mill handbook that got converted to a different publishing company now, so. Yeah, I thought that was really interesting that Honeywell… Of course, they’re a big conglomerate, so now they have air engines, but they’re also a part supplier. So, attacking it from a bottom-up, from that perspective, I think it’s really useful for the air space market. So, real cool.


Stephen LaMarca:

I find it interesting, we hear about big companies like Honeywell a lot. And of course, like Siemens with their medical equipment and their machine tool controllers. But for the average person, when you think Honeywell, it’s like, “Oh, man, that’s the cheapest window fan you can get at Walmart.” Or “oh, yeah, like oh, you have a Nest automated thermostat like Amazon or Google thermostat” or whatever. I’ve got a Honeywell.


Benjamin Moses:

Yeah. I’ve got a-


Stephen LaMarca:

The little dial thing?


Benjamin Moses:

Yeah.


Stephen LaMarca:

That’s a Honeywell.


Benjamin Moses:

The little five-day programmable thermostat.


Stephen LaMarca:

Just on the wall over there, we’ve got a… All around our office building, we have Siemens thermostats. I wonder how they feel about Nest coming in. It’s like, “Who’s this? You don’t make military equipment.”


Benjamin Moses:

That’s a weird thing about these large multi-billion-dollar conglomerates. They do such a variety of products, right? Like GE. GE has power generation turbine engines, right?


Stephen LaMarca:

Right.


Benjamin Moses:

But at the same time, they also make the little tray that I put underneath my washing machine. So, if it’s Christmas lights, I guess that’s neat. They make washing machines and the little tray does protect the floor from getting wet.


Stephen LaMarca:

Yeah.


Benjamin Moses:

It’s weird.


Stephen LaMarca:

At least it’s nice to know that a company like Honeywell is actually still a big company doing stuff. Because every now and then, you see a cheap TV that says RCA. That’s not RCA.


Benjamin Moses:

Oh, yeah.


Stephen LaMarca:

Somebody bought that brand name but, you know.


Benjamin Moses:

Honeywell keeps it real.


Stephen LaMarca:

Honeywell keeping it real.


Benjamin Moses:

The last article I want to talk about is from Inside Big Data, not the best website, but it’s fine. It talks about busting five common myths business leaders get wrong about AI. The first one is, “AI will take over our jobs.”


Stephen LaMarca:

That’s not true, right? We know.


Benjamin Moses:

The article talks about how AI is used to embody and empower the workforce and how, it’s just, there’s certain things that can’t be done by humans, handling larger and larger amounts of data, and categorize that data, just can’t be done by humans. This is… It’ll take years and years and years to process a quarter of that data, so.


Stephen LaMarca:

Right.


Benjamin Moses:

Now, rather than saying it’s going to replace humans, it actually re-emphasize the value of humans. The second one is, that I want to talk about is, “AI works like the human brain.” Nope, that’s definitely not true. They talk about the complexity of the human brain and all the variables that go into it, versus the mathematical processing of how artificial intelligence works. It’s all based on math.


Stephen LaMarca:

Right.


Benjamin Moses:

Which is repeatable and implement it through software. The last one before they get to some of the marketing side of the article is that, “AI is objective.” That gets a little hairy because, the way AI tools are built up is, there’s some level of bias automatically built into it because humans are teaching AI tools how to work. Once you’re teaching it something, you’re automatically implying some level of bias. So, whether or not it’s categorizing cats, dogs-


Stephen LaMarca:

That’s a good point.


Benjamin Moses:

AIs will never be completely objective because it’s based on some level of learning set that has bias built into it.


Stephen LaMarca:

Right.


Benjamin Moses:

So, I think that’s an important tool to think about too.


Benjamin Moses:

The last two things is, AI is difficult to implement. It’s seeing a very quick transition of being this ambiguous nebulous thing versus widgets and doodads that can be implemented very, very quickly. So, we have our software development team, and I’m working with Kalesh on implementing different tools to categorize research papers. A lot of the AI tools, more specifically machine-learning tools, are basically a software command line. So, you put in the data, you call in this command line, and you get the data back, the results back. So, it’s fairly straightforward for implementing it. It’s the data scrubbing required to get to a proper dataset to process, that’s the difficult part. And we’ve seen that a number of times in talking with Nina, our own data scientist that, actually running the command lines, running the mathematical tools, that’s 10% of the time. 75, 80% of it is scrubbing the data so it’s clean enough to actually run the process, run the mathematical tools.


Benjamin Moses:

And the last one they talk about is, “Businesses don’t need an AI strategy.” I think that’s pretty solid. That’s a fairly succinct advice in that, if you’re doing anything with data, you’re going to need some type of machine learning or AI strategy in that. How do you make your business better? How do you work faster? How do you work more intelligently? And in some cases, you’re not… And that was weird, I don’t know what that beep was.


Stephen LaMarca:

Why is there a phone in the studio?


Benjamin Moses:

I don’t know who put that phone in there. But in the end, you’re going to be doing something with artificial intelligence where they’re not just categorizing data or manipulating information to get forecasting information, repeated actions. It’s best to have a strategy, even if the strategy says, “We’re not going to do anything.” That in itself is a useful strategy to be aware about, so.


Stephen LaMarca:

Yes.


Benjamin Moses:

It’s useful.


Stephen LaMarca:

That is a good strategy because you’ve got… Too many times they’re like… Man, you don’t want to hear Doug, could go on about somebody saying like, “Oh, I want all of this data.” And it’s like, “Okay, what specific data points from your factory do you want?” “All of them.” It’s like, “Okay, well, what are you going to do with the data?” “I don’t know.”


Benjamin Moses:

Yeah, that’s a really big problem is they want to acquire the data but not solve a business need. I always say, the article kind of talked about that a little bit more is that, what’s the problem that you want to solve? There’s still… We had a meeting a couple of months ago with a variety of groups and we talked about one implementation of just dashboarding, determining whether or not your machine’s on and off.


Stephen LaMarca:

Nice.


Benjamin Moses:

You said that simple dataset is going to provide a process of improvement opportunities for the next couple of years for that factory. Why would I want to go above and beyond that when I… The simple fact is, what’s my machine utilization? Is it on, is it off? And how long was it on? How long was it running for? How do I increase that?


Stephen LaMarca:

Are you talking about what you heard from the joint technology summit?


Benjamin Moses:

Yeah. Yeah, in January.


Stephen LaMarca:

I love that. “We implement MTConnect.” “What do you use it for?” And it’s like, “We just use it to see if our spindle’s on or off.”


Benjamin Moses:

Yeah. I mean-


Stephen LaMarca:

And it’s still something. Then you’re doing it right.


Benjamin Moses:

I mean, in the end, you’re making more money because of that. That’s the key, right?


Stephen LaMarca:

Does it have so much more potential? Absolutely. Do you need it? No.


Benjamin Moses:

Right. You can’t just implement technology for new technology. Got to make money off of it.


Stephen LaMarca:

That’s right, man.


Benjamin Moses:

So, where can I find more information on these articles, Steve?


Stephen LaMarca:

Well, for my article that does not yet exist, I’m going to give them a link to the Harrelson Trumpet company, which is WhyHarrelson.com. And in the meantime, if you want to check out other awesome articles written by our Kathy Webster over in exhibitions, check out IMTSInsider.org?


Benjamin Moses:

.com.


Stephen LaMarca:

.com?


Benjamin Moses:

We’ll put a link.


Stephen LaMarca:

We’ll put a link. And then you can check out my test bed at… Or you can Google Adventures of an Amateur Machinist, or you can go to swarfySteve, all one word .blogspot.com.


Benjamin Moses:

And you can find me on LinkedIn for my latest news.


Stephen LaMarca:

Nice.


Benjamin Moses:

I think that’s it, Steve.


Stephen LaMarca:

Clutch move, Ben.


Benjamin Moses:

Bye, everybody.


Stephen LaMarca:

Bye.