Allan: Today we’re sitting down with Matt Menard and Michael Pursely from Summit Manufacturing Systems, as well as Steve Phillips, our founding father here at Philips Precision. We’re talking about the important technical details of an effective CMM program, as well as effective fixturing, and using the inspection process as a way to more efficiently control your production process, rather than just a way to check for compliance. Matt, would you like to introduce yourself?

Matt: My name is Matt Menard. I’m one of the founding members of Summit Manufacturing Systems. We’ve been in business now for about 3 years and this is year 4. What we really focus on is metrology, specifically when it comes to CMMs, and getting that technology used correctly throughout the industry.

Allan: Perfect. Michael, how are you doing today?

Michael: I’m doing great. I am Michael Pursely, Matt’s partner, one of the three partners at Summit. Like Matt said, we are focusing on metrology. Making metrology work for manufacturing is what we do. It runs the gamut from process, to CMMs, to SPC, to automation and all in between.

Allan: Excellent, that’s awesome. How are you doing today Steve?

Steve: I’m doing well. My name is Steve Phillips, the co-owner and president of Philips Precision.

Allan: What are you guys seeing as far as the current state of the industry as of right now?

Michael: Well, I think in the metrology labs you’re starting to see the environment change. What Matt works on a lot is putting inspection equipment at the point of manufacture. I think in the future you’ll actually see the large-scale lab start to go away, as that moves out to the shop floor as technology changes. I think fixturing plays a large role in that, especially the modular fixturing. It’s usually one type-part or one-part family going into these cells. You’re not running several different part-types in these types of cells, so having that modular fixturing is very advantageous, through simply being able to put one plate up, get it all setup, and store it as the part runs.

Allan: Yeah, things are heading more towards in-process inspection out on the shop floor.

Michael: They’re also lending stuff to automation too. You can get that organized, and Matt can speak to the fact that having an on-set fixturing system that you can simply load and unload on a plate is very advantageous to automation.

Allan: Absolutely, it definitely goes hand in hand. When you’re engineering for metrology and things like that on the machines – when you guys are coming up with your fixturing solutions, what is required of a CMM fixture?

Michael: The basic, fundamental, first most important thing is that it has to be static. It has to hold stationary in the same place. It can’t move around and can’t be adjusted. That’s the first criteria you must meet.

Matt: What we typically see a lot of times is: you get the same guys designing your CMM fixturing that are designing your milling fixturing. When they do that, it’s so overkill and expensive for what it needs to be. A lot of times a fixturing weighs five times with the part ways, or more, it’s crazy. If you can keep it as simple as possible with three points of contact, maybe a clamp to hold it down, that’s all you need to do. It reduces the cost of the fixturing, as long as you can still get to everything on the part, the CMM’s not physically going to move it, and the CMM applies very little force to the part. It doesn’t have to be much to hold the part in place and be able to inspect it.

Allan: Right.

Steve: That’s definitely a huge point. We had one customer that called us, and they had a thrust washer that they were making, and they had a pallet full of them that were rejected because they weren’t flat. They knew they were flat, but they couldn’t prove they were flat. What happened was, they were holding him with a chuck, a lathe chuck, on the table of the CMM. This made it so that when they tightened it, it basically crowned upward and went out of flat. That’s how we came up with the Spider- Clamp™. We very lightly held it in place and potted magnets, three points of contact, and we sent them one. We actually had one that we were experimenting with, so we sent it down to them to try. The very next day that pallet of junk parts was now good, so we added that to our product line.

Matt: Another common misconception that we see with fixturing – besides it having to be this big, bulky, massive thing – is that it needs to be the exact same every time. When I say that I mean that people have the misconception that, “Oh, I just have to be within a tenth when I put my part back on there,” and it doesn’t have to be. That’s the job of the CMM and the CMM program. If you’re within around ten-thousandths you can make a seek envelope large enough on your machine to find that part. Once you find that part, you’re leveling and aligning to the individual part, not the fixture; the fixture is just there to get you close.

Steve: Exactly.

Michael: That goes back to taking full advantage of the software you’re working with. CMMs are meant to do exactly that; they’re not meant to have you put the part in the exact same spot. They’re meant to go find the parts, multiple-part pallets – palletizing goes along with that. Utilizing the CMM to its full capabilities is something that we don’t see often. 

Allan: Right, that makes a lot of sense. I never actually thought about that myself, where the machine is actually there to do that precise work for you so that you don’t have to. That makes sense. That’s why they’re so expensive, it all makes sense now!

Steve: That’s a really common question that we get. We will have potential customers asking about the plates and they’ll say, “Well how accurate are they?” and I’ll have to go through the spiel of exactly what you just said. One time what I did, just to prove a point as I took a 6×6 Lock-N-Load™ plate, was: I put a credit card underneath the corner of it and literally just tipped it up to throw it off kilter. We put a ring gauge on it, measured it, got a number, and took the credit card out, did it again and got the exact same number. As long as the part doesn’t move and the CMM can come in and find it, you’re going to get a good number. As long as the parts not moving and the fixture’s secure.

Michael: And the programs programmed properly.

Steve: Exactly.

Michael: Yes, because in that case, there’s actually a way I could do that and change that number, depending on how you align to that part, and what you are using to align it. This goes back to the fixturing. There’s a fixturing error that goes into your MSA, which is your Measurement System Analysis or the amount of fixture error, and of course you want to account for that. This goes back to holding against static and not moving, which is really where your error is going to come from: A part moving within the fixtures, where the only error you can have is the fixture.

Allan: One thing I’ve noticed that Steve actually mentioned, when we’ve done some trade shows – and I’ve definitely seen it and picked up on it – is that, depending on where you are, you can always tell when you’re talking to the people that work in metrology and inspection and when you’re talking to folks that work in machining. It’s kind of strange to me, as I’m fairly new to this industry, where the paths don’t cross very much.

Michael: Oh, you’re being gentle. They don’t cross, they collide; they crash right into each other. 

Allan: Yeah, they crash.

Michael: It is a common problem. One of the things that we are proudest of is being able to foster a relationship between inspection and manufacturing, because a lot of times those two are not working together.

Steve: Right.

Michael: And it depends on how the role of inspection is being played at each facility. Are you checking to accept? Are you trying to control a process? Those are where you see it deviate, especially in groups when they are not getting good feedback to manufacturing. They can simply become pass/fail, move-part people. You need to be able to give feedback and information to the manufacturing floor. Then those two start to work together and depend on each other, which is what you’re looking for in a facility.

Allan: Yeah. We would go to a trade show like Eastec, for example, where everybody’s there, and most of the people come over and they’re saying, “Well, how flat are these plates?” Then we answer them with how flat they are, but we then have to follow up with, “In the end, it really doesn’t matter.” But when we were at Quality Show, nobody asked that question one time because they knew that it was a moot point.

Michael: That’s a good point. We very rarely run into a situation where management or upper level management really has an understanding of how to evaluate a CMM programmer, or what to expect from a CMM. It’s more of a situation where they hire a programmer and have to trust them.

Steve: Yeah.

Allan: What are some things that companies should consider when talking to and hiring a CMM programmer?

Michael: It’s tough. The environment is really tough right now. In a job market where, in general, you’ve got unemployment at a very low rate, along with the fact that we haven’t had a good history in the last 30 years of training CMM programmers. There hasn’t really been a trade or a college course that you can really get. There’s a few out there, but they’re rare, and they deal more with academics than how to relate metrology to manufacturing, which is what you actually need to have done. So, to find someone is very difficult. We’ve been promoting the concept of contract programming, as a means to alleviate that effectively and be cost-effective. When you think about it, you really only have a CMM programmer when you’re writing a new program, a first article, or when something new comes out. Other than that, you don’t need a programmer, you only need an operator, right? Most shops aren’t rolling out more than four or five first articles a year, so you really only need a programmer at those few times out of the year. The rest of the time you only need an operator, so why put someone on staff for a whole year when you can contract that out?

Matt: And that’s for the high production environments, where you’ve got a job shop that changes a little bit, but there are still ways to go around that. I guess what we would say when people are looking for somebody to be a CMM programmer is, if you have a high level inspector that is very good on a plate and understands how to lay out the part effectively on a plate, you can transition that into doing it on a CMM. Because essentially, you’re doing the same thing; you’re still aligning to a datum structure, you’re still making sure it’s stable, and you should still be doing a MSA on that to make sure that your program is correct.

Allan: Got it.

Michael: What’s a good point is that – what we see a lot of, me and Matt – is very few facilities actually incorporate MSA or GRR into their standard process. The reason for that is that it is time-consuming and expensive. It takes a lot of time, it has a machine up for quite a bit, but it’s worth doing if you’re going to take the information and use it. If you’re going to take it and use it to improve what you’re doing, it’s worth the time spent.

Allan: For those who don’t know, can you explain what Gage R&R is?

Michael: Matt, I’ll let you get that one.

Matt: With a measurement system analysis, you’re basically looking to find out what variation in your part is tied up by your machining process and by how you’re actually inspecting it. So, you want to try to structure it so that you’ve got a large variety of parts. So, when you’re trying to do your study, you want to make sure you’ve got parts from first shift and second shift, and that they are a wide enough range to cover the entire tolerance that you’re looking for.

Then what you do is, you have one operator measure one part 3 times, the second operator measure that same part 3 times, and the third operator do the same thing across 5 to 10 pieces, depending on the size of the parts, the time of the inspection, and how much money you want to spend on the study. A typical full Gage R&R would be three operators with 10 parts, three times each. Sometimes you don’t have all those parts to do that and you kind of have to mix and match. You can do this with traditional hard gaging and you can do this with a CMM. What we typically see when we do it with the CMM, is one piece of that Gage R&R – the R&R stands for reproducibility and repeatability –  the reproducibility piece goes away if the fixturing is correct because all the operator has to do is put it on the fixture and press the button. There shouldn’t be any operator interaction with the CMM itself because all they’re doing is hitting the “Go” button.

Whereas, if you’re using hand tools, an individual operator could squeeze the mic a little bit more, they could ratchet it a little bit tighter, and that’s going to change their value and you’re going to see an offset on the measurements of those parts from operator one to operator two to operator three. The CMM gets rid of that and you should see a consistent value across the board. Then what you’re doing is only looking at the repeatability of the program, which goes back to your alignment: if anything’s moving on the fixturing while you’re touching it and some other evaluation methods and the metrology behind your program.

Allan: Excellent.

Matt: Did that answer it?

Allan: That was great, yeah. Absolutely.

Michael: Well it’s important because in manufacturing I think accuracy is overrated because I can take one measurement system and that measurement system can say an inch is a foot. If I use that same measurement system throughout my process and it repeats, it doesn’t matter what the accuracy is, right? Repeatability is far more important in manufacturing than accuracy.

Steve: Yeah, one of the biggest problems that we have or – I shouldn’t say problems – hurdles that we have surrounds some of the custom fixtures we do. So, we produce a lot of custom fixtures and we’ll do 32 up 64 up fixtures and for these little, tiny, delicate plastic parts. We do a lot for plastic injection molded parts. And so when we do these fixtures, we have to make them so that they’ll pass a Gage R&R. In these companies that are paying us to make these fixtures, the first thing they do is a Gage R&R and it could take a week to do some of these fixtures, it’s very costly to do. It’s very important for us to get those fixtures just right. So, we have to go through each station and check them with a force gage and make sure that we try to get as much force as we can without deforming the parts. It is a really big challenge.

Michael: Absolutely.

Matt: Well, that’s something that we see with fixturing too. We have a lot of companies that would do something like a Lang Vise with a custom cut soft jaw. The problem is if you’re putting a part on there that can deform and you’re not using some kind of torque wrench on it, there’s where you see your operator-to-operator variation: operator one barely tightens it, operator two is a big gorilla out there and cranks the thing. They all do it differently.

Steve: Right. Yeah, and that’s where our Rapid-Loc™ comes in handy because you can make custom jaws. We’ve done jaws that go way beyond the platform of the vise. What we do is, we hold it very strategically and as the inspector you can dial your pressure down to almost nothing. Most of our pressure we’re running less than 5 PSI or somewhere in that range. That way, as an inspector, you can set it up and then hand it off to pretty much anybody and have them load and unload, and now it can be used as a production tool.

Michael: That’s the key to using a CMM. To get the advantage out of a CMM palletizing is the way to go and auto run. It allows you to run the parts overnight like you talked about before, the ability to set several up at once is an obvious advantage.

Allan: I’ve seen companies that have giant CMMs and with the modular system, they’ll run rails down both sides of it or multiple rails on it and then they’ll just lock their fixtures in where they need to. They can have like five or six parts set up at one time and then run them in a string. Set it up and then just leave and then the machine goes and moves – like first, second, third, and all the way around and then that’s like first article inspection for every part.

Michael: At a former company I worked at, we made small little lids that went over a laser crystal. They were gold-plated – nickel striking gold plated, so very touchy – and I would lay out a hundred of these on a modular plate just using standoffs and the pins. I’d do a hundred and I would set it up to go as I left at the end of the shift. I would set it up, let it run, come back in the morning and collect my data.

Allan: Yeah, that’s perfect. That’s awesome. So you guys are doing automation with robotics and stuff too, have you guys had any automated…?

Michael: Oh, yeah. I’ll let Matt cover this; Matt does a lot of work with automation and robotics.

Matt: Well, as Michael was saying earlier, what we see is kind of a shift from a traditional, “Okay, you make your part to bring them into this CMM and in a couple of hours later, you bring another part in,” and you’re constantly waiting for that data. So, you’re constantly running at risk. What we’re trying to promote, and what we see the industry going to, is that inspection at the point of use at the machine, whether you add robotics to it or not to try to improve your efficiencies, that’s kind of the next step that we try to take customers to but the first thing is data at the point of use. That way the operator is not waiting for hours at the CMM to see if he made the right adjustments or not, where he’s constantly going back and forth and tweaking his part. If he’s got the data at the point of use and we’re trending that on and SVC  chart, he can watch it as it goes.

What we’ve seen is he then makes fewer adjustments throughout the day, which means he’s actually making more parts. By having that inspection right at that point of use, we see an increased productivity without even having to do anything just by having the data there. The next phase of that is being able to tie that data right back into the machine tool so we can remove another potential source of error. I mean we’ve all done it, whether it’s as a CMM programmer or a CNC programmer, you go to make an offset – heck, even if you’re writing a report – you type something in wrong and now all of a sudden, you’re making the next part incorrectly. So, the CMM is going to produce a set of data for it, it’s going to go back to the operator, and the operator is going to say, “Hey, alright, I need to adjust X, Y, and Z,” and I’m going to come back to that point in just a second. So they type in their offsets, and hopefully they put the correction going the correct way. If not, they now have to inspect that next piece that comes off of it, make sure that they made the correct adjustment to the tool, and keep going.

What we can do from virtually any measurement system that’s out there is we can actually tie that back into the controller, using our friends at Karen Engineering and their auto-comp software. We can take that data set out as a text file and then we can relate that back to their cutting process in their tools to remove that operator from having to manually input those offsets and we can just keep running. Now if we add a robot to that system, the operator virtually doesn’t need to be there. The only thing I need the operator for is to change a tool, maybe if there’s a stop in the program for some chip rap if you’re cutting some kind of stringing material like an Inconel and possibly loading bar stock if it’s a bar fed machine, otherwise the robot’s going to pick up and load the new parts onto the milling fixtures and let it go lights out.

So like what Michael said, on the CMM, palletizing that and loading it up so you can run it overnight. We can now start doing the same thing with the CNC machines as well. The point I wanted to quickly get back to was making sure that the operator actually does the right inputs and offsets to the tool. What we’ve seen from multiple facilities is how the part is being held on the CNC is not the same way that the part is being held in the CMM. So, we tell them that they have to adjust their x-axis, that’s what the report says on the CMM, but in reality it’s their y-axis or z-axis that needs to be tweaked on the machine. This creates a ton of confusion and wasted time on the shop floor. So that’s something that we specialize in when we work at these facilities, and help out and do the contract programming, is being able to relate that CMM output directly back for the CNC. Whether we need to do some kind of little matrix that says, “Okay, X is your Y, and this is this,” once we provide that data to them it makes it a much more streamlined process.

Allan: Definitely.

Michael: Yes.

Steve: Yeah, typically on some of the more complex parts, when we do the inspection report from Calypso, we’ll make sure we put some sort of a visual symbol on there as a legend. We keep it very basic and it really does help because sometimes, just throwing a number at them just isn’t always the best way to do it.

Michael: You can see the advantage of taking that piece out of that process and letting it be handled by the machine. Of course, all of this flows right into your quality code and quality requirements, the contractual part; the code on the SPC goes along with this and all this stuff that you’re looking for from your process happens automatically because of this. Matt makes a good point too. We have a customer that Matt and Nathan have done a lot of work with that we have a case study on, and the numbers are surprising. You can’t imagine how much money you’re losing on the production floor. You’re chalking it up to the cost of doing business while you actually simply just lost profit. It’s not the cost of doing business, there’s efficiencies to be gained there – more than what most people can imagine. It’s kind of what launched Summit to be honest.

Allan: Got you, put you guys on the map!

Matt: It was basically putting an Equator out at the machine. So, we had one in process measurement device for 3 to 4 mill-turns in one facility at a company in New Hampshire. What we noticed happened was the productivity numbers went up, the scrap rate went down with the machines that were utilizing this. Then one month, it actually went backwards, and they stopped using this process and we were able to very quickly identify that those production numbers actually dropped for those machines that stopped using this in process, and the scrap rate went back up again. It was a very apples-to-apples comparison across the same milling machines, the same inspection equipment, same operators across all shifts. At the end of the day, it ended up being something around almost a thousand more of this particular part being produced a month per machine.

Steve: There’s a lot of money there.

Michael: I think 15 million dollars for one year.

Steve: That’s money well spent.

Allan: How much? 15 million in one year?

Michael: A million dollars per machine.

Allan: That’s amazing.

Michael: What they found is that it wasn’t just a matter of, (correct me if I’m wrong) but, it’s a matter of not stopping the machine nearly as often. It was probably the biggest increase that they saw.

Steve: Yeah, spindle down time is huge.

Michael: Exactly, and because you weren’t having a measurement that said, “Okay, adjust me here, adjust me here, adjust me here,” which is what you saw on the histograms; you would see how they’re adjusting straight lines. A majority of the time you shouldn’t be having to worry about more than a handful of dimensions of a part. Read the print just from a process standpoint, “What gives me trouble in the process?” It’s usually only tracking four or five dimensions, everything else should be pretty static. Having the SBC collecting that data tells you that you’re able to look back and make corrections far more effectively than if you’re looking at one part at a time and trying to react.

Matt: If you’re still running a traditional CMM environment (where you’re taking a part, you bring it in, and you’re checking every single feature on it) that’s overkill and it’s going to create a bottleneck. Start identifying inside your process what features you need to check/which ones might be out of control. You can even base it around: “Okay, I got a bolt hole pattern. I got 12 bolt holes on it. I’m going to check the first one and the last one. If those two check good, I know my tooling was correct, everything else in there is going to be fine. I’m not too worried about that,” or, “If it’s the same tool that’s cutting five features, I’m going to check the tightest feature that’s on there, and the last one cut. Now, I know that my process is still in control, I don’t have anything out, and I’ve minimized my inspection time.”

Michael: I’ve always said a 100% inspection is a symptom of a failed process.

Steve: Yeah, that makes perfect sense. I go into a lot of different shops and I talk to a lot of different people doing trade shows and things like that. I am amazed at how I can go into a lot of inspection departments that have a beautiful CMM and just a blank granite plate. I think to myself, “What are you doing?” We’ve got twenty-some-odd CNC’s in the shop, cranking out parts left and right and fixtures to hold these parts. Then you go on to the CMM and it’s like you have nothing. So why would you treat a $100,000+ machine any differently than you would a CNC milling center or any other machine out in the shop. Fixture it up and make it so you can quickly and efficiently swap from one job to another, so that you’re getting the most output out of that machine.

Michael: That’s a great point. That is a common task that we have to accomplish: highlighting that in a facility. Like you said, if it’s not highlighted you walk and anybody can see this, but they’ll be unsure of how it’s there. They’ll walk in and see that management is not pushing it along. It leads them to question, “How did you invest this much money in this equipment, it’s not being utilized.” Shining a spotlight on that and bringing attention to that is the challenge. That’s what creates our opportunity.

Allan: I think that’s a big challenge overall. As a solutions company ourselves here at Phillips, we’re trying to push the industry into understanding that there are more efficient processes and that there’s a lot more to the technology that they think they’re getting into. Like Steve said, they buy this machine that’s hundreds of thousands of dollars and they have a bare slab of granite on it. Then they end up writing it off and the corporate says, “Oh, that was a waste of money. That was a poor investment, we shouldn’t have done that.” You can’t go back but, it gets looked at as a poor investment. But it’s only poor because they didn’t utilize it properly. They didn’t go to the next step or tool it up. They didn’t program it right and they didn’t use it right. I think that’s a big learning curve that the industry is still in the midst of.

Michael: I think that’s a good point. I think you see that, not just with CMMs and inspection, but with technology in general. If you go to a trade show, like Eastec or IMTS, see all the technology and go to the shops, it’s evident they’re not able to utilize it. Technology is so far ahead of where manufacturing is actually at today. They’re a long way from catching it just going out today, not to mention what’s coming tomorrow. It almost moves so fast, that upper management is not able to keep up. The guys at the top don’t understand CMMs and Auto Comp. They don’t understand the different tools that they could be using. When they get them, they don’t have the individual technical skill it takes to make them useful. It’s hard to find.

Steve: We went without a CMM for a long time. When we finally made that leap to get a CMM in the shop, it made a huge difference for us. We were able to get it in the inspection department and the first thing we did was fixture it up. It’s where we learned how to do this stuff, but we had buyers coming in the shop and they’d say, “You have a CMM? Wow! Well, that’s all right. Let me see if we can get some more prints for you.” It’s like having a sales force working for you because you have that machine, and you know how to use it. You can actually check their carts and prove that they’re correct.

Allan: It’s amazing to me, coming from outside of the manufacturing industry where I do understand a lot of the newer technology. I do sales and marketing, so I’m not quite on the manufacturing aspect of it for the long-term background and everything, but even people that aren’t using things like modern email capability to communicate and companies that still prefer to send a fax like it’s 1985, amazes me. 

Michael: And we’re all talking about upper-level companies. We’re not even mentioning the job shops; the small and middle-sized job shops that really struggle with that stuff.

Allan: Yes, exactly. Actually, I think that the small and middle-sized job shops utilize the newer technology more so, because technology is advancing so quickly and so rapidly. As technology advances, the cost comes down, so it becomes more attainable for smaller and middle-sized job shops to be able to acquire newer technologies. This is actually one of the things I speak to students and other people occasionally about the industry. They don’t like the idea of robotics because they think that they’re taking jobs and everything, but the idea is it’s really creating an atmosphere where you have technology that is affordable for more companies to attain and to implement. This makes it so you don’t have one giant company with all the money paying all the people; you can have a bunch of companies with less people paying everybody higher wages because it spreads the wealth more.

Michael: It goes back to finding the people.

Matt: As far as robots replacing people: I’ve been in this industry now for 15 years, I spent 10 years of that with General Electric pretty much as a M.E. process improvement leader lean six sigma black belt. My job was to improve efficiencies, save money for the company, and we basically calculated that on, “Okay, here’s all the improvements that we’ve made. Here’s the people that we can get rid of.” Out of that time at GE, we never got rid of a single person. We always found something else for them to do; there was always more work out there.

Steve: I think that the robots are taking jobs. They’re taking the crappy jobs that nobody wants to do and they’re creating really amazing jobs. Now you need inspectors to inspect the robots when they’re built, you need people to design them, you need people to repair them, you need people to fixture them and make grippers for them, etc. There’s so much that needs to be done.

Michael: …. people that were doing that low-level job and getting them prepared to do that job. That’s the big hurdle.

Steve: Right, those people need to be trained, but into a much better job.

Michael: Oh yeah. The opportunity is there if we could just get the training and education.

Allan: I think the bottom level person that’s doing those repetitive tasks that the robots are taking over are actually the perfect people to go into the job for the programming aspect of that. They know exactly what is supposed to happen because they do it with their hands and now, they just have to learn how to be able to make this robot do it. The technology with robots is getting so incredibly amazing that it’s so simple to be able to get a robot and program it with a simple task. I think the people that are in those jobs say, “Oh, well, you got to be really smart to go and do that type of stuff,” and yes, they’re right, but I think everybody is smart enough, as long as you put some effort into it.

Steve: And a lot of them are teaching and repeat now, too, so you can just teach the robot by hand.

Allan: Yeah, grab onto and move it around. Absolutely. 

Michael: Like Matt described earlier, when you describe that sale and how it worked with feeding the system, that job becomes more of a monitoring type of job. You’re monitoring the process and teaching that person how to read an SPC and what it’s doing; that job changes a little bit. You stop having a traditional type of role and get more of a cell monitor type role. That’s kind of the way it goes. It does lower a little bit there with the technical skill needed. You’re more monitoring and reacting. It’s a little more process-driven. We’re construction driven, so it’s a little more defined.

Allan: Yeah. It’s no longer writing pages and pages of code to get a robot to do something. The programming aspect of it has become so easy. The user interface has become very similar to any other user interface you would use on a laptop or an iPad.

Michael: And the job becomes less tribal knowledge that way. From a shop owner standpoint, if I can get somebody in procedure and workers say that can pass from one individual to the next, it’s a huge benefit; I’m not tied to one person that knows that process.

Allan: Right. They’re predicting that over the next 20 to 30 years, over 30,000 new jobs that haven’t been invented yet will become available. So, there’s jobs out there that we can’t even think of yet, that we know are coming because the needs are going to be there. I made a joke that we have these cell phones, and you go to the mall, you have the kiosk with your cell phone accessories, but pretty soon we’re going to have a robotics accessory kiosk. We’re going to go be like, “Oh, I’m going to go get some new grippers from my robot because I don’t like the way that, that one grips my wine glasses when they load and unload the dishwasher.” I appreciate you guys coming on today. This has been fantastic, and I learned a lot. I’m around this stuff all the time and I learned a ton today, this was amazing. Thanks Steve. Thanks for coming in and sitting in on this.

Michael: We would be very disappointed if you said you learn anything.

Allan: Yeah, definitely.  Matt, thank you so much for coming in. Michael, thank you very much.

Matt: Not a problem, thank you guys for having us.