CMTC's Shifting Gears

We succeed because you do.

Season 3 Episode 2 - Robotics & the Small Manufacturer Part 2: How Robotics Can Increase Employee Safety & Productivity

Posted by Rachel Miller

Episode Show Notes

Episode 2 features Mario Persicone, Director of Operations at PNM Company and Owner of Pro Laser Graphics. Mario discusses the applications of robotics and automation and how they can boost productivity and make a workplace safer.

Mario Persicone is the Director of Operations at PNM Company and the Owner of Pro Laser Graphics — both located in Fresno, CA. Mario started his career in manufacturing over 33 years ago as a CNC operator and later became a CNC programmer. After several years as a programmer, Mario took on the role of product engineer and was responsible for managing the prototype shop at PNM Company. In early 2012, Mario was promoted to the Director of Operations where he manages three 24-hour shifts. One of Mario’s biggest accomplishments at PNM Company has been the implementation of their ISO program. With his training, he is now a certified lead auditor in ISO 9001 2015. Mario also serves on the Board of Directors for CMTC and as an Advisory Board Member for Fresno City College, Reedley College, Madera Community College, Duncan Polytechnical High School, and CTEC High School. Mario has helped to develop the curriculum for many of the schools’ manufacturing programs. He is also a member of the San Joaquin Valley Manufacturing Alliance and works closely with the Fresno Business Council, Fresno Regional Workforce Development Board, and Career Nexus. 


00:00:00 - Introductions

00:01:19 - Pressing operational challenges before the company began integrating robotics and automation technologies

00:02:03 - What types of technology have been implemented

00:02:19 - What processes are handled by automation and robotics

00:02:30 - How company determined which robotic and automation solutions were right

00:05:50 - Results of robot integration

00:06:25 - Benefits of automation

00:07:50 - How integration impacted employee safety

00:09:48 - Employee acceptance of new technology

00:12:05 - How company addressed programming needs to ensure that the robot is as productive as possible

00:14:25 - Whether or not grippers change based on different jobs

00:16:36 - Length of time design took

00:17:15 - Discussion about emotional side of risk factor

00:18:41 - Reliability of robots

00:20:39 - Expectations for improvement in productivity or reduction in cost

00:23:46 - Specific skills required for employees upon implementation

00:24:42 - What the company would do differently if they had it to do again

00:31:54 - Next steps in the company's automation journey


Gregg Profozich [00:00:00] In the world of manufacturing change is the only constant. How are small- and medium-sized manufacturers SMMs to keep up with new technologies, regulations and other important shifts, let alone leverage them to become leaders in their industries? Shifting Gears, a podcast from CMTC highlights leaders from the modern world of manufacturing from SMMs to consultants to industry experts. Each quarter we go deep into topics pertinent to both operating a manufacturing firm and the industry as a whole. Join us to hear about manufacturing sectors' latest trends, groundbreaking technologies and expert insights to help SMMs in California set themselves apart in this exciting modern world of innovation and change. I'm Gregg Profozich, Director of Advanced Manufacturing Technologies at CMTC. I'd like to welcome you.

Gregg Profozich [00:00:48] In this episode, I'm joined by Mario Persicone, Director of Operations at PNM Company and Owner of Pro Laser Graphics. Mario discusses the applications of robotics and automation and how they can boost productivity and make a workplace safer.

Gregg Profozich [00:01:05] Welcome, Mario. It's great to have you here today.

Mario Persicone [00:01:06] Thank you, Gregg. Thanks for having me.

Gregg Profozich [00:01:09] Mario, can you just take a minute or two and tell us a little bit about your current role?

Mario Persicone [00:01:14] Yeah. I'm the Director of Operations at PNM Company here in Fresno, California.

Gregg Profozich [00:01:19] Well, thank you so much for being here. I'm excited about our conversation. I'm really looking forward to hearing your perspectives and your insights. Let's get started. We're here to talk about robotics and automation. I really want to focus today on the applications of robotics and automation and how they can boost productivity and how they can make a workplace safer. What were some of the pressing operational challenges before you began integrating robotics and automation technologies? What were the things that you guys were dealing with?

Mario Persicone [00:01:45] Well, I think employee efficiency was one of them, just the availability of the employees, where we were having attendance issues, especially with COVID. COVID really played a big role in our automation implementation. But yeah, a lot of the inefficiencies of just an operator. 

Gregg Profozich [00:02:03] Got it. Then you guys started down a path from automation robotics. Specifically, what kind of technologies have you implemented at PNM? 

Mario Persicone [00:02:10] We implemented a FANUC M-10 Trinity Automation cell to one of our vertical mills, one of our Haas machines

Gregg Profozich [00:02:19] That includes what operations? What processes does the automation and robotics do?

Mario Persicone [00:02:23] It's really part tending, really delivering fixtures to the machine tool to machine CNC parts.

Gregg Profozich [00:02:30] Okay, so machine tending application. How did you determine which robotic and automation solutions were right for you given your production and operational challenges?

Mario Persicone [00:02:39] Well, we teamed up with, like I said, Trinity Automation. They were big in the food industry, and Bill Selway wanted us to really bring his automation company into the machining world. Our biggest factor in determining what to choose was what's going to fit best in a job shop environment. We have many different parts that we machine on a regular basis. For us, it was standardizing the automation to where we can bring it into a job shop environment.

Gregg Profozich [00:03:10] You're building other people's parts? You're building to print, essentially, right?

Mario Persicone [00:03:14] That's correct. Contract job shop.

Gregg Profozich [00:03:16] You're manufacturing in lots of 1 to N — 50, 100, 1,000, 5,000, whatever that is. Every job is different and has a different quantity. How did you have to think about if I'm going to look at a robotic solution, when is it profitable, when is it not kind of thing? Because if I'm running one part and I have to change over the robot and the machine, that's probably not as cost-effective as if I'm going to run 5,000 parts, and I can set it up and let it run for a day or two or a week.

Mario Persicone [00:03:43] Yeah. I think what we were looking for with the automation side was carrying on how our horizontals work for us. We have 7 horizontals with 96 pallets. We dedicate a lot of our equipment. We wanted to bring that same cell philosophy to the automation side. That was where us and Trinity really teamed up to bring on how we handle the horizontals and bring to the automation side. We had the experience already but just not with the robot.

Gregg Profozich [00:04:15] Okay. Just for those not familiar with the CNC industry, 96 pallets. Tell us, what is a pallet?

Mario Persicone [00:04:21] Well, a pallet is a block that we can dedicate fixtures to it, and they're stored. We have horizontal cells that have 36 of these pallets. They're either four-sided, two-sided, eight-sided. We can dedicate a lot of work right to those pallets. We eliminate all the setups, where we just type in a five-letter password and activate a job, and we're running production within 10 seconds.

Gregg Profozich [00:04:49] They're preset fixtures or jigs for a given type of run.

Mario Persicone [00:04:53] That is correct. They hold the parts that we manufacture on a daily basis. Everything's designed and engineered with quality control in mind. We can engineer a fixture that mounts to one of the pallets that holds the parts and allows us to run many parts at one time.

Gregg Profozich [00:05:12] How is that connected with the robot? The pallet allows the robot to do a prescribed set of motions or movements, pick up parts and put them down in a certain place? What does that do for you?

Mario Persicone [00:05:22] Yeah. The cell we designed for our robot was essentially... We designed 56 locations in a robotic cell that houses an 8 by 12-inch pallet, and the robot knows the position of every pallet. For us, programming the robot was more specifically to a location in a stocker shelf versus how to grab each unique individual part that's stored on those pallets.

Gregg Profozich [00:05:50] Got it. What were the results of the integration? How did it impact productivity?

Mario Persicone [00:05:55] Well, obviously, our efficiency went up. We're essentially taking 4 hours a day to load up 56 pallets of work. Then that operator is able to walk away and go run another machine while the robot tends to those 56 pallets. It runs for 20 hours. Instead of having 24 required man-hours, we reduced that down to 4 hours, where we're gaining 20 hours lights-out manufacturing.

Gregg Profozich [00:06:25] Wow. You mentioned COVID earlier with the difficulty in finding employees and with people being absent because they were exposed to various different issues. This has really been a great thing for you, it sounds like because less operators can still run multiple machines, right?

Mario Persicone [00:06:39] Absolutely. We didn't really have to train an operator how to physically operate the robot. The robot knows what to do. We just have to train the operators to change out the parts on the fixture. The automation side on the initial implementation was the programming of the robot, but after that, it's just like running a man CNC. The robot will deliver the parts. We just got to make sure the robot has parts to grab.

Gregg Profozich [00:07:05] Just like the CNC has its already preset G code for the operation it's going to do on this repeat part, the robot knows, too. It has its own set of instructions. It knows what to do and where. It's just a matter of marrying those two together correctly. That's where the operator training comes in. Is that what I'm hearing?

Mario Persicone [00:07:18] Yeah. As long as they're loading the parts correctly on the pallets, the biggest hurdle we had was dealing with chips and making sure that the pallets remained clean at all times because that affected the accuracy of what we were trying to accomplish. But we come up with some creative solutions in chip control, and that really helped out with our efforts.

Gregg Profozich [00:07:40] That's always one of the big problems in CNC machining. All the chips, all that metal that gets machined away and those little chips and shards is the problem.

Mario Persicone [00:07:49] It was a big problem.

Gregg Profozich [00:07:50] The other side of that, the results of the integration, how did the implementation integration impact employee safety?

Mario Persicone [00:07:57] It's really safe. Everything is enclosed in a cell. It's basically in its own cage. We have three zones to that cage, where an operator can access any given zone that the robot is working off of. If the robot is working off of zone one, they can open the doors and access any parts or pallet in zone one. They only have access to zones two and three, which keeps them out of the cell, keeps them safe from any movement of the robot. It's a very safe cell to operate.

Gregg Profozich [00:08:31] Do you have the yellow safety cages or are you using light curtains?

Mario Persicone [00:08:34] Yeah. It's light curtains, and it has doors with interlocks that you have to hit a button to request access. Then the software within the cell will determine if it's safe to unlock that zone.

Gregg Profozich [00:08:48] Okay. Proximity safety is ensured by the light curtains, and the doors, and the cages.

Mario Persicone [00:08:53] That's correct.

Gregg Profozich [00:08:53] Then other aspects of employee safety — repetitive motion injuries, picking up payloads, and moving things around like that. With the robot doing most of that work I'd imagine that risk is probably going down substantially?

Mario Persicone [00:09:04] Oh, big time, yeah. You're not lifting a lot of stuff. You're changing parts right onto the cell. You're not even lifting the pallets. No, the robot is essentially doing the work. We still have our challenges of occasionally having an Allen wrench slip, and you bang up a nut bolt, but it’s very minor stuff.

Gregg Profozich [00:09:24] Yeah. A whole lot less opportunity for lost time injuries, right?

Mario Persicone [00:09:28] Absolutely.

Gregg Profozich [00:09:29] As long as operators and technicians are being careful. You get the occasional accidental thing that's unavoidable almost, right?

Mario Persicone [00:09:37] Yeah. You still have to be mindful of what you're doing in your surrounding areas. But yeah, the chances of pulling a muscle in your back or smashing a finger is very, very, very minimal.

Gregg Profozich [00:09:48] Excellent. Then talk a little bit about organizational acceptance. With any new technology the operators on the floor accepting it, embracing it, if you will, is really critical. Did you have any apprehensions about bringing in robotics and automation and, if so, how did you deal with those? How did your employees accept? 

Mario Persicone [00:10:06] Well, at first, naturally, you hear, "Oh, they're bringing in automation to eliminate our jobs." That's not the case at all. Once we talked to our staff and had the conversation saying that automation is coming in to grow our business to where we can train you to operate these cells and pay you a higher wage, then the acceptance was almost immediate.

Gregg Profozich [00:10:31] Yeah. There's another podcast we did this season on robotics and the small manufacturer. Part one was more about the evolution of robotics technologies and where it was. We had a conversation there. I spoke with SK Gupta from USC and Craig Schlenoff from the NIST Laboratories. They were talking about the fact that robots have difficulty doing what a human two- and three-year-old can do without thinking about it, often. They're not androids. They're not going to take our jobs.

Mario Persicone [00:10:57] They never will.

Gregg Profozich [00:10:58] Walking up to a bin full of parts that were poured into a bin and picking out one is a very difficult thing for a robot to do. A human three-year-old could go, "Yeah, I could grab the green one or the yellow one. No problem." The placement, the position, the orientation, all those things have to be computed for the robot through some kind of sensoring, and visioning, and manipulating the gripper or the end effector the right way. All that stuff's very difficult to do for a robot. They're way behind us, if you will, in their evolution. I think that's a key message.

Mario Persicone [00:11:24] I think it's always going to require some sort of manpower to keep the automation running at optimum. Our main guy on it now, at first, he was just so against it. He didn't even want nothing to do with it. Once he’s seen how it worked, and the training that went into it, and the advancement opportunities that were available, he wanted to be a player in it again. I think once he saw all the shop tours that we do, and all the educational centers that come through here, and how the automation is such a focal point of those tours and our guests that come, he wanted to be a part of that. I think that was a big factor for him, too.

Gregg Profozich [00:12:05] You mentioned some of the things about training: how to set up the robot, how to do some programming. From an operations perspective, how do you address the programming needs to ensure that the robot is as productive as possible?

Mario Persicone [00:12:06] Well, in our environment programming an automation cell can be tremendous if every part's different. First thing we had to do is figure out the end of arm, how we can effectively pick up 56 pallets that have 112 different parts on them. For us, programming was easy. The automation company did that for us. But we still have to service; we still have to be able to manually move the robots. The programming is a small part of what we do because, like I said, all the locations are fixed in our cells. Our technicians, our operators, are learning more on how to service, how to maintain, and how to keep the robot happy than they are for programming.

Gregg Profozich [00:13:03] You've got a set of programs that is meeting the needs for all the parts you're going to run based on the cell. It's machine loading. I'm picking up from a given place; I'm putting it inside the machine; I'm closing the door and pressing the button. Well, the robot's doing that. The machine is running; the machine finishes; the robot opens the door; grabs the finished part; puts another blank in; closes the door; presses the button; puts the finished part somewhere else.

Mario Persicone [00:13:23] Yeah. The operator's biggest role of that is just making sure that the correct program number is assigned to that pallet and that location within the cell. If he's running part ABC in stocker shelf 10, he's got to make sure ABC program is assigned to that pallet. If he's going to run XYZ in stocker shelf 10, then he would make sure that that appropriate cut program is assigned. For them, it's more program management of the actual cutting program versus programming of the robot.

Gregg Profozich [00:15:57] In the same way a machinist without a robot would load a part in and make sure he calls up the correct cutting sequence program. Still have to do that. Then you have to make sure that the robot program is married up to that correctly, as well.

Mario Persicone [00:14:09] Yeah. The robot knows the stocker shelf locations; they know where the pallets are stored; it knows the ready position. It'll take it from the stocker shelf to the ready position to the machine. It knows all the key locations within the cell.

Gregg Profozich [00:14:25] Let's go into the design just a little bit more and some of the details here. If I'm a small manufacturer and I go to buy a robot, I can buy a FANUC, or an Yaskawa, or an ABB, or a Universal, or Productive Robotics, or any of those out there. I'm basically buying the arm. Then there's the end of arm tooling that's different. It's the arm, not the hand, if you will. That end of arm tooling, that end effector, that gripper — all the different terms are used for that. There's mechanical versions of those, and pneumatic versions of those, and suction cups. Do your grippers change based on different jobs?

Mario Persicone [00:14:54] No. That was the beautiful thing. The collaboration we had with Trinity was what can we do to where we don't have to change grippers. We designed a failproof dovetail system. Our end of arm tooling basically dovetails. That's how it's picking and pulling a pallet is on a dovetail system.

Gregg Profozich [00:15:17] Okay. Your technicians don't ever have to change end of arm tooling. They just literally have to change programs and make sure that the program of the CNC machine and the program of the robot match and they match the part that I'm running.

Mario Persicone [00:15:28] Then aside from that it's just load the material to the pallet. After that, the robot picks all 56 pallets up in the same fashion, same way, same dovetail. It's a pretty unique system we came up with.

Gregg Profozich [00:15:42] Wow. How long did that whole process take? It sounds like there was some design and some research that you had to do. I'm guessing you probably had to take a look at your whole portfolio of parts that you do and say, "You know what? The robot would work great on this portion of them because we could all do the same. We could use the same end effector for all of these. Therefore, we want to automate that. Things that are more particular that require changeovers add time, add cost." Is that the logic you went through?

Mario Persicone [00:16:07] Yeah. There was a lot. With Trinity not having the machining experience that we had and us not having that automation experience that they have, there was a great deal of collaboration that took place of developing a cell, what were we trying to get after. It was a good team. The guys over at Trinity really did a great job with us analyzing all those kinds of things. We did learn a few things after the fact that we would do different, but that's part of implementation.

Gregg Profozich [00:16:36] That whole process with all those complexities and all those things that you learned in that collaboration, was that a two-month process, a year-long process?

Mario Persicone [00:16:43] This was the first cell that Trinity really did. We took a while on the software side of things. They had to learn how we use the horizontals to our advantage. They had to write some software custom for us. They had to build the cage. There was a lot of trial and error. Total, I want to say it was about a six-month process. But I know now that they've done quite a few of these cells, I think that that implementation process for them is a lot quicker now.

Gregg Profozich [00:17:15] Through that process, talk about the risk factor and the emotional side of that. We're doing this. We think we can make it work. We're taking the risk, though we're wasting time if it doesn't work. What did that feel like?

Mario Persicone [00:17:25] It was scary for us at first. We were new to automation. We didn't know how it was going to be handled if we had breakdowns. The initial investment upfront wasn't bad at all. Automation has come down in price. We weren't scared in that aspect. We were scared about what if it goes down? What if the robot goes down? How do we fix it? Well, that was one of the advantages of teaming up with Trinity. They're located there in Southern California. We're a three-and-a-half-hour drive from them. Getting a technician here right away has never been an issue. That really helped us with some of our angst on implementation and getting us into building. But it all boils down to the relationship you have with your automation company that's helping you.

Gregg Profozich [00:18:09] I think that's a really important point. You've got to have a good, trusted, collaborative relationship and really build the communication, I think, because if they don't fully understand your problem, the solution they come up with won't match. Right?

Mario Persicone [00:18:21] Yeah. In a job shop environment, we're losing money if our spindles aren't turning. With every piece of machinery, we buy we team up with the companies that are going to provide the best service to us, instant service. Whether they have to come here or they're a phone call away with support for us was a big defining factor.

Gregg Profozich [00:18:41] You mentioned a minute ago the reliability aspect. I want to delve into that just a little bit more, if we can. The robots, how reliable are they? What happens if they break down? Is it outside staff or internal staff that can fix them? What does that whole thing look like?

Mario Persicone [00:18:55] Well, they run. We don't run them to the max of their capabilities, because we feel steady always wins the race. Are we running the robot at 100% of its speed capabilities? No. We run them at about 50%. The reason why is we would rather just be safe and make sure that they're taken care of. I would rather sacrifice an extra 10 seconds on a pallet change for a robot that houses 10 parts, because that 10 seconds extra spread out over 10 parts, it's a second more per part. That's not going to kill us. Taking care of it was a big deal. But breakdowns, most of the time we have issues, which are seldom, it's because of lack of maintenance, to be honest with you. For us, we handle that side of it unless it's a major PM that we're doing. Then Trinity will come in and perform that. We do that once a year. That's usually a one-day service, where they'll come in, change all the oils in the arms. They will just do their yearly PMs on it. We do our daily, and weekly, and quarterly PMs on it. It's been working out well. Our downtime literally has been very, very minor. We were so caught up. One of our issues was the machine was never turned off. Then when we finally turned it off, we didn't change out the batteries and then lost some of the parameters. That was our fault, because we never had the battery changing on the maintenance side of it. We got that up and running right away. It's been rock solid.

Gregg Profozich [00:20:30] Okay. I believe you guys are a 24/7 operation, right?

Mario Persicone [00:20:33] We are. We're 24/7.

Gregg Profozich [00:20:35] Were you 24/7 before the robots were brought in?

Mario Persicone [00:20:38] We were.

Gregg Profozich [00:20:39] If I was a manufacturer, even a 24/7 operation, if I brought in a robot, I could see a range of X percent improvement in productivity or reduction in cost? What would that look like? What should I expect?

Mario Persicone [00:20:50] On our vertical mills we operate OEE. We're looking at our overall equipment efficiency, and we're looking at our employee efficiency. Two totally different things we monitor here. We were on a manned vertical mill on an OEE. Our machines are running anywhere between 58% to 68% of the time. The other is part changing, setup, job changeover. Now our uptime on our spindle, it maintains anywhere between 85% and 91% on our robotic cell.

Gregg Profozich [00:21:31] Wow. that is quite an improvement.

Mario Persicone [00:21:33] That is an improvement. It's a huge improvement. The best part of that is only a small percentage of that is labored hours. Now I'm able to move resources to other machines. 

Gregg Profozich [00:21:44] Right. You probably also have increased flexibility. Your best machinist can work that one-off part that's the most tricky instead of having to tend to the big job that has to go out, because he can set up the robot and let the robot do the work.

Mario Persicone [00:21:56] Absolutely. It's all about having the right resources on the right jobs. The advantage of the automation and how we did it is we're able to also control and ramp up customer spikes. If we have a customer that buys 20 parts a day, 20 parts a day, 20 parts a day, then all of a sudden he runs 100, it's really easy for us to adapt to that with automation. We either add more pallets to our cell, and we just activate more. It just runs, runs, runs. We're able to accommodate spikes up or down.

Gregg Profozich [00:22:32] Then there's an aspect there, too, of consistency. I'd imagine the robots are doing exactly the same thing every time. They're just machine tending. But it's very rarely or is it happening often or occasionally that there are issues with the program and the accuracy of it? Do they drift out of phase?

Mario Persicone [00:22:49] No. What we do is we have a locating system. We're using Schunk vices, and it repeats in the Y axis very accurately. What we do is every time a pallet is delivered to the machine from the machining side, the very first thing we do is we bring up a spindle probe, and we're digitizing the X-axis, because it does have a little bit of variance of where it delivers the pallet to. We want to make sure it's perfect every time. We do have a probing operation that takes place before it machines any given pallet. It's more for security than anything else. It just really depends on the tolerances you're working with is how you need to approach it. But for us, pretty standard stuff is plus or minus 2,000ths to 5,000ths you really don't need to probe anything, but on those jobs that are 1,000th or less, just to be safe we added probing to it.

Gregg Profozich [00:23:46] Got it. Any specific skills if I adopted a robot into my operation? Any specific skills that my employees would need to acquire? Where should I go to get those skills?

Mario Persicone [00:23:56] Well, I think that's where your integration company comes into play. None of us had those skills. The integration company will train you, whoever you choose. They should train you and have a pretty good idea of what level of expertise you need to run them. But that was probably our biggest takeaway from this whole implementation was how automation scared us at first. We to this day don't understand why, because we didn't have to be PLC programmers or robotic programmers. Your integration company takes care of all that. It's no different. We're just machinists. It was that easy for us.

Gregg Profozich [00:24:35] Just using the same skills. The skills I have as a machinist can apply.

Mario Persicone [00:24:39] If you can service a CNC machine, you can service a robot.

Gregg Profozich [00:24:42] All right. Let's get into some of those if I had it to do over questions. We're talking to the small, midsize manufacturers out there who are listening. If you had to go back to day one, what are some of the things you'd consider in terms of the emotional side, the fear, the cost justification, the how do I strategize and think about this, what parts do I apply it to, how do I choose an integrator, any of those things, whatever comes to mind. Let's go through some of that.

Mario Persicone [00:25:05] I think definitely number one, the last one you said, is choosing the right integrator. That one's key. You have to team up with them. They become an extension of your company. Aside from that, then the next thing you really got to start looking at is what machine is this going to be integrated to. That was one of the takebacks we have on it. We have a great system, 56 pallets, and we put it onto a vertical machining center that only has 40 tools. We ran out of tool capacity way before we ran out of real estate on our pallets. That was a problem because now we're cutting our opportunities down. We had to really in our environment find the right job. The initial jobs we put to the cell ended up being transferred off of that cell. As we went out and quoted new jobs, we started specifically looking for jobs that would fit the cell, fit the amount of tools. There is a lot of things that we overlooked at the beginning. That was machine selection, tool capacity, and just having the right work for it. You got to know what you're after. Ideally, having the real estate we have, 56 pallets, we have the opportunity to dedicate a lot of different parts to the cell. But volume plays a big role in that. But every business is different. It depends how you run your business. Like you said, we're a contract shop. All of our parts that are on contract we're obligated. Our lead time is 48 hours. We got to have the right jobs on that cell so we can meet our deadlines of our lead time.

Gregg Profozich [00:26:44] I think I heard you say that you had to find the right jobs for the cell, almost like you invested in the capacity, and you thought it would do one thing, but it didn't quite do that. But there was a way to still make it very profitable and maybe even open up new business areas, too.

Mario Persicone [00:26:57] Yeah. The first thing we did is we had a mindset that we were going to run many different jobs. First of all, we just run 100% aluminum on it, because that minimizes the amount of tools that you got to have. Our main driving factor on this was tool-driven. Whatever we put to this machine's got to be made by the same 40 tools that are dedicated to the machining center. That was the challenge there.

Gregg Profozich [00:27:23] Has that been limiting is the question.

Mario Persicone [00:27:26] Yes, it has. We're currently looking at maybe a fifth axis that has larger tool capacity to put the system on. We also designed this system with Trinity. Going back on how long did it take to implement, we wanted to make sure we designed the cell that can basically adapt to any vertical mill or fifth axis machine. This is actually a mobile unit we can roll from machine to machine if we adapt our other machines to accommodate it.

Gregg Profozich [00:27:57] It sounds like you've come to a clarity. Robots apply in my business here. They make a lot of sense right here. I could use them for other things, but I haven't seen the business case be compelling enough yet. But I see the business case be compelling here. Machine tending.

Mario Persicone [00:28:11] Machine tending is the best area, because that's the area where we have the most inaccuracy. Robots don't call in sick, don't get COVID. Those are key areas, the human factor. If you can eliminate a lot of the human factor, that's where the automation comes into play.

Gregg Profozich [00:28:28] Yeah. They don't complain about the work they're doing. They'll do the same mundane job over and over and over again. The other side of that is mentally unchallenging jobs tend to have high turnover. We as human beings want something more out of our work than picking up a piece and moving it over there, picking up this and moving it. It's mindless. You want something more. A robot doesn't mind that. On those dirty, dull, dangerous, mundane jobs, on those jobs with repetitive motion injury risk, a robot's a really good solution. Like you said, they don't get COVID; they don't take breaks; they don't go on lunch. They'll work 24/7 and not complain.

Mario Persicone [00:29:01] That's one of the things that we do with our operator. We rotate between two or three different people, because it is the same 56 jobs right now that they are running on a daily basis. It does get old for the operator. Hey, let's transfer you off of this for a month. You go out, run other jobs. Then as we dedicate and make new pallets for the cell, maybe next month when you come back, there's some new jobs involved in it. It helps keep them fresh. It helps keep them dedicated to the cell to the point where they want to be there, not because they have to be there.

Gregg Profozich [00:29:38] It's better for morale; it's better for productivity, and it's better for focus and quality. They're more mindful when they're engaged.

Mario Persicone [00:29:47] That's it. We're looking at collaborative robots now for parts, maybe more for our turning centers. We totally envision two or three machines in a row that has a robot, to where one person is tending to the robots; they're not tending to the machine. It's more job appealing to tell someone that you work with automation than you just load material in a machine. It's a little bit more fun to advertise what your job role is.

Gregg Profozich [00:30:17] Absolutely. We covered the if I had to do it over again questions a little bit: choose the right integrator, figure out what the right machine is to automate, find the right jobs for the cell, machine selection, tool capacity, part volume, etcetera, job selection, and then the whole employee acceptance piece. I had a question about what advice you'd give, but I think I've already got that answer: choose the right integrator, and choose the right parts.

Mario Persicone [00:30:37] I think that's it. I think what the integrator will bring to the table and then you. No one knows the parts better than the company itself. An integrator is going to have problems deciding what's the best part. It's what you're trying to accomplish. Being a small manufacturer in a 48-hour lead time for the majority of our products, we have to be able to get a call right now, be able to manufacture a part, send it out for plating, and have it in our customer's hands within 48 hours. The way to do that is to eliminate setup. The way to do that is not worry about employee attendance. The way to do that is to go out there and type in a five-letter password, and you're running production in 30 seconds. You get those parts off to the platers within an hour, and they're back in your hand by lunchtime. Our main goal behind this was to minimize the amount of finished goods inventories we got to carry on our shelves, to where we become that famous just-in-time manufacturer. Give me an order at 7:00 in the morning, I'm running parts by 7:05. I have them on the truck at 8:00 in the morning to go to anodizers. I can have them back by 1:00 in the afternoon. That's a huge advantage for us.

Gregg Profozich [00:31:54] You've mentioned looking at cobots now. What's next for your organization in your robotics and automation journey? Are you considering other technologies or more robots? Where are you going?

Mario Persicone [00:32:03] We're actually looking at having a more mobile UR arm that can adapt to many of our turning centers so we can make it universal and have two or three of these part tending robots that are easily adaptable to multiple machining centers, just for parts tending only. That's really easy to accomplish on the turning side. I think with our experience we have with our current automation, the collaborative stuff, I think, is going to be a pretty easy implementation for us. It's not necessarily having to select the right jobs. It's real easy to pick a round part that goes into a chuck, the same chuck. It's a lot easier to integrate. I think what we're going to do is get a robot that's more usable for multiple machines. I think that's where our staff is going to be more involved with programming those collaborative robots versus our current robots. They're going to get a bigger taste on the automation side and how it operates more, because they're going to play a bigger role in automation than they currently do right now.

Gregg Profozich [00:33:09] You're going to mount the collaborative robot on a cart and move it around into set positions from machine to machine.

Mario Persicone [00:33:14] Yeah. We went to IMTS, the last IMTS they had in Chicago, which was — I don't know — a few years ago. We saw a pretty nice little setup that was on a cart. You're able to load your raw materials, your finished goods on the cart. You wheel it up; stick it up next to the machining center; you teach the key point locations, and you're off and running. We're really looking at that technology to bring into our company for the future.

Gregg Profozich [00:33:42] Mario, we've covered an awful lot of ground today. Is there anything else you feel it's important to cover before we wrap things up?

Mario Persicone [00:33:48] Yeah. I think it's just the fear of automation. I think for a lot of small manufacturers, it's impossible. The cost of it is so astronomical. I think integrators have paid attention and listened. I think it's not as expensive as people think. The fear is just a lack of knowledge and understanding of how to implement it. It can be very beneficial to you.

Gregg Profozich [00:34:13] I'm going to try to do some summary points on my notes here from what I gleaned out of our conversation. It looks like for your experience of the robotic cell that you guys put in has been a huge asset in a number of areas. Number one, productivity. It's allowed you to set up for and do lights-out operation on an ongoing basis. I can do a setup and let the job run for 20 hours, and nobody has to look at it again. My key people with their skills, my key machinists with their skills on setups and on key parts can focus other places. I don't have to tie up resources to do that. I can maximize my spindle time. The second thing I heard you talk about was the safety aspect. That was one of the key things we wanted to talk about today. I think we covered it pretty well. There's a reduction in lost time injuries. The robot's handling all the parts, picking up the heavy stuff, moving it around. Nobody's going to wrench their back reaching, just hurting themselves type of thing. The robot's doing all that work. You have the safety curtains and the cages that you built into your integration. It's impossible to get near the robot and get hurt with an industrial robot that's running. You built the safety in in a couple of different ways. You've also got that safety morale side in. You're not losing people because they're doing a dull and unrewarding job, doing something mundane. The robots are doing the mundane jobs, and the humans are doing the things that require a little bit of brainwork. Organizational acceptance we talked about. Initially, it's natural for people to resist change. The robots are going to take my job, all these fears that come in. Then we see the reality is that robots actually can't do everything humans can do. They're always going to need us. But they can be another tool in our tool belt, another asset for us to be able to increase our productivity and increase our wages, essentially. The more profitable a company is, the more stable the company is, the longer I've got a job, the better chance of getting the raise if the company is profitable. There is some training required, some programming but mainly for technical service and maintenance. Then really just making sure you're doing the setups correctly. You talked about the importance of having a good integrator, finding the right integrator, an integrator that's going to work with you well, that you're going to be able to collaborate with, an integrator that's going to understand your desired application and the aspects of your business that you're trying to automate and why so they can make sure it's the best fit. Then they're going to do most of the programming for you in all likelihood. You don't have to have programmers in-house; you really just have to have people... As you said, if you can run a CNC machine, you can run the robot. We talked a little bit about end of arm tooling and the creative ways that you've tried to minimize the need — and totally eliminate the need, in your case — for end of arm tooling changes. That's been one of the things that's really made this such a success for your organization. That kind of design thinking, it goes back to the collaboration with the integrator. How do I think about it; how do I make sure the integrator understands what I'm trying to do; and we minimize the complexity? Let's make this as simple as possible, as effective as possible. Design of the cell was a piece of that. What range of my products can I apply to that cell? If I've got 100 parts in my portfolio, maybe I only want the robot to be able to tend 20. If it can do machine tending on just those 20, if those 20 are profitable, it's a good investment; let's do it. Then you think about maybe there's another 10, or 15, or 20 in my portfolio. Another robot or another application might be... Can I group it that way? That design thinking of step back and figure it out from a how do I maximize the productivity and the benefit perspective. I think those are some of the key takeaways that people listening would want to have. Maintenance we talked a little bit about. Robots run. They really don't break down often. If there's issues, it's probably a lack of maintenance. Some internal discipline is going to be required to make sure you're maintaining on a regular basis. Most of the simple service it sounds like technicians and internal staff can do. The integrators can do some of the more major things and the annualized maintenance type of things. Huge improvements in productivity can be realized, both production flexibility — the ability to handle spikes in demand, the ability to quickly change from one product to another to meet short-term expedited orders potentially, etcetera, etcetera — then also just the efficiency on the mill, going from the 58% to 68% OEE on the manned vertical mill to the 85% to 91% on the robotic version of it. That increase in productivity is huge, as well. Did I miss anything in there?

Mario Persicone [00:38:15] No. You took really good notes.

Gregg Profozich [00:38:16] Well, thank you.

Mario Persicone [00:38:19] I think your summary is spot on. Those are all the things that we went through as a company and we're going to go through in the future as we adopt more. I hope this helps others. I hope it gives a little bit clearer perspective on not being so afraid of it.

Gregg Profozich [00:38:36] Mario, it was great to have you here today. Thank you so much for joining me and sharing your perspectives, your experiences, your insights with both me and with our listeners.

Mario Persicone [00:38:43] Gregg, thank you for having me. It's fun. I love talking about it. It's exciting technology. I really appreciate giving me the opportunity to talk about it.

Gregg Profozich [00:38:52] We really appreciate your being here. Thank you so much. To our listeners, thank you for joining me for this conversation with Mario Persicone and discussing how robots can increase employee safety and productivity. Thank you so much. Have a great day. Stay safe and healthy. Thank you for listening to Shifting Gears, a podcast from CMTC. If you enjoyed this episode, please share it with others and post it on your social media platforms. You can subscribe to our podcast on Apple Podcast, Spotify, or your preferred podcast directory. For more information on our topic, please visit

CMTC is a private nonprofit organization that provides technical assistance, workforce development, and consulting services to small- and medium-sized manufacturers throughout the state of California. CMTC's mission is to serve as a trusted advisor, providing solutions that increase the productivity and competitiveness of California's manufacturers. CMTC operates under a cooperative agreement for the state of California with the Hollings Manufacturing Extension Partnership Program (MEP) at the National Institutes of Standards and Technology within the Department of Commerce. For more information about CMTC please visit For more information about the MEP National Network, or to find your local MEP center visit

Topics: Robotics Increase Employee Safety & Productivity

Tell Us What You Think