Enabling Automation Podcast: Episode Four
We’re excited to bring you our first-ever podcast series, Enabling Automation. This monthly podcast series will bring together industry leaders from across ATS Automation to discuss the latest industry trends, new innovations and more!
In the fourth episode, host Simon Drexler is joined by two special guests, Mike Healy and Klaus Werner to discuss What you can automate.
What we discuss:
- Why you must know your process before automating
- Is your product designed for automation
- The importance of having a roadmap
Host: Simon Drexler, ATS Corporation (ATS Products Group)
Simon has been in the automation industry for approximately 15 years in a variety of roles, ranging from application engineering to business leadership, as well as serving several different industries and phases of the automation lifecycle.
Guest: Mike Healy, ATS Corporation (Life Sciences)
Mike Healy is the VP of Innovation and Applications Engineering for the Life Sciences Division at ATS. He has been with ATS for his entire career, starting as an Engineer right out of school 22 years ago.
Guest: Klaus Werner, ATS Corporation (Industrial Automation)
Klaus Werner is the Innovation Manager for the Industrial Automation group at ATS. He joined ATS in 2022 to help with their innovation journey.
——Full Transcript of Enabling Automation: Episode 4——
SD: Welcome everyone to Enabling Automation. In our monthly podcast series we join industry leaders from across the ATS group of companies to share helpful insights into automation and discuss the latest trends in innovation. In today’s episode, we’re talking about what you can automate. My name is Simon Drexler. I’ll be your host for today. I’ve been a part of the automation industry for about 15 years and my current role as the General Manager of the products group inside of ATS Automation.
SD: We’re really fortunate today to be joined by two guests for our topic of What Can You Automate and two of the primary drivers of innovation inside of ATS as a whole.
SD: Mike, can you tell the listeners a little bit about yourself?
MH: I’d love to, Simon. Thank you very much. My name is Michael Healy, I’m the VP of Innovation and Applications Engineering for our Life Sciences Division here at ATS. I’ve been fortunate to work for ATS right out of school as an engineer and have worked up to my current position, so 22 years with the company.
SD: And we also have Klaus here today with us for our discussion.
KW: Thanks, Simon. I’m Klaus Werner. I am the Innovation Manager for the Industrial Automation group here at ATS. I’ve been there for about seven months now, as we start our innovation journey at Industrial Automation and looking for exciting things ahead.
SD: And I think even the mention of exciting is kind of why I am excited about today’s episode. You know, innovation is such a huge driver in automation as a whole right now. We’re starting to see lots of changes, new technologies, new approaches coming into our industry. And so let’s get started into our conversation.
SD: You know, broad question, you know, what can be automated today, guys? And Mike, maybe you can go first and lead us off.
MH: S As an engineer, it’s a detailed question and I like to dive pretty deep, pretty quick. But I to answer in a very general terms, mostly anything can be automated. I think the question that we need to look at is can it be automated with the proper ROI and cost effectively? We’ve had many projects over the years that very, very difficult products to manage and handle. But the amount of engineering, the amount of effort to do it has just hindered that progress. I think as we ask the question, I think there’s a lot of details we can dive into. And I’m sure, Klaus, you’ll have some comments. But it really comes down to me about how well do you know your process, how repeatable of a process do you have? And those two questions really drive how you can automate.
SD: So how well you know your process and how repeatable it is our primary drivers to that that ROI.
MH: A lot of our customers come to us and they’ll have something very exciting. Hey, we want to automate this we’re doing this on a bench in a lab and we’ll go and look and doing something once is very different than doing it a million times. And a lot of the journey we take with our customers is that exact journey of how do we take you from 1 to 1000000 and achieve the consistency that you want to see. And really, when that when that consistency isn’t there and maybe it’s the product design, it’s the
process to put the product together. That’s when things fall apart and that’s where you learn, you know, maybe this isn’t the right path or maybe we have to introduce different ways of ramping up production.
SD: So that’s an interesting thread about, you know, something we haven’t talked about yet in this podcast is very much the difference between low volume production or making one versus high volume production and automating a million. So Klaus, you know, under the whole umbrella of what can be automated, you know, your background, have you seen lots of differences and discrepancies inprocess and repeatability around the change or shift or scale from 1 to 1000000?
KW: I think that’s a really interesting point on that scaling question and I like what Michael said, which was around how well do you know your process and how well do you know your product? Because I think a lot of people asking the question of what can you automate might be coming from that manual, very hands on manufacturing space and you don’t know the tribal knowledge in people’s heads. It can look like a very simple process at the at the beginning. And when you go to look at that question of can I automate this, the answer, as Mike said, typically is always going to be, yes, you can, but how complex is that process? And are there other things that you really need to look at in product design, in process, in scalability to get from that one 0 to 1000000.
SD: And I think a lot of our listeners are in that world, and especially with the driving trends across, you know, the global economic conditions that more people are looking into automation than ever before, we’re applying automation to things that we never have before and so Klaus, would you have guidance for those people that are moving from their initial manual process into evaluating what can be automated?
KW: I think the first step, and I think a lot of the people that we’re talking about in that space are some of the kind of interesting people in our world right now too, are the startups, the people who are trying to be disruptors, and they’re going from a team who’s very highly skilled and very technically minded. I like Mike’s comment earlier about being a very engineering focused, wanting to get in and deep dive really soon and a lot of these people do. I think the key part to start is, is what are what are the important parts of your process when you’re going in there? Is there there’s things that are going to be critical to getting it right and getting it wrong. Do you understand that well enough? And then start with the simple steps that you can pull out of that. Can we start automating a part of this process and get those learnings under your belt as you go? Or working with a partner who can help you learn that process, one way to the other. Really map it out and go this. This is what makes sense. This is where the ROI is now. And as you evolve, the ROI for the rest might come along with it.
MH: I totally agree with that. I think I always think back to a story I had a customer here and, and we were doing a sales presentation and you know, you’re trying to win the job and customer turning to me during the meeting says, well, what’s the number one risk with this program it’s a very complex part. And I turned and said, you as the customer, you need to know your process well. You need to learn it. You need to design your product for manufacturing. You need to look at these types of things. And if we can nail that down, then I think we can have a successful program. Maybe a career limiting move at the time but, but it went okay, we won the project. So, but I think that that other idea of design for automation I think is important early on. So Klaus, you talked a lot about, you know, understanding and being involved, but thinking about a lot of those start ups and a lot of those when you got into something new and exciting, everyone’s excited. Everyone wants to go. Everyone’s focused on the technology and not focused on the end game of making money as much. Right. And producing these a million times. I think bringing in early to the process, someone who can look at your product and say, in five years from now we want to make 5 million of these a year. What do we have to make changes? Because let’s do them now. We get customers come to us too often and say, we’re ready to ramp. Let’s go. And the first thing we tell them is you’ve got to change your product. It just doesn’t work and you’re adding complexity to the automation that doesn’t exist. So I think having that early view point, you know, whether it’s pre automation studies or whatever it may be, is absolutely critical.
KW: And I agree with you that just to jump back in on that point, that design for automation, I agree, is something everybody doesn’t really dive into and think about until it’s there. You think, well, I can put this together, I must be able to automate that. And the part that also people lose and whether it’s a first time automation opportunity, a startup or even somebody who’s been in this longer in a maybe a simpler operation, where there’s more hands on is, I think the cost of that automation for the complexity that they make is the part that surprises them and then maybe even scares them away. And not because that’s actually what it’s going to cost. But if they’re not willing to change, it can drive up those dollars as well.
SD: And so, like I said, I think a lot of our listeners are in that space. And so one of the strategies is really to engage on that design for manufacturing, design for automation early on in the process. Would you recommend engaging with an automation provider early in the process to do that, or is that a resource that these companies should be more consciously looking to bring into their team early on in their growth cycle?
MH: So I can personally, I think it could be either or. I do think there is an advantage with an automation company in that they’re seeing the latest and greatest. They’re up on the trends. They’re seeing they’re seeing technologies across industries. So we have a lot of applications where someone will bring a new product to us and we’re able to reach out, not just in let’s say, because I come from Life Science, but reach out to our Klaus and his team in the IA group and say, have you guys done something like this before? And they may be able to look at automation they’ve applied in a completely different type of application and help us. So I think there’s an advantage to coming to an automation house that sees a lot more, but absolutely critically important to look atsmall things like the way that you tolerance parts, where you put chamfer, like it’s in the details that can add millions to an automated solution at the end.
KW: I’ll jump on to the end of that, too, which is the I agree part of that partnership with an automation supplier is fantastic for can you automate it. But I think the other the other great part about having an internal resource for some of these people who are starting is maybe not necessarily having to have that depth of knowledge, but having what my history is like to phrase a smart buyer in that organization, someone who can interact with that automation supplier to really dive down deep into the product details, really understand what the feedback they’re getting back and be able to be the translator to the rest of that excitement in the organization because there needs to be a two way translation. A lot of organizations are very focused on that end product. And I think one of the interesting parts there is automation is a black box to a lot of these people where I’ve got a part, of course, you can put robots to put this together, but it’s never that simple. And that that individual who can bridge that, whether it’s in the organization or whether an automation supplier has a really strong person who can help, that translation part is really important in can I automate this.
SD: So I really like where this is going and you know, perhaps, maybe my own career limiting move, Mike, you know, dealing with some customers early stage in their automation journey who have potentially maybe raised venture capital and are really investing in scale. I often tell them they don’t know what they don’t know. And Klaus, what you just touched on is having an internal resource to really be the funnel of information, start to engage with experts and start to ask the right questions and then translate information back to the organization, because you can absolutely get caught in the details of automating a process and it can end up costing significantly more. Where if you engaged early in the design for manufacture of the part as an example, you can avoid those additional expenses by adding a chanfer- for adding a parallel surface for the robot to grip or something like that that becomes nearly impossible late, but could have been, you know, a no brainer early.
SD: I’m going to come back to our two innovation leaders here and we talk about the challenge of designing automation, the design for manufacture. Are you both seeing changes in technology around, you know, the flexibility of automation or the adaptability of automation for gaps in the design for automation, in a product design, are you seeing technologies adapt or is this something that is still a challenge in our industry?
MH: As so, starting off, the flexibility is probably the number one issue I see when customers approach us now. You know, I have the benefit of seeing every new opportunity we look at in Life Science and a lot of customers are asking for flexibility in the automation solution because they’re thinking we’re going to have, you know, the mass manufacture of products just isn’t happening as often now. Customers want customization, so they want a lot of SKUs, low volumes. So how do we deal with that? Especially in a startup, they’re talking about scaling and their product is in flux. So what you see today, a year from now is going to be totally different. And so they want to invest, but they’re trying to figure out how do I provide- buy an asset that I can then later modify to still allow me to run and not and not throw that money down the drain. So it is a big pressure in the market. A lot of investment I’m seeing going into electronics, electro mechanics, how we bring more servos. We’re seeing them move away from pneumatics in a lot of cases where you have two positions, how do you provide flexibility in your automation?
MH: And the other thing I think is important, we talked about design for Manufacturability. It’s also, you know, having some insight when you approach it. We have a lot of customers come with two very different products and say, hey, can you automate something that does both of these? And to be honest, the cost to do that is well more than two independent systems in a lot of cases. So I think also there’s an education that needs of how do we look at your product and look early on to say, okay, where are the areas of flexibility you want in terms of design? And how do then we construct the automation, so you talked about islands of automate. Maybe we maybe we don’t automate a whole piece of it because we know, you know what, let’s just keep that semi-auto, right? So I think I think that’s a piece that’s also missing. I know you can’t see into the future. Our customers can. But I think they generally know what areas of an assembly are going to be going to change. That makes sense.
KW: It’s really interesting, Mike because you’re really talking about I said it can you automate is really developing that manufacturing strategy into product roadmaps which kind of goes back to my comment about the black box of automation. If you’re looking at typical product managers in say direct to consumer goods or anything like this, they’re really forecasting, yeah, what’s the next thing that’s going to make it interesting for the end consumer, they’re not thinking on that roadmap of how does it affect my automation, how does it affect my manufacturing? But I think some of the big scale from where Industrial Automation is going at this point, and we watch the electric vehicle space, for instance, we’re seeing so much technology in battery and that same flex you see in in a startup space but you’re seeing it in very established organizations now that the technology cycle is not years like it was necessarily in automotive where they could put out a car and ten, ten years later, you change it so you can get your ROI. It’s now the flexibility, but then also modularity. How do I take what I’ve got and modularized it to address that, that portion? And that’s kind of where that islands of automation, which is if you start that journey and your roadmap has modularity involved and your solution has modularity, you can very build up and you kind of have your Lego brick set up to, to automation.
SD: You touched on something that I’m very passionate about, especially inside of the team that I drive, you know, that modularity, those Lego blocks of automation, the enablement of scale and the fact that you identified that this is happening both in small companies and large companies today is really, you know, indicative of the environment that we’re in inside the automation industry, because even the giants of automation, the giants of manufacturing, they’ve been automated for decades, are still finding themselves in this position where they’re designing the manufacturing process as they’re designing the product. And so for any of our listeners that are, you know, maybe feeling alone in that, I just want to stress that you aren’t you know, it’s happening across the manufacturing world.
SD: And so Klaus, you know, talking about modularity, talking about islands of automation, what guidance could you give to someone who’s starting into their automation project, whether it be, you know, part of a large company or a small company, but trying to automate something new, do they need that overall approach to the manufacturing process or do you start in bits and pieces in modules?
KW: I think the overall it goes back to the road map comment. The overall road map needs to be there. You need to have an understanding of where you want to go. Does that mean you need to tackle every one of these like stepping stones or something that you don’t need to tackle every one from end to end? No, I think you can look at that very precisely. But it goes to some of the comments earlier, too, around what’s the goal, what’s the end goal, what is the process for the future. Mike talked a lot about the flexibility and the multiproduct and the many, many SKUs. If you don’t know that when you are- if you don’t know that that is the goal at the end of your journey, you’re going to start trying this modular route and go, well, I tried to be modular here, but this is exploding on me. So why would I use this? Again, just undermining some knowledge and capacity there. But having the vision before you take on each one of those small, small elements would probably be my biggest input to those companies and also not trying to get to the endpoint first. I had a personal experience just because I’ve got other stories. I’ll add mine, which was in a startup environment, we had this fantastic goal of we’re going to get the end product, which was big, which was bulky, which was complicated and that’s going to be our first step along the automation line here. And what we discovered very quickly afterwards was we don’t understand anything along the way properly to actually get this full solution in and ended up scaling back to something that was 100th of the size and all of a sudden success started building. And from that success, the iteration process and the modularity process could build up, which was a very exciting but very eye opening and in some cases very expensive learning lesson.
SD: Kind of comes back to the you don’t know what you don’t know until you dive in and start learning. And Mike does Klaus’ story and approach, you know, right from the very beginning, your framing thought was you need to understand your process and you need to understand your product. You know, understanding your process, starting with that end in mind and then working back. Does that align with your experience in dealing with, you know, effective implementations of automation as well?
MH: Yeah, I think a few thoughts on what Klaus was saying. First one case, we see a lot of his customers coming to us and saying, hey, we want you to concept an automation solution for these five products, but we don’t want these for right now. We’d like you to just look at this one, build it, and then we’ll retrofit in the field. Well, as you talk that through, a lot of our customers aren’t thinking, okay, when I do that, I got to take the machine down for three months to do that, I’m going to lose production. So I think forward planning, thinking through be simple, don’t you know you said it right, don’t overcomplicate it. Driving in multiple products sometimes does overcomplicate the solutions and thinking too far ahead you’re planning for things that may never happen and that can really drive-up costs and complexity. So I think understanding your long term vision, you’re not going to get everything right. But simple, simple, simple. Don’t try to overcomplicate.
SD: Being part of several scale ups myself, one of the best pieces of advice I ever got from a mentor was about hiring. And you hire for the job that exists today, not the one that might exist a year from now because it may never come to fruition. And you just touched on that because when you invest in automation, you’re essentially hiring a resource for your business to scale. And it’s a really similar thought process that if you get too far ahead of yourself, that future may never happen and you may have invested time, dollars, effort into something that just goes by the wayside.
MH: Another thought that came to mind interesting was around process and knowing your process. So, again, maybe it’s bringing it back a little bit, Simon, but I was thinking we get customers all the time coming in saying, hey, we want to automate this. And in the life science, very regulated, a lot of inspections, a lot of quality checks in place. Customers don’t realize this. Once you start putting cameras on their product and start looking for things, you find all kinds of things you never knew existed. So when we talk about understanding your process, I think it’s also, you know, I had a classic one the other day where, you know, someone came, says, hey, we want to glue these two parts together and then we just have an operator look at it and say is it good or bad? Okay. So now it’s well, what does that mean? What does good or bad mean? Because a vision system doesn’t have that ability, that thinking and processing quite yet that a human being does. So, and then we start putting under magnifications that weren’t normal and lighting conditions to find things. And now we’re finding more defects and so I think there’s a lot of work that, again, knowing your process, understanding what’s really important from a quality perspective as well, what is essential to making a good product is important and we define those upfront.
SD: And so that’s a really interesting example. And when you start to move from the human part of the workforce that can make judgment calls and they’re aligned to the overall objective of trying to put a quality product or trying to solve a problem. They can make those judgments. When you translate that to a piece of technology, you need to be able to write it down because you have to program that piece of technology to make a decision. It’s a go, no go. And when we talk about understanding your process, it’s really about that. It’s about being able to write down the criteria that historically one of the team members has just been saying. Yeah, yeah, that’s okay, that’s good enough, this passes
MH: And we’re doing a lot, and Klaus knows too, we’re doing a lot of work in ML and AI to try to improve that and have self-learning algorithms on our vision systems and in our, you know, on our HMIs to learn. But again, in regulated environments, you can’t just change things so understanding that and learning upfront is critically important.
SD: That’s a that’s a great example. So, you know, again, kind of coming back to the innovation leaders inside of of ATS, talking about what can be automated, framing that out in understanding both your product design and your process, I would say constraints or a process definition, you know, Klaus, what would you close off with some guidance or some framing thoughts for the listeners of the podcast today?
KW: I think there’s two things and without opening up a whole another side tangent right now, I think the one thing as you’re as you’re getting into automation and what can you automate, we were very product focused here, but I mean IA also other part of business has been very much in taking manual construction, large scale projects out there. So if you can define the task well, if there are these criteria that we were just talking about with Mike on what is good look like, is it standard process? Is it- are they repeatable processes? Taking those first notes down? And as you’re kind of first steppingstones into automation and going, yes, I can check a few of these boxes, it’s time to start looking into automation. I think are really good starting points for this and especially innovative ones because a lot of times you don’t think about those as you’re developing it. But, but taking that one step back, looking for those starting building blocks and then getting back to those experts and whether that’s internal or a partner to build them into the next phase of automation is really where you can take this.
SD: Thank you very much, Klaus. And Mike, how about yourself? Do you have some closing thoughts for for the listeners of the podcast today?
MH: I’m going to jump off Klaus and say 100%. My background is systems engineering and know your process, know your process, know your process. And you know, I’ll give another quick example. You know, I see all kinds of videos come across my desk where again, person takes two parts, puts them together, and you see them do a little wiggle with their hand. Right? And you don’t realize that that little wiggle creates a lot of issues and a lot of problems. You can’t really just wiggle something right. And you’re relying on the operator feel. So that’s a great example of bringing in some automation people. Let’s identify what’s actually making it work. What is the root cause of why that part is going together or being assembled properly? And then let’s figure out how we automate that. And I think that’s a step that is missed a lot of times. And I think that’s which can can help people. And the other piece I’d bring is not everything’s going to have an ROI, right?
MH: So, you know, again, a lot of examples where, you know, it just makes more sense to have an operator and that’s okay. Right. But let’s do the proper analysis. Let’s look at it. Let’s see, you know, adding a, you know, a $5 million little automation cell where you could really have two operators do it. Sometimes that just makes more sense. So I think that that’s an important thing too. I think that’s a key driving thought because when you look at what can be automated and we talk about, you know, really anything can be automated, it’s whether or not it should be automated (Mike: Right.) and no is okay, you want to invest in the part of your process, the part of your business that’s going to help you scale most effectively. And consulting experts driving into your both product and process understanding will help you better understand where that impactful implementation of technology is. Is that a fair summary?
SD: Absolutely. Nope, I agree. Couldn’t agree more.
SD: Well, Klaus, Mike, we’re so fortunate to have you on the Enabling Automation podcast, so thank you very much for joining me today. Thank you to those for listening to this episode of What You Can Automate Today.
SD: Next month’s episode will be discussing what to know before you start your automation journey. So we’ll actually probably build on a lot of the topics from today, around definition of product design and process. So I’d like to thank you once again for joining this month’s episode. And I would encourage you to subscribe to our YouTube channel to never miss out on the latest podcast. We look forward to seeing you next time.