How to get started in automation

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 third episode, host Simon Drexler is joined by special guest Ryan Tavares to discuss How to get started in automation.

What we discuss:

• How to get started on your automation journey
• Determining when you should engage an automation partner
• Defining what you want from automation

Host: Simon Drexler, ATS Automation (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: Ryan Tavares, ATS Automation (Life Sciences)

Ryan Tavares is the Director of Pre-Automation Services within our Life Sciences group at ATS. He has been with ATS for 15 years, in a  variety of roles from engineering to project management. He now leads the Life Sciences Pre-Automation Services segment, focusing on engaging customers in the early stages of their path to automation.

——Full Transcript of Enabling Automation: Episode 3——

SD: Welcome to the Enabling Automation Podcast. This is episode three, How to Get Started in Automation.

SD: My name is Simon Drexler. I’ve been in the automation industry for about 15 years in a variety of successive roles. I’m currently the General Manager of the Products Group inside of ATS Automation. We’re fortunate today to be joined by one of the experts from the ATS Group, Ryan Tavares.

SD: Ryan, can you take a minute and tell the listeners about yourself?

RT: Thanks for having me, Simon. My name is Ryan Tavares, I’m with ATS Automation Life Sciences out of our Cambridge office. I’ve been with ATS for about 15 years as well, with a variety of different roles from engineering and project management. Now I’m in our Life Sciences Pre-Automation Services segment, which we lead. The focus of our pre-automation services in our life sciences segment, is around engaging early with customers who are thinking about their roadmap or their path to automation and looking at ways to help the customers get there, specifically by addressing missing information or identifying areas of risk that could potentially come up in the future and that we can investigate and address early on.

SD: I think you’re the perfect guest to have this conversation with Ryan. You know, as the leader of the Pre-Automation Services Group inside of ATS, you’re constantly engaging with people who are trying to get started into their automation journey, regardless of where that is in the automation pyramid, whether they’re just getting started, they have nothing and they want something or they’re actually at the high end of the automation curve and they’re trying to do something new or something that’s never been done before. And you get to engage with all of them. And so the background that you have to be able to offer, insight and best practices to our listeners, I’m really excited for our conversation today.

RT: Thanks, Simon. It’s an exciting place to be. We see a lot of different opportunities with customers on different ends of the spectrum. We might have customers who are brand new to developing a product in the R&D space, and they’re thinking about a long-term roadmap in terms of getting that product to market and what that looks like. And maybe it’s the first time that they’re thinking about things in that way and they’re looking for that kind of expertise to help them get there. And then at the other end of the spectrum, we often deal with some really great customers who are established and have been in the industry for many years and know how to get there, but look for different ways to augment their expertise or bring in some industry knowledge from an outside perspective like ATS that can kind of help them bridge those gaps to where they are from, where they are today and where they want to get to.

SD: Yeah, that’s perfect. If we’re going to jump right in, and this is normally how we structure these discussions. So, let’s start from the top. How do you get started on your automation journey? I’m certain that we have a listener today, that has a scaling business. They have a product that’s partially defined or mostly defined but works really well for manual and they’re looking to dip their toe in or maybe jump to feet into automation. How do you tell them how to get moving down that path?

RT: I think the first thing to really understand and to think about is, what are the long-term goals and where are you at today and how do you get from where you’re at today to achieve that long term goal? The automation often kind of comes out of that type of an assessment and the automation becomes a way to address certain needs of the business. It’s not always the case that automation is the answer but thinking about things from a from a longer term perspective and looking at establishing what those goals are. Those goals could be things like a cost reduction play. It could be using automation to improve product. So certain things about a product that are complex or tricky to assemble by hand where maybe automation can help with. Are there particular processes that are not well-suited to be done by an operator with an operator’s hands? And can automation help in those areas? Automation is a natural candidate for helping to improve manufacturing capacity, so in the right phase of a customer’s lifecycle, introducing the automation to get more product to market. And there’s also other things that automation can help with, including improving safety. If there’s particular chemicals or products that are risky for an operator to be exposed to, automation can help by handling or dealing with those products in a more contained or controlled environment to isolate any interaction between that product and humans.

SD: Excellent. There’s a few things there that we can dive into in more detail. I think one of the most important things that you said, we’re on a podcast entitled Enabling Automation. But one of the things that you said is automation isn’t always the answer. You might not actually need automation in the current state of the business, and I’ve seen that from time to time as well. Where there are other ways; there’s always multiple ways to accomplish the objectives of the business. And there are times where full automation isn’t the right thing. You know, there’s technology augmentation. There are different tools that can be offered. There is process implementation. There’s outsourcing to different experts. There are often and commonly other ways to solve problems. So we want to make sure that we really narrow in on when automation is the right time.

SD: The other thing that I really liked, what you said was, you start with the end in mind. I’ve been fortunate to engage with lots of people who kind of walk in and say, in our business of scaling, we’ve got to go lights out and lights out is a great goal, but it normally starts significantly before that.

RT: That’s a great point.  I think we often get involved with conversations with some of our customers that are thinking and want to go to lights out automation. But what we uncover when we go and we think about that, in many cases, it’s likely a very big jump from where they are today. Could be significant upfront investment in both upfront capital as well as floor space and infrastructure to be able to support a large-scale system. But the exercise of understanding and getting that vision in mind, I think is helpful because then you can look at that might be a longer-term goal. So maybe that’s something that you want to have in mind as a target for the next, maybe six years out or 5 to 10 years. But with that vision in mind, it can help guide what the immediate next steps are. In the near term, for example, lights out automation system may make sense down the road, but to minimize the hurdle of getting there and that upfront investment, you might want to break that down into smaller pieces and say, are there elements of that full scale system that we could look at and assess in terms of their feasibility to be implemented today? And does it make sense to bring those automation modules forward on a piece-by-piece basis on a smaller scale and start to bring those into your organization today? So there’s a few things that can help with that within organizations that are less used to automation or don’t have an adopted automation and complex equipment at a high level, starting slow and starting with individual modules upfront and starting early in specific pockets or areas where automation solves particular problems can start to get the organization used to and familiar with running automation and then starts to prepare them for adopting and bringing on larger scale automation in the years ahead.

SD: And I think that learning curve is a really important part of taking sort of one step at a time towards a broader view of what Future State shopfloor might look like.

RT: That’s right.

SD: I think what would be really valuable for our listeners today and under the topic of how do we get started in automation is to really understand how they start to identify what modules might be important because I’m certain that there are lots of people listening that they’d like to jump to maybe the automation that you see on how it’s made, but it doesn’t always happen that way. So how do you pinpoint where you start to engage a partner or where you start to do your analysis on? Is automation right for me?

RT: That’s another good question. I think the first part to that question is really understanding what metrics or the thing or the objectives that are important to the organization. If there are a handful of things that matter, for example, is it reducing the operator, the number of operators that you’re utilizing today, is it reducing the cost to manufacture the goods on a per unit basis? Is it improving safety? Is it increasing the capacity of your overall manufacturing at the facility level, in other words, addressing a bottleneck system? I think what’s important is understanding what those potential objectives are and then assessing your current state operation in individual process areas or modules of assembly or work centers. And looking at those and saying, okay, based on those KPIs which of those particular areas have the largest opportunity to improve those specific KPIs or metrics that are important to our organization. And then on the other side is looking at, okay, if we were to assess each of those individual areas and say hypothetically replace one with an automated system, what would those metrics look like and on the flip side, where you take that out and you introduce automation? I think starting with broad strokes and organizational level or an operational level through your manufacturing process and looking at each of those pockets and saying, you know what if we were to do this and take your existing operation out and put in an automated system and then reassessing in that theoretical environment, what would be the opportunities?

RT: I think that starts to paint a picture about where the likely opportunities lie, which could help guide the prioritization of what should happen first in terms of investing in automation. And again, that’s more of a near-term picture and something that is common as a customer is starting to close the gap between where they are today and ultimately getting to that full scale lights out automation vision that they might have.

SD: Yeah, to get to where they want to be. And what you just described is so valuable to someone who’s listening today, you know, if you’re an operations manager, a manufacturing engineer, anybody who’s trying to justify investment to drive change, that’s a great process to walk through because what Ryan just described is how you drive alignment across a number of stakeholders. You choose what’s important to the business, you list them out, and then you identify and evaluate how driving that change will drive an improvement or a decrease to the metrics that you care about. And so that’s something where you can communicate across stakeholders and really get alignment to something that’s going to help move the business forward. And automation or technology or process advancement is really no different than driving a variety of other changes inside of an organization. It’s just driving different metrics.

RT:  And it’s important to mention as well on those metrics there, you know, a lot of them are financial, right? There’s typically a justification or a business case that that should make sense for the investment. But then also there’s a lot of non-financial metrics as well that look at it from a complexity perspective, looking at it from a human capital perspective, a safety perspective. In some cases, even though financial metrics may not justify investment, there’s still a reason to do the automation because that’s part of an overall vision or an overall strategy to adopt automation at a higher level.

SD: And so for those listening, we’re starting to walk the journey, the process of how you get started. So we’ve now defined our goals. We’ve looked at the metrics that are important to the business. We’ve evaluated a number of different projects that we might do, choose that first one that’s going to make the biggest impact to the metric we care about the most. Now we start to look at how we define equipment requirements and define what we want. So have you seen a best practice or a best approach in the work that you do, Ryan?

RT: This is really key to getting to the next step where we can actually start to drive actions out of these initial assessments or roadmap planning. So the answer to your question is yes, the definition around the product in the process is really key. And the reason for that is that really starts to form the foundation for an engineering team about what they’re going to do in terms of design and build of an equipment, of a piece of machinery. So understanding everything about your product and your process, how it comes together, what you look for in a good product or a good device. What makes a product become a good product during the assembly process or one that may be something that becomes a reject or waste, understanding critically how you measure those and what causes those things to happen are ultimately key to really understand, because those are the things that inform the basis of a design for equipment. Defining those requirements is not just around the technical. There’s also definition around where the business requirements for that particular piece of equipment are. There is operational, financial or other technical requirements outside of the product and process itself. Those are key to really understand because it’s great to have a machine that does one thing, but it’s also even more important that that machine is well adopted by other stakeholders within your organization, aside from just the operations or the technical team.

SD:  The line that you just drew around being able to define product and process, I often look at that as the line between a human based approach and whether that’s people with technology augmentation versus a full technology approach. And if you can write it down, if you can define it, you can likely move to the line of full technology implementation. Is that a fair framing of being able to define product and process?

RT: It is, yes, it is. One of the terminologies that we often talk about when we’re talking about requirements is characterizing the process. If there’s a process that’s being done today where an operator is doing the majority of the operations on that particular product, if the operator is doing a lot of things that can’t really be quantified or there’s a lot of judgment involved in what the operator is doing, those could be viewed as more complex processes or maybe more difficult to automate. On the flip side, if those could be characterized and defined and codified, so to speak, then it might be a little bit easier to translate that into a technology which can then reproduce or perform those same steps that the operators doing today. I think the extent that processes can be well characterized upfront and well understood, the better the likelihood of success during an engineering and design phase when you’re working on those pieces of equipment and designing those equipment, that that equipment is actually going to do what you expect it to do in the end.

SD: Right. And I’m certain in the role that you have in pre-automation services, you’re going to have worked with a number of people who are either new to defining their processes or they’re not able to. And what happens when you’re not able to define your process?

 

RT: That’s a great question, Simon. So an organization who’s working on developing the product or the process, it’s great to understand what that long term vision looks like. But moving ahead too quickly into automation, too soon, where a product and process is not well understood, could open up the risk of potentially needing for a lot of change in the future as that product evolves or revisions come to play or the process becomes better understood and the realization is had that there’s certain things that need to change on the equipment now, on the manufacturing equipment to accommodate those new product or process changes. For organizations who are in those early phases of product and process development, that work can be done in an R&D environment or in a laboratory setting and that work is often done without automated equipment. But there’s also opportunities where smaller scale automated equipment could help them to develop that. So again, with that automation goal in mind or what a lights out system could look like, taking a slice of that or a sliver of that today and having that in a very small scale setup, providing that platform for that customer to do that continuous product development and continuous improvement on the particular product or process, then allows them to have the basis upon which they can scale up that particular process or process automation step in the future.

SD: This conversation really touches on two parts of the discussion so far. You know, it circles back to the beginning where we’re discussing how you get started. And part of how you get started is to be able to define the process. And if you can define the process, it’s very likely that you can apply technology and potentially automation to that process, and it will normally be a return on investment type decision. Is there enough impact to your KPIs to justify the investment in the technology? And then there’s a secondary part of what we’re talking about now. Where is, if you can’t define the process to that point, the discussion becomes, can we create technology or is it worth it to create technology to take that role in our organization right now? Is that a fair framing of the two sort of paths?

RT: Yeah. So, if I can I guess think the way I think about that is so there’s multiple paths that can be taken. And I think the key circling back to, as you mentioned, the understanding where things are at today, how far along or how mature is the product or process? How far out are you thinking about automation? Is that a five-year plan or is it a ten-year plan? So, to some extent, there’s going to be some market data that’s going to help to drive or paint a picture about where you want to be in the future. And the extent to which you accelerate that roadmap is really going to be driven more on where you see the market demand today. If it’s a double in five years or if it’s a quadruple in six, those paths are very different. And those are going to drive different behaviors in terms of how quickly you’re going to want to develop and nail down your product, your process, and start to adopt that automation. I think what’s interesting is even though the forecast or the outlook for the market needs for particular products or devices vary between organization and vary depending on what they view the market demand as for those products, I think what we see is the phases and the steps that organizations take are similar in terms of how to get there. What changes is how quickly they need to do things in terms of expanding or contracting that timeline to align with and meet the needs of the market.

SD: That’s interesting, so as someone listening today, it’s about going through similar steps regardless of you being a small organization or large organization, but it’s how quickly you need to accomplish your priorities based on the metrics you’re trying to change.

RT: That’s right. Different size organizations obviously are equipped with different resources to be able to do things a little bit quicker or some require a little bit more time or investment or funding before they’re able to do certain things. But in either case, I think there’s  still good fundamental practice and judgment that still makes sense. Whether it’s an organization that has a lot of capital or a startup, there’s still an interest of doing things in a cost effective way up front that make sense, but still drive the results that allow you to move to the next step. In other words, even for large organizations, it doesn’t really make sense all the time to go lights out automation on day one, it still makes sense to go through that initial R&D and a pilot line or a low volume production phase to still test the waters to make sure that it’s a viable from a market demand perspective before you actually start to look out at larger volume lights out automation.

SD: I think that’s great advice because we’re all trying to move to the next step. It’s very rare that you meet a business that isn’t trying to grow. And it’s normally not just the next step, it’s the next step and the one after that and the one after that. So, Ryan, again, I know in the role that you have, you engage with many customers trying to do this. If I’m a business and I’m trying to plan for growth at the same time I’m trying to implement new technology, implement new automation, how do I build that into the metrics that I’m choosing? The first project that I choose, the equipment requirements that I write, how do you build in that? I know I’m going to change, I know I’m going to grow. What should I be asking? What should I be writing down?

RT: That’s a that’s a good question, Simon. I think when we think about having that end goal in mind or that that ten year vision in mind and bringing each of those individual modules back and assessing and saying, what of those particular modules makes sense to actually implement today. I think what often comes out or comes to light of those types of assessments is, there’s going to be some very, relatively attractive opportunities for doing some  more traditional things like maybe offsetting costs by reducing headcount if it’s a particularly laborious process today, there may be opportunities of automating maybe very process steps that are a very low complexity on a relative scale and that can allow for an organization, again, just to start to think about and bring automation into play while not being overly concerned or worried about automating very risky processes or processes that are critical to their operations or making good salable product. Looking at things like good candidates from a technical perspective, that that would be an immediate fit for automation today, looking at opportunities where, you know, where can we address and maybe it’s a bottleneck in a system. So maybe it’s a simple process or assembly step that’s happening today that is a current bottleneck and maybe automation can be introduced there to give a little bit of a lift in capacity in throughput of that system and ultimately leading to greater production capacity, in the end of the day. So those are some, I guess, quick wins, if you will, that can start to help bridge the gap and get an organization under the roadmap to the next step in getting them closer to lights out automation.

SD: Ryan, really appreciate the time that you spent chatting with me today about how to get started in the realm of automation before we close off today’s episode, would you want to leave our listeners with any closing thought?

RT: Absolutely. Thank you, Simon So I think what I would leave everybody with is really, you know, when you’re thinking about that roadmap to get you from where you are today to where you’d like to be in the future, it’s really about thinking about building in options, flexibility, and off ramps that allow you to pivot or change direction along the way. Because that roadmap or that plan can often be, you know, multiple years, ten years, maybe ten years plus. And no doubt things will change as time progresses. But the real key is to understand what that end-to-end vision looks like and then dial it back to focus on the actionable decisions that you can make today that are going to start to take you to that next step. And then it’s about revisiting. So take that plan again and revisit it in six months. Revisit in a year. Because like I said, it’s, you know, it can change. But the key is to keep that in mind and to keep adjusting as time progresses to make sure that your plan is always being refined and is in alignment with the, you know, where you see strategically the direction of your organization heading ultimately to meet the demands of the market.

SD: Great advice and an awesome way to finish off today’s episode. Ryan, thank you very much for joining us today and to the listeners for episode three of Enabling Automation, thank you for joining us. If you found today’s discussion insightful, I highly encourage you to join our next episode, which is episode four, which talks about what can you automate. And we have two experts, Mike Healy and Klaus Werner from the Innovation arm of the ATS Group to talk about what’s possible. Thank you very much, everyone.