The Future of Automation in Manufacturing

In our second episode of the Connected Technologies podcast, hosted by Paul Norford, guests Allan Anderson, and René Schrama delve into the world of machine vision and its transformative impact on manufacturing and supply chain processes. They discuss the evolution of technology, the importance of collaboration, and the challenges faced in implementing machine vision solutions. The conversation highlights real-world use cases, the significance of ownership and integration, and the future trends shaping the industry.

Connected Technologies Podcast, by Peak Technologies.

Transcript

Paul 0:03

Hello, and welcome to the Connected Technologies Podcast by Peak Technologies, where supply chain meets innovation. My name is Paul Norford, and in this series we'll explore how technology is reshaping the way critical operations run every day. From warehouses and delivery routes to inventory management through to retail and the customer experience. The pace of change across the supply chain and logistics landscape is accelerating. We'll be speaking to the people living that change first hand. Supply chain leaders, IT specialists, and mobility experts who understand what digital transformation really looks like on the ground. Today we'll explore the technology trends that matter, the strategies that keep operations moving, and why the areas like data security and sustainability are fast becoming essential foundations for modern supply chains. You'll also hear the real stories, the challenges, the breakthroughs, and the partnerships that turn complex problems into smarter operations. In today's fast-paced manufacturing landscape, the adoption of technology is essential for maintaining competitive advantage. One such technology that has gained traction is machine vision. But what exactly is machine vision and how does it impact the manufacturing and supply chain sectors? Today we'll explore key insights from industry experts Allan Anderson and René Schrama, offering a clear understanding of machine vision and its applications in manufacturing. Are you ready to explore what's possible? I sure hope you are. So now let's dive into episode two of the Connected Technologies Podcast by Peak Technologies. It's really good to have both of you gentlemen on the Connected Technologies podcast by Peak Technologies. Allan, René, really elated to have you here. Thank you so much for making the time. Allan, I'll come to you first. Can you just let me know who you are, who you work for, and what you do in the industry, please?

Allan 2:08

Yeah, sure. So my name is Allan Anderson. I am the founder and uh managing director of Clearview Imaging. Um I have been in machine vision for nearly 30 years, which makes me feel very old. So started off as an engineer, worked through into sales, and then founded Clearview 18 years ago. Um so yeah, been in the machine vision industry for most of my career.

Paul 2:28

Fantastic. And René, we all know that you're a stalwart of the industry. Uh, however, could you just let us know who you are, what you do, and how long you've been in the industry for?

René 2:38

Sure. So I'm René Schrama, I'm the chief commercial officer for Peak Technologies. In my daily responsibilities, I manage Europe and Australia and clearly heavily involved in our overall corporate strategy as well. In terms of our industry, it depends on how I define industry. If you're talking about automation technologies, then like Allan, uh I am also around 30 years in this industry with various uh companies and the last five years of those with Pete.

Paul 3:08

Fantastic. Gentlemen, it's really good to have you here. So, Allan, I I want to come to you first. So you mentioned the term machine vision. Now, in my head, I have got a machine with a camera on the top and it's it's looking at whatever it's looking at. Could you maybe just kind of describe what machine vision is and the impact it has on the manufacturing and supply chain industry, please?

Allan 3:29

So, I mean, machine vision, the the great thing that I love about it, it covers lots of different facets. Ultimately, you you're talking about a camera and software. So, you know, camera being the eyes of the system, the kind of software being the brains. And therefore, those two things in harmony are used in lots of different ways to automate what is effectively a human process, automating whatever you you want to do and your imagination is really the limited factor in that.

Paul 3:57

So, would I would I use it in my the things that I do? I mean, can you just kind of describe the different use cases or maybe one or two use cases where it's used?

Allan 4:07

Yeah, so machine vision, I think for me, tends to be more industry. Computer vision, you know, you can literally go to a social darts club. There's bars in London where they'll have computer vision systems tracking the darts as they fly through the air. That's computer vision. So very much at the at the forefront of kind of you know people and doing lots of different funky things. Machine vision, I see it as more factory automation. So we can think of a of a robot pick and place all the way through to label verification, checking that the things are printed correctly. That's more machine vision in our industry, if that makes sense.

Paul 4:45

Yeah, it does. So I thanks for kind of creating that line of delineation between the two. Um I I love the example of kind of computer vision and darts. I get that that resonates. When it comes to machine vision, you talk about kind of automation on assembly lines or or kind of things like that. So that that gives me that gives me some some sort of context. Why would machine vision be used? I mean, what's the what's the point of it?

Allan 5:08

Yeah, I mean, ultimately you're you know, we're looking at uh things in in the UK, you know, where I'm based and productivity, you know, we have a real productivity issue in in the UK and and we have labor shortages. So, you know, where vision can help there is that vision doesn't sleep. So, you know, if it's automating, it's not gonna be tired and make errors. So it's not necessarily about replacing jobs, it's just that where there's issues in manufacturing where people don't have workforces to do that, we can we can put vision systems in and if used correctly, they can help manufacturing processes run more efficiently. Spot, spot errors, save money, that kind of thing.

Paul 5:49

Are you seeing that there are real issues with lack of integrity for want of a better word, with these vision systems? And let me rephrase that question. So when it comes to these vision systems, what errors do they pick up? I mean, are are there are there inherent errors with without using vision systems? And if someone doesn't put them in place, like what things are they missing?

Allan 6:13

Yeah, I I guess it depends on on what you're trying to do. So as an example, let's think about a uh a part manufacturer in an automotive uh supply chain. There could be an instance where you're making parts that go into a radiator for a large truck, and down the supply chain, the manufacturer makes those parts and they arrive at their their customers who's then going to assemble everything. It could be parts, tolerances that are wrong in that manufacturing process, and ultimately then you know the the person fitting the system can't do that properly, and then you have to go back, scrappage, whatever. So there's an example where machine vision used by the the part manufacturer to check what is being manufactured and to make sure it's manufactured correctly, that's gonna stop that scrappage, that's gonna stop that that end thing being manufactured incorrectly, or time taken to rework. There's a kind of very classic example of where, you know, vision can help manufacturing solve a problem that maybe they've been dealing with for many years.

Paul 7:16

So that example that certainly resonates. You know, I'm a I'm a bit of a petrol head, I like trucks and cars and and that sort of thing. So are there real situations, real stories that you've heard of where clients haven't implemented machine vision and they've been having challenges and issues during the manufacturing process? And then kind of conversely, can can you maybe kind of tell us some some stories of of where a client has successfully implemented machine vision and they've actually saved money or saved resources, things like time and and like human energy?

Allan 7:50

Yeah, I so I think you know we tend to get involved when when something has gone wrong. So there's two types of companies that that think about using machine vision, the forward-thinking companies that maybe are, you know, they get they don't get to the point of the error and they're thinking ahead of how vision can be used to automate and and make things more efficient. You then get, you know, for want of a better phrase, maybe kind of ambulance-chasing business where something has gone horrifically wrong and you know, the company in question is going to lose a contract or the cost of escalated so much, so that it's becoming a real, real problem for them. So then you know they call vision companies and they try to implement that. But the the better situation is always the ones that are forward thinking. You know, they have a spec, they're engaged in trying to think about how they automate processes, how they how they stop problems happening before they happen. But in both cases, it is valid. So I don't, I don't necessarily, you know, I can't think of cases, you know, where I can give you examples where you know we haven't engaged, and you know, I can tell you about you know problems that have been happening, but I can I can conclude that typically when we get involved with some of the you know ambulance chasing business, it's because there's there's been problems. And so, you know, you know, sometimes that's that's absolutely fine and we can we can work with them, we can resolve issues. Sometimes the biggest thing we come come up against or combat is that those companies sometimes don't do anything, right? So they they they look at vision, they they might not go ahead because they're kind of worried about the costs or they're worried about adoption or nervous about the knowledge they need to do that, so they will they will not do that. They will go and get three quotes, they'll they'll do a whole kind of study of those different things, but they don't do anything. And then we might hear from those same guys six months later or a year later because they've tried to solve the problem again in the same way that they've always been doing that. Um, so look, I I think you know, in certain cases, there's probably places where vision isn't needed and they can continue doing things in the same old way. But having done this for nearly 30 years, I see, you know, I see more and more traditional manufacturing companies coming to us looking to proactively solve a problem or reactively solve a problem, and then actually going through into you know kind of full implementation. And then look, there's still, we'll get on to it, I'm sure, well, there's still risks in terms of implementation, in terms of you know, people that that make mistakes or choose the wrong products or the wrong partners. So this there can still be a pitfall of issues, and we we see those. We sometimes we're involved later in processes where vision hasn't been implemented very well as well. And we not that we have to rescue things, but we'll go in and help and and make sure it's it's done correctly.

René 10:42

If I can add to that, actually, because what we're seeing is it's it's a traditional technology adaptation curve taking place here. So, like Allan is saying, you know, he he started with this type of technology already 30 years ago, and certainly when within peak, we didn't have this technology already on our shelves 30 years ago. But what we're seeing, and I think it's the the proof point of the maturity of the technology right now, there is an acceleration of adoption that is taking place because the use cases are well documented typically, the business benefits are well understood, and instead of waiting for something to go wrong, we see an increased momentum of people accepting that this is now best practice, and therefore for us to stay competitive, we should be proactively looking at making sure that we replicate these business cases also in our own operations. And I think that is what is a really interesting momentum in the marketplace, which is one of the things that we have been observing, is and that's why I said in my intro, we we have been in automation technologies, and what perhaps was a barcode scanner in the past is now being replaced because we can do so much more now with machine visioning, and that is predominantly because of two things, I would say, and I'm probably less qualified well, not probably, I am definitely less qualified to make that statement than Allan, but from my perception at least, I think the camera technology has moved on enormously and is really powerful these days. And at the same time, the progress around AI and deep learning and the capabilities of the software element that sits behind that has opened up even more use cases now. And that is the really exciting and interesting bit.

Paul 12:26

So actually, so and René, thank you for that. I think you're absolutely spot on. I want to get your view on the kind of changes that you've seen over the last 30 years, and you alluded to the fact that machine vision is now becoming a de facto standard when it comes to innovating in the manufacturing lines and kind of technology kind of building blocks of processes. Allan, let me let me just jump back to you. What would you say is the inherent cost of not implementing a machine vision system?

Allan 13:00

Yeah, so okay, so a couple of scenarios, right? So if you look at um, you know, food production labeling, right? So if those guys don't have verification systems in place, you know, and they mislabel something, you know, stressful Friday night rush, whatever, right? The end result is that you mislabel something, it arrives at the retailer, and there's an emergency product withdrawal. So that could everything then get sent back. So there's all the scrappage of those fresh goods, and then an EPW fine could be anywhere from £15,000, £50,000. And then some of the retailers might have some kind of traffic light status. So, you know, you do that two or three times and you lose the contract. So and what's an EPW? Emergency product withdrawal. So something is mislabeled, you know, some from fresh produce, some poultry, some meat or something like that is mislabeled. You have allergen issues, you have sell by dates, use by dates that could be incorrect. If that is wrong and that arrives on the on the front line of a retailer's shelves and an end user or a consumer picks that up, there's obviously huge amounts of liability. So retailers will will do everything they can to stop something like that happening. And the pharmaceutical industry is way ahead of lots of other industries that we're involved in. You know, we we take it as granted that when we when we go buy some medicine, that you know, it's the right stuff. And so, you know, when we buy food and allergens and issues like that, we don't want it to be mislabeled, we don't want it to be out of date, you know, we want to we don't want to make sure that we're protected as consumers, right? So the cost going back up to the producer is an EPW fine, or they could lose a contract. And so you're talking about huge amounts of of risk. And if you've just got people that are checking that, I think that is is more and more a risky thing to do. So that's that's one example, and it's easy because you know we're talking about fines, we're talking about loss of contracts. When you look at then in classic manufacturing, we then get into the debate of return on investment, and in those calculations, we then have to think about how much is it costing to make the part that the person is producing, and then you're getting into scrappage. So can you can you quantify how much those parts are there? They cost a pound, and your scrappage rate is you know 50%. You can put a numerical value on that, and the financial controllers, directors can actually understand what the return on investment of something is going to be. So that again is is most customers are switched on to that, but some aren't. And some industries are higher margin, some industries are lower margin in terms of the products they're producing. So pennies might count, pounds amount might count. It really depends. So I think it's they're they're the kind of different ways that I would frame it in terms of financial risk and reward. And the better the customer that we work with has that identified, to have that well understood, and it's part of the conversation and the thought process up front. So I don't know if that gives you a good um illustration of the different cost, rewards, risks on that side.

Paul 16:25

It does, and it opens up a whole bunch of different questions and conversations from that. Renée, I wanna I want to bring you in at this point. Uh so there's clearly a collaborative relationship between Clearview and Peak Technologies. When did that relationship first start?

René 16:40

Uh I'm gonna say probably two and a half years ago, Allan. It was actually we we acquired a uh another business in North America, um which uh was distributing very similar products to Allan. I would say the industry is is not a massive industry in terms of size and number of people that are in working in that industry. So the network between these companies has always been traditionally very strong. So the acquiring business that joins the Peak family, the owner of that business was I had very strong personal relationships with Allan. And because we are also based in the UK, we naturally found a dialogue forming to replicate the relationship, uh, not necessarily of course acquiring, but that didn't mean that we couldn't collaborate in the market and leverage the expertise and product sets and solution designs from ClearView in our go-to-market strategy over here in Europe.

Allan 17:44

I think that there's a nice link also with with Zebra. So, you know, Peak, uh huge partner of Zebra, and so from from our side, you know, we you know in my career worked for Matrox Imaging, which for for many people isn't really a known name. And um about three years ago Zebra acquired Matrox Imaging, which I think was a massive investment for for Zebra. And I think that kind of partner ecosystem has been has been great. And so then working with Peak kind of intertwined with that a little bit has been has been fantastic.

Paul 18:13

Yeah, absolutely, absolutely. And it's it's it's really it's really wonderful to kind of hear that collaboration as technology moves on, that companies like Zebra Technologies can acquire companies to pull those into what they offer the industry and to keep innovating and pushing things along. So, René, what what do you see as the is the most important part of deploying a machine vision solution from the peak technologies perspective in manufacturing? What what what do you see as the is the most important, either set of rules or set of kind of best practices to really bring that to life?

René 18:49

Well, I always work with with our team to first of all understand the business outcome that the customer tries to achieve. Because ultimately we do more than machine visioning. So understanding truly what the business outcome is and then select the right technology, that using our wider expertise to think that this is the likely technology that might realize the business outcome is is definitely step number one. Plus, uh clearly, if the business outcome is, let's say, worth K, but we size the investment requirement to be a million, we can very quickly say to the customer, that is very nice. Technically, we can probably resolve this. However, you'll never get approved for the investment here. So that very early qualification of selecting the right uh the tap the right technologies for automation and making sure that the investment case would have a good chance of getting through the approvals, I would say is absolutely the first step. And then when it comes down to the machine visioning part, so let's say machine vision is indeed the technology that uh is most likely. Uh the second big question I always ask my team is is this a science project or is this something that we can replicate? So one of science projects, because no matter what we think, there's quite a heavy uplift at the front end of the design, testing, making sure that all the components could work together, setting up a POC, which even though we often and nearly always actually charge for it, we rarely charge the real value of all the things that we put into a POC. So we just ask for a financial commitment just to make sure the customer is serious, really. Um, so if it is a one-off, then clearly we designed something, we sold something, and then we have to start again. So it becomes far more interesting for us if it is something that we can see repeatability behind it. So a more common business case that probably doesn't just sit within that one customer, but is more for similar-looking companies, they were likely to have the same business challenges. So between those, as soon as we have green ticks, then I think we're all in.

Paul 21:02

Yeah, yeah, absolutely. So having having spoken about that, Renée, I want to stick with you. Having spoken about that, uh what what are the common mistakes that you see? Because you're you're clearly experienced in this, but what are the common mistakes that you've seen with companies looking to deploy or or trying to implement a machine vision solution?

René 21:22

I think Allan was touching on that earlier. I think the the most common mistake is that the number of specialists in the market that have a good technical understanding and more importantly have the battle scars of when things go wrong and therefore know where the pitfalls are. Finding a good partnership to engage with as an end user, this is is probably the hardest challenge because it it is quite easy to talk about machine visioning at a high level and and portray that you know everything. But as soon as it comes down to actual design of solutions. Solutions, that broad experience set is what you need to make sure you have a reasonably fast design because otherwise you have to test everything five times. Simple things like lights, you know, machine vision. We can't underestimate the importance of light. So the condition of light, it needs to be constant, it needs to be a certain frequency or a certain light source. Just just that little aspect is a complete domain expertise that is even more rare to find. It's one of the reasons why we're we're partnering so strongly with Clearview, because they have that 30-year set of expertise already in-house. They they've seen so many things not work. And that is just so valuable to us because when we see something, we think, oh yeah, that makes sense. And of course we we very easily miss something. So now we are clearly investing ourselves in building up our expertise alongside and we're going really quick, which is great, because thankfully we're we're a very well-resourced organization. But I always have been a strong, strong believer of building the right ecosystem around you to make sure that collectively you can just make s shorten that curve in terms of making sure that the the solution comes to life. But more importantly, single out all the things that the customer needs to know. So for a customer to find that kind of partnership in the market, I I just don't think there's there's a lot of availability currently. And in fact, if anything, I think the current demand because of the acceleration of the adoption is probably outstripping the talent that is available in the wider market today.

Paul 23:39

Okay. Okay. Allan, what would what would you say to kind of support René's René's comment there?

Allan 23:46

Yeah, so so I go back to to my experience in the the um the last few years, I think is is really enlightening in terms of of the change that we've seen, right? So if I go back to 10 years ago, let's say, and let's say uh a really large manufacturing company wanted to to implement At Vision, they would typically turn to an integrator, which is fine. But the the type of integrator would have been like a two, three-man person company, right? And they they have their place, very, very knowledgeable, um, been in the business for a long time, engineering guys. But sometimes, you know, the the the bit the that's missed is that that's a huge risk for the for the customer. You know, you know, what what if that company fails? What if that company isn't around in in 10 years' time? And yeah, I can literally give you examples of um big pharmaceutical companies who have entrusted single person organizations who have been really good at what they do, but the guy has retired. And so, you know, what what do you do with the with the 20 vision systems you've deployed? Yeah. So so now I kind of see a shift in in like enterprise. I don't know if that's the right word for for peak. I don't want to insult Renée, but I kind of see like enterprise level companies like Peak, who are, you know, a few hundred million dollars in revenue, you know, been around for a long time, good collective workforce, you know, so low turnover of people. So people have been there a long time and have good customer relationships, care about the customers. And they are the trusted advisor of that customer. So again, I'm not doing this to to be polite because I'm you know on with with Renée, it's what is what I actually see. And it's you know, it's in action where these guys will go from people go on on the kind of factory floor, walk, walk around with with the customer, and there will be the trusted advisor, you know, why don't you do this? We can help you with this, and then pull that back together. And so, you know, I'm proud of of Clearview and our knowledge, but we're great in our domain of of advanced machine vision and knowing where things work, knowing where things don't work, and that's our place. You know, we we haven't got the the breadth of sellers or the size that peak have got to be able to go and have those you know extended conversations and relationships and to be a real big, serious, trusted company, you know. So you know, and it's a bit like from Matrox to to Zebra, you know, ultimately the the Matrox uh the technology that's there isn't so different to the to the technology that that Zebra has now. But it's like that old adage that no one gets fired for buying IBM. You know, there's there's the technology is exactly the same, but the companies behind this now are really big, serious companies where there's trust. So whether it's implicit or whether it's subconscious or whatever, there's something that has tangibly changed. So for me, the technology hasn't changed. The technology is still there. What's what's happening now is that is that you know, really serious big businesses are are are taking this forward and allowing companies, manufacturing companies to de-risk this just by the fact that they've got you know a breadth of um of team. You know, what happens when something goes wrong? You've got service engineers that are gonna be there like within you know however long the SLA is. And so I think it's it's those parts that I see as a massive shift in the last, you know, five years, especially, and it's accelerating all the way through. And I think this is the really exciting thing that's gonna really lift. So, you know, obviously customers, manufacturing companies, they have to want vision, but but all these kind of little, you know, it's not at the forefront, you know, sort of subconscious risk elements are being taken away by the fact that companies like Pete, companies like Zebra are are kind of leading the charge now on this.

René 27:42

I think I'd like to add one aspect to it as well, because in manufacturing, um, the manufacturing sector inherently is international. It's it's quite often that a manufacturer has sites across the globe. And when a solution gets designed in a specific location, let's say in Ireland for argument's sake, um, generally speaking, the team from the customer side is actually a multi-country team that is working on a specific use case under, for instance, the industry 4.0 framework or initiatives. And then once they have actually got the solution verified, accepted, proven, and deployed, they're looking for repeatability across their entire state of similar production lines. Yes. And that is this is clearly a play where we excel because we have expertise to work literally across the grow sorry across the globe from our previous background because we already do that for these kinds of companies for other technologies. So that is another another reason I think that when you link the design and engineering expertise and the deep domain knowledge from ClearView, and you add the scale of being able to deliver and maintain and support, more importantly, after that, is a very strong combination, particularly in the manufacturing sector.

Paul 29:05

Yeah, you've absolutely hit the nail on the head with that, René. It's not just about procurement or deployment, it is about kind of support and longevity and standardization across global sites. So, yeah, everything that you've said completely resonates. So, Allan, just back to you. The technology sounds amazing. You've been in this field for 30 years. I have to say, it doesn't show at all. You know, I mean, some people may be able to kind of view the a short video here with you in, but it doesn't show. You look great, you know, 30 years.

Allan 29:36

Because I have no hair. But that's the easy. Silver.

Paul 29:39

Let's let's go with silver. It's not grey. It's not grey. How do you see so so clearly over your kind of 30 years longevity of being in this space, how do you see machine vision is evolving over the next few years? And I want to kind of split this down in two parts. The technology and also kind of what you've alluded to with regard to the support and the intangible parts of deploying and maintaining a machine vision solution.

Allan 30:06

So you know, my my job, you know, in terms of managing director of community, I have to keep an eye on on where we're where we're going and and what our position in the market is. And so there's uh a very large European show, The Vision Show in Stuttgart every two years. And it's always a very nice kind of moment to reflect on what is happening. You see the trends, you see the names, you see different things there. So when I walk through through through that kind of show, going back several years, you would see technology trends happening every year. Now that show is every two years, but you would see big technology leaps happening every every year. That's not really happening so much now. So if I think about the last few visits to that show, and we we exhibit also, you know, 3D has been a big theme, and and then sort of non-visible imaging and AI, and there's these have been the themes. So now what's kind of interesting is there's not anything that's like, wow, you know, that's that's new and and amazing. The the reality is that for me, you know, is technology is not the limiting factor. You know, the technology is there to do 99.9% of any kind of task you would you would need. The issue is knowledge. And so knowledge in terms of customers' understanding of vision and knowledge with with the ecosystem and people that create and deploy vision systems. So, you know, we're on a on a mission to kind of take that knowledge gap and just you know bring it right right down. So then, you know, our then position is to make sure that we've got really clever certified machine vision professionals that can help train people, so train people on how to even create a spec. You know, so some of the biggest issues when you deal with with end users is they don't know how to pull together a spec. So just on that part, all the way through to training how to how to create stuff. So got kind of broad brush stuff that I'm I'm covering there, but the technical side of things, I genuinely believe that you know there's there's not many limiting factors now, and the the price point of certain things is is is coming down. It's it's the knowledge on how to how to make sure that goes through is the key.

Paul 32:15

Rena, you've got a you've obviously got a a let me say a much more wider view simply because clear view are one of the kind of collaboration partners that you work with when it comes to machine vision. So almost a similar question to you. You've you've been in this industry for years. Many people know who you are and they've seen you evolve and change. So, what's your view on kind of where machine vision is going from both a technology perspective, but also how peak peak technologies are helping support those intangible elements such that a customer gets a great return on their investment?

René 32:51

Yeah, so I think from a technology perspective, I I don't think that I'm the right guy to uh I'm gonna just lean on Allan's answer here, if that's the right value. But what I what I because I'm more of a you you know me from old, so indeed. I am more of a customer-focused, go go-to-market what would work, problem-solving individual. Yes. And and I think what we've spoken about so far about manufacturing has been very much uh manufacturing process-centric. And what we're seeing uh popping up is that and what Allan just said, it's really interesting. He said the technology is just there, and what is now realizing is that you can use that technology in so many different ways. So if you just think about the back end of the manufacturing process, there's shipment, and outbound shipment validation is actually a massive growing area for us. Because there is a very interesting thing happening, it's not so much compliance or making sure there's no errors. This is about a chain of custody that is changing because as soon as it leaves your dock door, it is with a haulage company or a freight company, and then it arrives at the end user or the next uh station in the supply chain. If there's something that is gets damaged in that process, and this is where machine vision systems are now also really strong, is not just about inspection, it's all about also about having the repository to prove afterwards. Yeah? Yes. So so shipment validation, and we've got quite a few use cases now building up where the business problem typically is chargebacks. So somewhere else in down the chain is charging back penalties because they say you didn't ship it or it was damaged when it arrived, etc. So to be able to prove what you have shipped and the condition in which it was when it was shipped is massively important. So we're seeing a complete evolution of people being creative and quite often by the customer itself, right? They have the experience now in their own facilities, they see what they can do, and then they come to us and they say, Could it also do this? You know, and that is that is the really exciting part for us. Because generally speaking, the core technology stays the same, but then you have to design, for instance, if you ship a pallet, you need to design a gantry through which the pallet needs to go, and that's then where we add expertise into the whole delivery process of the technology. So the core technology is probably the same, but how you actually deploy that and how you maintain it after that, that again then plays to our strengths.

Paul 35:34

It does. One thing that really the supply chain industry has been again seeing tr seeing trends and and and things like that in supply chain is simply one word. It comes down to visibility, it comes down to being able to see exactly what's going on at each individual time, and to the word that you use, proving that things have been maintained in a certain situation or a certain scenario that things have been maintained and one is able to prove, no, this is what happened, or even going back in time, you know, this is what happened at this point, so therefore things happen further down, further down the line. It's a really interesting space to be in, supply chain, right now. It really is a very interesting space to be in. And as we kind of wind up the podcast, I I there's probably one kind of final question that I want to come to you with both. And Allan, I'll I'll start with you. So if if if there is a company who is looking to improve what they're doing when it comes to their manufacturing process and handing over to a three-peer at the end of a line, what what kind of three things would you say to them if they are looking at or even on the fence of deploying a machine vision solution?

Allan 36:46

Yeah, so I think the first part is be really, really clear what you want to achieve. So you know, it sounds obvious, but technically and commercially or from um, you know, internal flow perspective, you know, the where where things succeed or fail is is that really, really clear upfront you know, working document, if you will. So that's the first point. But then it's the stakeholders within the business. So I've seen vision systems you know work really, really well. The upfront part is great, the specification is great, deployment's great. But then there was not enough stakeholders in the business. So the the the guys on the front line, the manufacturing guys, whatever production guys, they were literally sticking screwdrivers and stuff in the in the vision system to make it fail because they feared that that was gonna, you know, you know, have a detrimental effect on on their working practices, careers, whatever. So just a lesson in terms of getting buy-in, you know, across the board. You know, you you sometimes see you know, guys from IT, guys from production kind of battling stuff like that. So there are things where you know a business has to get you know cross-party collaboration, you know, up from clarity on what you want to do, full stakeholder engagement throughout the business. Um, and and then it's look, it's engaging with the right people, right? And so, you know, I don't pretend that you know we're the only people that that know what we're doing, but you know, it's it's choosing that right that right company to to work with. And I think you know, in this case, there's a there's a really strong partnership where we play to to strength. So we know what we're very good at, advanced machine vision, and then speaker greater what what they do. So I think I think it's those it's those three things. If I was in the shoes of some of our um end customers, they would be the things that I would really be um taking care of.

René 38:31

Really? I would say I would summarize that as ownership, integration, and collaboration. And let me go deeper on those three. So ownership, who exactly owns which part of the process, where is the exact handover, but also ownership of the IT infrastructure. Because all these systems need to be integrated into what we call operational technology because it is not always evident and obvious who maintains what and therefore who has got responsibility of what. So ownership is a wider topic, I would say, than integration. Where as as soon as you start to design a system, the data needs to go somewhere. So the whole integration piece is that into a specific PLC? Is that the does the repository in the cloud, does it need to be on premise? What are the security aspects? Who would have the veto rights in that whole process because they've got corporate standards? So the whole topic of integration, and sometimes the data needs to go into two directions. It could be going to both the manufacturer and the 3PL. So integration is always a really big one because it's again never clear-cut in my experience. And then the last thing is collaboration. If if you're fine, if you're finding yourself in a place where the manufacturer and the respective third-party logistics are already in a conflict situation, you're you're almost doomed to fail with whatever project you're going to deploy. So to make sure that the relationship between those two parties is collaborative and they're both having the aim to actually design the right outcome for both parties. That is the ideal environment to actually design a solution into. So ownership, integration, and collaboration, I would say.

Paul 40:22

Today has been absolutely fantastic. René, thank you for those three kind of takeaway words. OIC, I suppose you could boil it down to the ownership, integration, and collaboration. Fascinating conversation, Allan. It's been great to have you on. I think there's been some really, really good understandings and deep dives into certain areas of machine vision, why we have it, and the cost of potentially not having it as well. And Allan, just quickly, if we were to do kind of more of this, would you be open to coming back onto the podcast to talk a little bit more about certain areas of machine vision, please?

Allan 40:57

Absolutely. I thought I'd be happy to do that. It's been fun and uh happy to talk about different applications or challenges that that I've seen and we see collaboratively with Peak. That would be a pleasure.

Paul 41:09

Fantastic. Uh, and of course, René, we would love to get you back onto the uh back onto the podcast as well to talk about just the whole refinement of go to market using all sorts of technologies, not just uh not just machine vision, but also in the integration of AI and some of those intangible aspects about the uh best practices and having the scars of rolling some of these things out. So, would you also be open to coming back onto the podcast?

René 41:38

Yes, of course, naturally. Likewise, I really enjoyed this conversation. I think there are so many more topics that uh we could cover off or or exchange our thoughts on. So, yeah, absolutely.

Paul 41:50

Allan, René, thank you so much for being on the podcast, and we're looking forward to getting you on the podcast again fairly soon. Thank you, gentlemen. Thank you. Machine vision is no longer a future capability. It's here and now. Its true value lies in proactive risk mitigation, preventing cost errors and liabilities before they happen. Shifting away from reactive fixes to strategic investments, reducing fines, recalls, and reputational damage. The real evolution of machine vision is about knowledge and ecosystem maturity, where investing in training, partnerships, and ecosystem sustainability becomes the differentiator and less about the technology itself. Ultimately, companies that proactively adopt it gain a real competitive edge, while those that delay risk higher costs, lost contracts, and missed opportunities across the entire supply chain. A massive thank you to my guests, Allan Anderson and Renée Charmer, for bringing their real-world experience to this conversation. The insights they shared from the trenches of actual implementation are exactly why we did the show. If today's discussion resonated with you, do consider subscribing so you don't miss out on what's coming next. If something we discussed sparked an idea for your own organisation, please don't keep it to yourself. Do share this episode with a colleague who may be facing similar challenges. Better yet, reach out to us and tell us how you're applying these insights in your own world. Those stories often become our best future episodes. Thank you so much for listening and do catch up with us next time on the Connected Technologies podcast by Peak Technologies.