Data-driven manufacturing at Arca Continental: With Insights Hub and Senseye Predictive Maintenance
Show notes
Welcome to the Trend Detection podcast, brought to you by Senseye Predictive Maintenance – which gives you visibility and insights into all your assets, from single machines to full plants to help you reduce downtime, increase knowledge sharing and accelerate digital transformation across your organization.
In this episode, we’re joined by experts from Siemens live from Hannover Messe 26 to explore how Arca Continental, one of the world’s largest Coca-Cola bottlers, is driving a data-led manufacturing transformation across its global operations, and how combining transparency, predictive maintenance, and a strong digital foundation is unlocking measurable operational impact at scale.
What you’ll learn in this episode:
How Arca Continental is scaling digital transformation across 45 sites by starting with transparency (OEE, cost, loss drivers) before moving to advanced use cases
Why predictive maintenance success depends on strong cultural alignment and early digital maturity — not just technology
A real-world example delivering impact in days — saving 13 hours of downtime on a critical production asset
How Insights Hub provides a unified data foundation to connect use cases, systems, and workflows at scale
What’s next: expanding with AI, contextualized data, and prescriptive insights to drive continuous operational improvement
You can find out more about how Senseye Predictive Maintenance can reduce unplanned downtime and contribute towards improved sustainability within your manufacturing plants, by visiting: www.siemens.com/senseye-predictive-maintenance
Transcript
Welcome to the trend detection podcast powered by Siemens. Hello everyone. And thank you, Alex, thank you Simon, for joining this conversation about ArA Continental and the data-driven manufacturing transformation and how technology people and insights can help organizations to enhance operations. But why don't we watch this clip before we start this conversation? Arca Continental is the 2nd largest Coca-Cola bottling company in Latin America and one of the most important in the world. We operate in the US, Mexico, Ecuador, Peru, and Argentina. As you can imagine, having 45 manufacturing sites around these 5 different countries involve having also different maturities in terms of technology and data management. There's a lot of people to be impacted in terms of the implementation of digital transformation use cases. So understanding the impact we have on humans that are still part of this loop, it, it's pretty important. So we try to have one excellent center in terms of digital transformation in at least one manufacturing site of each country where we operate that allows more people to be involved and have like a local geographically located example, you know, so you can just go and experience what we mean with digital transformation. Thank you. Excellent. Well, Alex, my first question is for you. Do you remember when we first approached ArA Continental, what were some of the challenges that they were facing at that time? Yes, absolutely. So it is actually we had a meeting with ArA Continental when their journey started here in Hanover. It was a couple of years ago and they were actually facing their challenges which are very similar to what a lot of manufacturers see. It's the Traditional productivity improvements were leveling out, so they had to actually use new technology to actually get there to slope up again. Another topic was in some areas they were concerned about that they were not generating the yield which they wanted to have, so higher spread rates than it was actually desirable. And then a very interesting one is, is very specifically, so when they of course are filling their products like Coke, etc. And they track very closely now the cost per case and understand what is actually driving these numbers and that was also a major challenge for them having this under control to match this with the rather tight margins which they sometimes get there so that they can actually do this as a profitable business. Interesting. So when those challenges were clear, do you remember how did they make that decision? Where did they start instead of trying to tackle everything at once? Yes, so. Very clearly they started with to get in first basic transparency. Do they actually know where do they have to go to make the changes to actually get also then the reward for they did. So what they did is the main use case for them at the beginning was OEE for getting this transparency where is actually the equipment which they have utilized to the best and where is there room for improvement and in these cases what are actually the main loss drivers and then. On top of that, they already knew at that point that they need to manage their energy costs more closely, and so they introduced building on our energy manager solution, energy management system based on ISO 50,0001. That's the starting point. And as they learn then where are the lost drivers. For example, which failure modes cause downtimes, what are the main reasons for scrapping parts of their production? That is then when they were able to go to the next stage and then adopt topics like predictive quality or predictive maintenance. Right. So now, Simon, my question is for you. What role did predictive maintenance play at Arca Continental to achieve their goals for digital transformation? Yeah, and, and predictive maintenance is, is obviously is a big buzzword, and it's also probably one of the ones that's the hardest to adopt and make successful with a customer. And we always know there's, there's certain key points we look at for a customer to see, is it going to be successful, and the key one for that is a cultural maturity. Support from the shop floor to the, to the top floor, and Archi Continental very much had this in spades, right? And we very much look for are they already on a digital transformation journey, because if they're not, and their first entry point is predictive maintenance, that's quite a tough ask, right, cause if they are not even doing OEE transparency yet, how are you gonna measure what predictive maintenance is, is achieving, right? So, once they had initial success with transparency use case. Then we started deploying Sensi predictive maintenance, which is a machine and vendor agnostic, highly scalable predictive maintenance solution, which we deployed with them. And because they had that great sort of strategic framework to deploy this and a lot of buy-in across the organization, you can actually get successfully quickly, and, you know, within 6 days of going live. They save 13 hours of downtime on a filler labeler machine, right, and you can imagine over 13 hours, that's a lot of Coca-Cola, that's not being produced, right, so, it's not always that quick, but with Alca Continental, because they had that, that right mindset and strategy across the organization, it's been really successful to help them along their digital journey. Excellent. So this question is for both of you. So how does insights of and predictive maintenance work together to achieve operation transparency and predictive insights? Yeah, well, I, I, I can start, Alex. Predictive maintenance never exists in isolation, right, it's, it's always part of a broader ecosystem of other products, other maintenance workflows. Enterprise workflows, and also often a customer will start with a certain use case, maybe transparency or quality, and then move to like predictive maintenance. And for us, Insights-Hub provides that backbone, that's where we store all our data, all our context, and we try to align across that to give a single experience to the user. So, a bit like Microsoft Office, right, you can start with Word. That works well, then you move to PowerPoint and you want the experience to be similar, and you want the data to be kind of interoperable and things like that. So that's, that's what I see ins Shub as as being. I think that's already a very good wrap up and what comes on top of it is, of course, if you moved into more use cases, you still have the same data foundation and that's an important piece for consistency. Another topic is also if you do this basically on the same platform set up, it's the. which you still have to build with CMMS on the one side, MES on another side, kind of, you do this once and then it's basically working similarly for the all the use cases, taking a lot of extra effort out as you're expanding with adopting new use cases and addressing additional production challenges. Well, you kind of have been with Arca still in the very beginning in this journey. Do you remember what were some of the early wins that the team made them feel like they were on the right path? Exactly. So when they, when they started, as mentioned, the focus was a lot on transparency, but that doesn't mean that it's just information and then nothing tangible comes out. I mean, one of the elements was that The cost per case could be managed more effectively and they could reduce the cost position there, which of course directly turns into additional profitability. They were able to increase OEE by 3% points and if you compare this basically to the large setup which we saw in the in the early video, this is actually a significant additional. Output which you can then sell on the market and then the last step is also that with more knowledge about what is actually causing disruptions, they could also increase the stability of the processes and then of course now with the adoption of the predictive use cases we are now actually adding now towards this and uh yeah and it's, it's actually a really good journey and uh we hope that ArA can actually keep this pace. Excellent. Well, Simon, one of my last questions about people and the partnership between ARA and Siemens. How is the collaboration and what, how has it been a factor of the success? Yeah, and, and, and as, as I hinted earlier, adopting predictive maintenance requires a real cultural shift, and we've been doing this now for over 10 years, and we have stories that are positive and negative. And you know, normal maintenance workflow is you're reacting to machines failing, and you're extremely busy doing that, and at times you'll probably get a quite a rush with being able to fix the machine quite quickly while the whole production line is down. So you're asking them to adopt a totally different workflow, stop rushing around, and use a system that can inform you how to schedule your maintenance activities. And to minimize that that crazy period of reacting to things. And doing that is tough. So, you, you need the team to fully support that. You need from shop floor to top floor to to support that. And certainly for the first year, we work intensively with our content, with our customer success team to actually almost be like the customer. We're almost using a system like they use it to help them adopt it, and when they get success, to make sure it's promoted and it spurs on further. Further expansion and and success, and our continental team has been very receptive to that, you know, we work very closely in in collaboration between our sort of customer success team and and their engineering team. Thank you. So looking ahead, what are the plans in the roadmap for Insights-Hub and Sens I predict maintenance? Maybe I start quickly so fast. It's actually there are 3 major building blocks. Number 1 is, of course, we get a lot of feedback from customers and that goes directly into product. That is always the most important thing, getting the customer feedback in. Number 2 is contextualization. If you look around over here, there's a lot of different elements going beyond the analytics which make up operational software and Analytics only gives the full value if you actually get this contextual information, no matter if it's manufacturing, execution, quality management in and you actually augment the data with this to actually get the greater insights. And then the last one, it's of course if you look to the innovation hub what we are doing there already, it's expanding on the AI capabilities which we have with the production co-pilot with agentic AI. Yeah, and, and, you know, the, the key thing is to keep being best in class on, on this, and I, I look at technology now, you know, you have the sort of the first generation stuff which is more condition monitoring and at least from a predictive maintenance perspective. Second generation is sort of the first AI wave, and now you've got the, the G AI wave, and it's gonna be completely transformative over the next few years and what we can achieve and it'd be even more prescriptive with what we give to our, our customers. Thank you. Well, this has been a very interesting conversation. Thank you for sharing your perspective on Arca Continental. And if you wanna learn more about Insight Hub or Sensite Predictive maintenance, you can come and meet with our team, the experts, and they would be very happy to, uh, share with you, do a demonstration. We are in the station 122 and 139. Please like and subscribe on all major podcast channels and find out more about Sensei predictive maintenance at Siemens.com.