The World Cement podcast: a podcast series for professionals in the cement industry.
Hello, everyone, and welcome to another episode of the World Cement Podcast with me, your host, David Bizley, senior editor of World Cement. With the background of the first robot half marathon being held just a few weeks ago in China and the winner completing the course in fifty minutes, a full six minutes fifty four seconds faster than the human record, we're diving into the surprising crossover between the worlds of robotics and cement. To guide me through this exciting topic, I'm joined by Thierry Obadiah, Chief Revenue Officer of Anibiotics, and Robert Christinger, Head of Automation at Vigier Cement. I just wanted to take a moment to remind you to register for World Cement. It's free of charge and gives you access to the latest issues of World Cement, both in print and online.
David Bizley:Every issue comes packed full of regional analysis, technical articles, project case studies, and the latest industry news. Simply head over to worldcement.com, click the magazine tab, and register today. It's as simple as that. Happy reading. So Thierry, Robert, welcome to the World Cement Podcast.
Robert Christinger:Great to be with you. Hi.
David Bizley:Hi, David. Now before we get into the detailed discussion today, let's begin with some introductions. Starting with you, Robert, tell us a bit about Viger and your role at the company.
Robert Christinger:Okay. I'm Robert Christinger. Viger is a cement plant in Switzerland, an integrated cement plant with an annual capacity of approximately 900,000 tons. We have about 20% of the market. We have an integrated plant, so we from quarry through to the out to the market.
Robert Christinger:We have three bore molds and precalcina.
David Bizley:Okay. Thank you. And Thierry, same question for you.
Thierry Obédé:So I'm Thierry Obédé. I'm the chief revenue officer at Anybotics. Anybotics provides best in class end to end autonomous industrial inspection solutions. So my role is to translate what is actually a breakthrough technology into commercial reality, building the go to market engine that takes autonomous inspection robots from a compelling innovation into a scalable and trusted solution for asset intensive industry worldwide. My responsibility is all about commercial, including sales, business development, customer success.
Thierry Obédé:And also my job is also about making sure that the right industrial companies find us, understand what's possible and then stay with us long term transformation partners, not just the technology buyers. Okay.
David Bizley:Excellent. Thierry, sticking with you for a moment so we can get a a clear idea of what we're actually talking about today. Give us an overview of the system being deployed at Vizier. You know, in broad terms, what is it designed to do? And for the benefit of those listeners who haven't actually seen a picture of one yet or seen one in person, what does it actually look like?
Thierry Obédé:Yeah. So let let me start with the physical because it's actually striking the first time you want. The animal is a four legged autonomous robot, roughly the size of a large dog, about 50 kilogram. It works, not on wheels, not on tracks. It works, which is a critical design choice because industrial facilities are not built for robots on flat surfaces.
Thierry Obédé:They have stairs, grated floors, terrain, tight corridors. So animal navigates all of that autonomously and without a remote operator guiding it, it carries also a sophisticated sensor payload, cameras, thermal imaging, acoustic sensors, gas detection, and it operates continuously in conditions that will be uncomfortable or dangerous for a human inspector. Now what is actually doing at Viger at its core, it's performing the inspection rounds that will otherwise require a technician to physically walk the plant on a scheduled basis. But the moment you move from human rounds to autonomous rounds, four things change. And this is where the real value unlocks.
Thierry Obédé:So first is about coverage. So fixed sensors cover what they cover and nothing else. An email reaches the assets in between. So the ones that don't justify your permanent sensor, but absolutely need monitoring. Second is perception.
Thierry Obédé:The robot sensor suite detects things the human eye simply cannot. Thermal anomalies, acoustic signatures of early stage bearing wear, gas leaks at trace concentrations, problems that were previously invisible until they became failures. Third is safety. In the cement plant, there are areas that are hot, dusty, noisy, and hazardous, and every routine inspection round that animal takes instead of a technician is a reduction in human exposure to those environments. That matters for compliance and it matters also for the people.
Thierry Obédé:And fourth, and this is the one that which is underappreciated. It's about knowledge to become permanent. The experienced inspector who knows that a particular motor sounds slightly different on Tuesday afternoon, that tacit knowledge when it retires is gone. With animal, every reading, every anomaly, every baseline is captured and retained. The plant's intelligence compounds over time rather than walking out of the door.
Thierry Obédé:That's what's being deployed at VVA, not just the robot doing rounds, a fundamental shift in how the cement plants manage reliability, safety and cost.
David Bizley:Excellent. Thank you. Robert, so we've seen some interesting changes in cement plants over the years with increasing digitalization and automation and decarbonization technologies coming online. But even so, autonomous robots walking around is still quite different from the norm, right? So how did this collaboration even come about?
Robert Christinger:Johnny saw antibiotics the first time, I think, on LinkedIn. Johnny is Johnny Stoyan is our chief maintenance manager. Whenever he sees things like this, he comes to me, asks me what I think about it. I had a look at it, and I thought, yeah, this could be something, not only what it is now, but what it could come one day in the future. So we were invited to Zurich to see the robot, and we were impressed.
Robert Christinger:We also saw the robot in another cement plant, which I'm not even sure if it was only there for show or if it has been deployed there. But when we saw it and we saw the different advantages it has, we thought we have to get into this. There's been an initiative with WIKA since two, three years that we must start thinking of innovation, digitalization, data platforms, and so on. And at the beginning, oh, where do we start? But when we saw this antibiotics, if you think of what it could do, even if it doesn't do it, it's obvious that there's huge potential in this technology.
Robert Christinger:So we started off with a pilot project. Antibiotics, if I understand correctly, borrowed us the robot. They showed us what it could do. So they came, measured out the plant, and technicians, I must say. We don't have scientists.
Robert Christinger:I'm not a scientist. I'm an engineer. The scientists we see at Antibiotics and we are users and that works fine. And Johnny wanted our employees to get closer to this to see what a robot is. So we presented the robot to everyone who's walking around around the mill, showing that it's it will avoid you.
Robert Christinger:It will stop to get them familiar with this, which was actually a very good move by Johnny to help us move into this new era and to accept that they are robots and to try and understand what it can give us.
David Bizley:Your robots have been in operation at Vizier since 2023, I believe. And, you know, as we all know, a cement plant could be quite a hostile environment. Were there any challenges or teething problems that you had to overcome to get these robots up and literally running?
Thierry Obédé:First, deploying an autonomous robot into a working cement plant is not a plug and play exercise. And this year, any serious industrial operator didn't expect it to be. So cement environments are quite hostile, fine particulate dust that they, that gets into everything. Extreme temperature differentials around kilns and clinker coolers, complex multilevel structures, area with very poor lighting. So animal is engineered for exactly these conditions, but every plant has its own specific layout, its own operational rhythms and its own edge cases.
Thierry Obédé:That's true of any serious industrial deployment. So what I think is worth highlighting and what actually differentiates Anabutics as a company is how we approach that reality. Speed of deployment was a priority from day one. We don't want to believe in the multi year implementation programs that deliver value somewhere in the distant future. The expectation we set with Viger and with every customers is that you are operational and generating data quickly.
Thierry Obédé:That discipline for us to be rigorous in our onboarding process and precise about what needs to be configured before go live. But speed doesn't mean abandonment after the uncheck. The customer support model we operate is actually continuous. When you deploy any more, you don't get a robot and a manual, you get your team. Our customers and engineering teams stay close to VGA through the early operational phase, because that's when you learn the most.
Thierry Obédé:But where does the robot need route adjustments? Which readings needs by lines calibration for this specific environment? What does a normal look like in this plan versus an anomaly worth escalating? So those answers come from cloud collaboration, not from the deployment documentation. And then there is what I would call the compounding advantage.
Thierry Obédé:The constant update cycle. Anibiotics pushes regular software and capability updates to be deployed, automatically. So the robot VG operates today is meaningfully more capable than the one we commissioned in 2023. Detection algorithms have been improved. Navigation has been refined.
Thierry Obédé:New sensing capability capabilities have been added. This isn't a capital asset that depreciates. It's a platform that gets better and that requires constant connection with the customer to make sure each update is integrated smoothly and that the Vizier teams understand what's new and how to leverage it.
Robert Christinger:I can also confirm there. Of course, we know best the industry we would like, couldn't we check this with the robot? Couldn't we measure that? And I must say, Antibiotics has very good engineers. They supported us all the way.
Robert Christinger:And if we said we would like to do that, they could tell us, yeah, there are plants doing that and they would help us get to whatever measurement or inspection we wanted. Okay. Thank you.
David Bizley:Robert, on a typical day at Vizier, what are the robots up to? What is their routine? I know we touched on this a little bit. What are the advantages over using a human for these tasks?
Robert Christinger:We have one robot at the moment, will probably be more in the future. He has daily tasks. He does, at the moment, 200 measurements. This will probably become more. The main advantage is whatever he checks and inspects is documented, Which previously you had a piece of paper, very good engineers and they wrote down everything, but this paper is somewhere and it's just gone.
Robert Christinger:Whereas the data platform of antibiotics reports everything, you can search for it and they have all kinds of means of presenting it and it's a lot of data. But it helps you navigate and still see the trees although there's so much forest, which is what we use it for. We have also health and safety issues. We measure certain gases. Maybe in ten years time someone says I was exposed to these gases over the years and that's why my left leg doesn't work or something.
Robert Christinger:And this is a huge advantage that we might only find out years later. At the moment we are limited. We only measure two gases, that is ammonia and CO. But the way I see it, there's no limit to what this antibiotics payload will accept one day.
David Bizley:Okay, excellent. Thierry, so we've heard a little bit about the daily routine and zooming into the hardware involved on the platform, how are they actually navigating the plant? Can the robots react to changes or unexpected obstructions on their route? What kind of sensors or systems are they using during inspections?
Thierry Obédé:Yep. This is where I would invite anyone who hasn't seen anyone in operations to watch the footage because there is a moment when you watch this robot navigate a staircase or step over and unexpected to stay cool or something clicks. So this is not a remotely controlled machine. This is not a joystick, no operator guiding it through. It's actually thinking about where it is and what to do next.
Thierry Obédé:So the foundation of this is AI artificial intelligence. And I want to be precise about what that matters because AI gets used loosely. So at Anybotics AI, it's what makes autonomous inspections readable in environments where it matters most. For our customers that translate into two concrete, two very concrete things. A robot that can go anywhere in their plant and a robot that knows what to look for once it gets there.
Thierry Obédé:So on the navigation side, animal builds and maintains the three d map of the facility. It knows its roots. It knows its waypoints and it executes its mission autonomously, but it is also continuously perceiving its involvement in real time. For instance, if a pallet has been left in a corridor or if a door is unexpectedly closed, or if there is a vehicle that wasn't there on the last round, detects it, adapts its path and continues. It doesn't stop and wait for a human to intervene.
Thierry Obédé:And if it encounters a situation, it cannot navigate safely and obstruction, it cannot run, it cannot route around or a condition outside its operating parameters. It doesn't guess it stops its signals or it returns to its base. Safety is not a feature layer added on top. It's embedded in the decision. What is the robot actually sensing, as it moves through the plant?
Thierry Obédé:This is where the modular payload architecture becomes important. Animals is not a single purpose device. It carries a comfortable suite of sensors, depending on what the plan needs to monitor. At Vigee, we are talking about depth cameras for spatial position and navigation, LiDAR for precise three d mapping, thermal imaging to detect its anomalies across equipment surfaces, acoustic sensors pick up the high frequency signatures of bearing degradation or compressed air leaks long before they are audible to the human ear and gas detectors for environment safety monitoring. So what makes the combination powerful is that these streams are fused.
Thierry Obédé:The robot is not just logging individual readings. The AI layer is correlating them, trending them over time and surfacing what's meaningful. A thermal reading alone is data, a thermal reading on a specific asset cross reference with acoustic data, trended against the last 13 patient cycle and flagged against the degradation threshold. That's intelligence. That's what animal delivers.
Thierry Obédé:And that's what changes the conversation from inspection rounds to asset intelligence.
David Bizley:Okay. Interesting. So obviously a lot of data is being collected then. How is this data managed and then made available to plant personnel?
Thierry Obédé:So this is the question I'm glad comes out because the robot is the visible part of the story. It's what people photograph or, and posts out, but the data layer is where the abolition value actually lies. And frankly, in my experience across industrial technology, this is where a lot of promising solutions have historically fallen short. You just can't do anything useful with it. So antibiotics approaches these very differently.
Thierry Obédé:It starts with a principle. The data has to flow to where decisions are made, not create a parallel system that someone has to remember to change. So the platform includes what we call antibiotics data navigator, essentially the operational intelligence hub where all inspection data are centralized, visualized and trained it. Plan personnel can see every reading from every inspection run track as a condition of the time and immediately identify what's changed since data cycle. So it's designed to be intuitive for maintenance teams, not for data science tool, or it has to be an operational tool.
Thierry Obédé:So the person doing the morning rounds can put it up and know exactly what animal flagged overnight and why. But more powerful element is what actually changes the maintenance workflow is direct integration with existing CMMS or ERP systems. This matters a lot because the last thing a maintenance team needs is another systems to manage. So what they need is for the information to arrive where they already work. So when anybody takes an anomaly, a thermal spike on a motor and acoustic signature consistent with bearing wear, that finding doesn't sit in the dashboard waiting to be noticed.
Thierry Obédé:It can be triggered to a work order into the customer CCMS or ERP systems automatically in the system. The maintenance teams already uses already trust, already reports from. So no manual data entry and no translation step between inspection finding and maintenance section. So it's the loop process itself that we bring to vision to some other customers.
David Bizley:Okay. And Robert, we've touched on this a little bit already, but how has the work of the robots practically changed the daily operation of your human personnel on-site? What would you say is the biggest before and after change that you've seen?
Robert Christinger:In fact, we have completely changed how the maintenance department works. Because we detect signals, we generate work orders from what was observed. Previously you had the older generation like me or even older which walk through the plant grease and dirt and every day. For us, we've got to get young people into our industry and keep them with us. We have to generate and make a workplace that is interesting, that people can see the you know, that we're living in a modern world.
Robert Christinger:And that's a that's one of the main reason why we wanted this robot, to keep the youngsters with us. They automatically have the enthusiasm to, oh, yeah. This could do this. They become in charge of the the system. They define which measurements are going to be done or I can ask them, don't you think we can check?
Robert Christinger:Are there if the fire extinguishers are still in place? Everything they can, they do on their own. And they are making a far more interesting workplace for these people. And that is one of the main things, advantages that I see. And that work orders are generated automatically that makes and is predictive.
Robert Christinger:So we don't just change motors because it's November. We change a motor because there is a problem. And we have, as you know, in cement plants, I estimate 1,500 rotating machines. So you do have to have systematics behind it that you don't lose control because there's so many trees and that the data platform helps you.
David Bizley:Excellent. Now sticking with the human element for a moment, one of the concerns that's been seen with the adoption of new technologies, specifically things like industrial AI, is that employees might feel that decisions are being taken out of their hands or even if their jobs are at risk. How have the personnel at Vigier reacted to their new robot colleague? Was there any skepticism at first? And if so, how do you manage this kind of thing?
Robert Christinger:Obviously, there are those that are skeptic. Maybe they've also got nothing to do with it because they work because they're working in a different department. Cement plants are large, you know. We try to promote communication. That's why I'm involved with Johnny, although I'm not in maintenance.
Robert Christinger:And we very well see when he detects problems if we fortunately, we have not ever had a leak of our fuel tanks that come by rail, but they are only inspected once a day. If the robot goes past there from time to time, it ever detects a spillage, this is a huge win for us that we can prevent problems from happening or a disaster if that ever would be.
David Bizley:Now, looking ahead, I've got a couple of final questions for each of you. Starting with you, Thierry, are we on the verge of a robot revolution? Can we expect to see autonomous robotic systems in some implants becoming widespread over the next decade, for example?
Thierry Obédé:Short answer is yes. So let me elaborate. So actually from my background, I've spent thirty years watching industrial technology goes through cycle of transformation and premature hype. I've seen ERP and CMMS change how plants manage assets. I've seen industrial IoT promise a connected plant future and finally take longer than expected to fully deliver.
Thierry Obédé:So I would like to bring here a certain rigor to the world revolution. So what's different with autonomous inspections is that the technology has crossed a threshold. Animal isn't a prototype or a pilot program anymore. It's a production systems running in real plants, in real conditions, generating real data that drives real maintenance decisions. So what we do with Viger is not an experiment.
Thierry Obédé:It's an operational model. And what happens when something works in production is that it becomes a reference and references accelerate adoption faster than any marketing campaign. So yes, I expect autonomous robotic inspections to become a standard capability in seven plants within this decade. Now, what does the future actually look like? And then a few things can come to our mind.
Thierry Obédé:First, animal itself will continue to evolve. We are on the verge of launching Animal X and I want to explain what these new announcements will provide to us as a significant improvement. Animal X would need the world's first and only like the robot to carry ATEX zone one certification. So for those less familiar, zone one certification means the robot is cleared to operate in environments with potentially explosive atmospheres. This is an entirely new category of access, area of cement and industrial facilities that are completely off limits to any of the autonomous systems are about to become inspectable.
Thierry Obédé:So that's not an incremental improvement. It's a new frontier for us to address new market differently.
David Bizley:Great. Excellent. Exciting times then.
Thierry Obédé:Indeed.
Robert Christinger:Yes. If only two years ago, I would never have imagined that this is going that we will be able to do this. Personally, I'm not afraid that people will have to leave here because we have a robot. We are very happy that we have a robot that replaces or helps replace those that have left. The decisions at least in our plant replace what should we do here.
Robert Christinger:This is all taken by those engineers that look after the robot and of course gives them far more authority and data to be able to decide. And we see things before it happens. So it makes a far more pleasant workplace and you can plan, you can arrange a lot of the jobs are not done by us, gearbox experts and that, but you can already schedule, I think, the next month this might break down. And if we stop the kill and say in two weeks, they could come in and do the job. So it's, I'm not saying it's magic and will replace everything, but it does give you a lot of advantages.
Robert Christinger:I cannot tell you the figure of how many hours downtime we have gained, but I would estimate this is if you do the calculation and you know how much downtime costs within the cement industry for us, it's a $56,000 US, then very quickly this is, of course, makes sense to have it.
David Bizley:Yeah. Excellent. And Robert, with now several years experience of working with these robot colleagues, what advice would you give to other cement producers who might be on the fence about adopting these new kinds of technologies?
Robert Christinger:My advice is not to wait myself also, until I realize that this this can do something. It's not the only system we have. We have online monitoring of of bearings and gearboxes. It's also an excellent system, but I never thought before that this would be possible. You know, we listen and we I know what Fourier transform is and all that, but this brings all this together.
Robert Christinger:And it's scalable. It's okay to measure a a machine once, but if you know that every day it will be inspected and they do not forget it, and he's not on holiday and so on, it gives you really a huge advantage. And I've not seen anyone being replaced. And to the contrary, keep people by having this.
David Bizley:Okay. Fantastic. And and Thierry, you know, what advice would you give?
Thierry Obédé:I see three things which are quite important. It has nothing to see with a sales pitch. First is don't underestimate change management. The robot is the easy part, but getting your organizations ready to trust it, use it to build new workflows around it. That's the real one.
Thierry Obédé:So you need to pay attention to operators, training, maintenance teams needs to understand what the data means and how to act on it. And critically, this needs visible sponsorship of plant leadership. If autonomous inspection is positioned as an IT project or a pilot, nobody would care about it. It will handle. So if it's championed from the top as a strategic shift in how the plants operate, transforms, yes.
Thierry Obédé:And then communication, training and management sponsorship are really essential for the adoption. Second is also thinking big, thinking fleet, not necessarily on the one plant. So the long term vision should be ambitious. Autonomous inspections should become a standard capability across all your operations and sites across a large company. So this is really quite important.
Thierry Obédé:And third is also to choose a vendor who treats your success as their success. So you have to ask them the odd questions about what happens after deployments. What does the support model look like in month six, in month 18? How frequently the software will be updated? How the robot will be maintained and so on and so forth.
Thierry Obédé:So at the Animetics, customer success is very, very strategic program, not a promise only. And that's build a long term partnership with our customers.
Robert Christinger:For me, you said a lot of very correct things, How to convince, and then once you get it, you also need the team that takes ownership, that looks after it. You buy any tool, which is the best tool, and then eventually you ask after two, three years, where is it now? But antibiotics and results that it gives us is in our daily it's what runs the show. In the meetings, technical meetings, they can report with a real KPI, not just made up figures, and figures that were derived by your measurements and your database. And this is, for us, a huge advantage.
David Bizley:Okay. Excellent. Well, thank you, guys. I think that's a great place to bring this episode to an end. So thank you for your insights into this fascinating crossover between robotics and the cement industry.
Robert Christinger:Been a great pleasure. Thank you very much.
David Bizley:As always, if you like this episode and want to hear more, make sure to rate, review, subscribe, and check out some other episodes from our ever expanding back catalog. Thanks everyone, and goodbye for now. I just wanted to take a moment to remind you to register for World Cement. It's free of charge and gives you access to the latest issues of World Cement, both in print and online. Every issue comes packed full of regional analysis, technical articles, project case studies, and the latest industry news.
David Bizley:Simply head over to worldcement.com, click the Magazine tab, and register today. It's as simple as that. Happy reading!