Solving Automotive’s Biggest Challenges Through Predictive Maintenance- with Mikhail Shaferman

Trend Detection Podcast ·

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.

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. And now your host. I am your host Senia, and in today's episode, we are going to explore how the automotive industry is navigating its most pressing challenges, from production complexity and labor shortages to unplanned downtime and sustainability demands. And joining us today is Mikhail Schafferman, sales vertical manager for Vertical automotive at Siemens. Mikhail, it's really great to have you here today. Can you start by telling us a bit about your old Siemens? Absolutely, Cenia. Thank you for having me. My focus is working closely with our automotive customers to make sure their production lines keep running day in and day out. In practice, that means helping them harness advanced digital technologies like IoT and edge computing to spot potential issues long before they cause a problem. Over the years I've seen just how transformative even small changes can be. A single new sensor or let's say analytic tool can radically improve up time and efficiency, and I'm truly excited to share some of this real world experience today. That is really great to hear, and I think those real world examples will be very valuable for our listeners. So perhaps let's start abroad. What do you see as the biggest operational pressures facing automotive manufacturers today? Unplanned downtime is still the single biggest pain point. And it's not just an operational concern, it's a business level risk. So you're saying it's not just a factory floor issue and it's really something that impacts the entire business. Exactly what's really striking is how it intersects with other challenges these days. Automakers are asked to do more with less managing, you know, high production targets despite labor shortages, volatile supply chains, and the constant pressure to keep costs down. And in the in that environment, every minute of stoppage can feel like a crisis, not just for the plant, but for the whole supply chain. Industry research puts the average cost of an hour of downtime at over €20,000 sometimes much more than you factor in logistics and penalties. That's why maintenance has moved from the back room to the boardroom, and it's now absolutely crucial for performance and competitiveness. Right, and I think that's exactly where predictive approaches come into play. And as more manufacturers go digital, yeah, predictive maintenance enters the conversation. But maybe let's discuss where exactly does it fit. Is it just a maintenance tool or a broader business level? It's a great question, and the answer is that predictive maintenance is so much more than just a tool. It's a really strategic lever for business. And of course the first goal is to catch and even predict failures before, you know, they happen, so you avoid unexpected downtime. And beyond avoiding breakdowns, what's the bigger picture benefit? The real power of predictive maintenance is how it links technical availability, which simply means machines are ready and working when you need them, with broader business KPIs. For example, we've seen companies cut unplanned downtime by up to 50%, which has line effects across quality, delivery times, and ultimately customer satisfaction. It's less about fixing what's broken and more about creating. A stable and high performance environment where teams can focus on making great vehicles rather than putting out fires. Right, that makes a lot of sense actually. But before we even get to prediction, I think there is one fundamental step before that, the data itself. So how big of a hurdle is connectivity in automotive plants? Connectivity is absolutely the foundation, and honestly it's often underestimated. You know, many automotive plants, especially ones with decades of history, rely on brownfield equipment, machines, and controllers that predate today's connected factory models. So now you're saying that just pulling reliable data out of those older systems is already a challenge. Exactly. The first real hurdle is simply getting secure, reliable data out of those machines. And at Siemens we work side by side with customers to bridge this gap. Using industrial age devices, secure gateways, and of course standardized protocols to ensure safe and seamless data collection. And once the data pipeline is established, we are able to feed predictive platform like Sensai, transforming a house of raw machine data into valuable, actionable insights. And I would add this isn't a purely technical step. The conversation about data ownership and cybersecurity is just as important. All right, let's summarize it. Connectivity is step 0 and predictive is step 1. And once both of that are in place, what makes Sensi predictive maintenance particularly suited for the automotive industry? You know, since I was purpose built with the realities of automotive in mind. First, it's cloud native and scales easily from a few critical machines up to entire fleets. Covers several plants, and that's unique. You know, is that it doesn't require users to manually set up thousands of rules. Instead, it leverages machine learning to understand each asset's normal behavior and automatically identifies anomalies. This is a huge benefit for multi-shift, high volume environment where, you know, where just isn't time or often the in-house expertise for endless programming. And there it's a conveyor on line one or a robotic arm in final assembly. Sensii integrates right alongside existing automation, so manufacturers gets the benefits of predictive insights without disruption. Yeah, and I think that is really important because ripping out old systems isn't usually an option, right? But let's say what happens if teams don't feel ready internally? What about plants that don't have data scientists or condition mentoring experts? That's a fantastic point, and quite honestly, it's a challenge that comes up along That's a fantastic point, and quite honestly, it's a challenge that comes up all the time. You know, not every plant has a team of data scientists or condition monitoring specialists. And that's perfectly OK. Our data enhanced service contracts are really designed with those customers in mind. With this approach, you don't just get access to powerful analytic software, you get a team of Siemens experts who partner with you. And they, these experts, handle the data analysis, flag early signs of failure, recommend concrete steps, and sometimes could even support implementation. For resource rich plants, this as a service partnership means you still get the predictive insights and proactive support without needing to build out a whole new department internally. Yeah, and that's definitely a major win for manufacturers who are already resource stretched. And now that more companies are scaling digital efforts, how well does predictive maintenance scale across plants, regions, or even globally? You know, scalability is one of the real game changers in digital maintenance. Meaning once you have solved it in one plan, you can actually reuse this model elsewhere. Yes, exactly. Once you have developed a predictive model for, say, a specific type of fan or robot, you can replicate it across identical assets at other locations, accelerating rollout dramatically and even better by aggregating health data across an entire fleet, which could mean thousands of machines in different countries. Automotive decision makers can benchmark performance and spot risks from a single centralized dashboard. Imagine being able to prioritize maintenance not just plant by plant but globally, aligning efforts to maximize up time and efficiency across all plants. That's the power of scalable, predictive maintenance. And they think that leads now perfectly into existing systems. How does this complement things like clean or TPM? Absolutely, it's really about amplifying what approaches like Lean and TPM already do well. Predictive maintenance, hence teams real-time data-driven insights, making TPM activities much more targeted and efficient. Instead of relying solely on periodic checks or even intuition, maintenance teams know exactly where to focus. And if a component starts showing abnormal vibration, for example, they can proactively interfere before failure. And this is not only boost technical availability, ensuring machines are up and running, but also frees up valuable time and resources by directing energy where it's actually needed. And the closer we get to final assembly or just in time delivery, the most I can ever. Oh, sorry. And the closer we get to final assembly or just in time delivery, the more every second matters, right? So how does predictive maintenance support those processes? Absolutely. Nowhere is timing more critical than in final assembly and just in time delivery. In those areas, even a 3-minute delay can throw off the entire flow, potentially leading to missed shipments or costly rework. And predictive maintenance empowers teams to anticipate issues. They conveyor motor trending toward failure, so repairs can be scheduled during plant breaks rather than disrupting the line. Can you perhaps share a real world example? Yeah, sure. In a recent case, we worked with a customer to identify and correct a bearing issue during an overnight window. Preventing what would have been a full morning's lost production. And that's the kind of precision and reliability just in time operations demand, and it's where predictive maintenance really shine. Yeah, that's definitely a critical point. And of course, as we know, companies also need to justify the investment, right? So is it ROI or risk-based? This is the heart of the business case. The financial ROI is clear lower downtime, less crap, better asset utilization, and very often, you know, reduced overtime and rush orders. But then there's also the risk angle, right? Correct. By catching problems before they cascade, you actually avoid last minute firefighting that can lead to missed targets or worse reputational damage. What's more, you know, teams become more confident. They know the data has their backs, which allows them to focus on continuous improvement rather than just holding the line. Yeah, I think that makes total sense. So Mikhail, let's perhaps close out by looking to the future. Where is predictive maintenance headed in the next 3 to 5 years in automotive? Looking ahead, we are already seeing the next evolution fully integrated, semi-autonomous maintenance. Where predictive systems will soon not just create alerts, but automatically generate work orders, order the replacement parts, and even slot maintenance into production schedules. For instance, if a key machine starts drifting from its performance baseline, the system can kick off a maintenance request, check spare parts inventory, and coordinate with operators without waiting for manual intervention. This level of integration means maintenance becomes both proactive and seamless, truly driving digital transformation in automotive. Wow, that was packed with insights. Thank you very much, Mikhail, for walking us through both the strategy and the practical side of predictive maintenance in automotive. Sure, thanks for having me. Always a pleasure to share how digital services are delivering real results. And to our listeners, thank you for joining us on the trend detection podcast. Be sure to check our SiMmens website for more stories, insights, and tools to help you on your digital journey. Thanks for tuning in and we'll see you next time. Please like and subscribe on all major podcast channels and find out more about Sensei predictive maintenance at Siemens.com.

Full transcript (WebVTT)