Filip Kramer - Taking Ownership, Maintenance Strategies, and Driving Innovation
A podcast with a wizard who shares so many gems
Intro
Welcome to our latest podcast episode featuring Filip Kramer. At the time of recording this podcast, Filip was the National Reliability Maintenance Engineering Manager at Amazon. He his a true visionary in the field of industrial engineering. In this engaging conversation, I sit down with Filip to explore his extraordinary journey from a machinist in the Czech Republic to leading one of the most advanced reliability engineering teams at Amazon. Filip shares invaluable insights into his career progression, highlighting key moments of learning and innovation that have defined his path.
Filip's story is not just about career advancement but also about pioneering change in the industry. He discusses the challenges and opportunities in transitioning from reactive to predictive maintenance, the importance of building a vision and roadmap, and the critical role of cross-team collaboration.
Listeners will gain practical advice on improving operational efficiency, enhancing team communication, and leveraging data for proactive maintenance strategies. Whether you are an aspiring engineer or a seasoned professional, this episode offers a wealth of knowledge and inspiration to drive your own career and organisational success.
Podcast Links
Spotify:
Apple:
Some of our Favourite Parts
Filip shares so many gems in this podcast, some of our favourite parts where;
- His Career Journey and Management Philosophy (00:02:11): Filip shares his unique career journey from a machinist to a senior manager at Amazon, emphasising the importance of hiring and developing a team that surpasses the manager's own abilities.
- Transitioning to Predictive Maintenance (00:22:47): Filip details his first implementation of predictive maintenance, explaining their simple approach to accurately predicting equipment failures and improving maintenance windows, highlighting the power of data-driven decision-making, and the impact predictive maintenance can have on sites.
- Cross-Team Collaboration (00:33:27): Filip discusses the integration of operations and maintenance teams, fostering better communication and collaboration, which significantly improved troubleshooting efficiency, operating performance and overall team performance.
- Mentorship and Learning (00:44:24): Filip offers valuable advice to young engineers, stressing the importance of getting hands-on experience, asking questions, and learning from seasoned professionals to build a successful career in engineering.
Full Episode Transcript
Disclaimer: This transcript has been edited for clarity. Timestamps may not perfectly align with the content, and guest dialogue may have been refined for accuracy.
Time: 00:00:03 JP Picard (Factory AI): Filip. Welcome to the show.
Time: 00:00:06 Filip Kramer (Amazon): Well, thank you. It is my pleasure to be here and share the journey and all things around.
Time: 00:00:17 JP Picard (Factory AI): That's good because it's also my pleasure. I was telling you a little bit about this off air, but I found you on LinkedIn as I find a lot of people and thought, Wow, there's gotta be a lot of good stories there.
Time: 00:00:29 JP Picard (Factory AI): And thought, Wow, there's gotta be a lot of good stories there.
Time: 00:00:34 Filip Kramer (Amazon): Yeah, and weird stories. Yes, I can promise.
Time: 00:00:38 JP Picard (Factory AI): Ok, cool. Well, a place where I'd like to start, actually. Hopefully this doesn't cause too much blushing, but one of the things that interested me. You had a lot of recommendations. This is not typical of someone highly skilled. There's specifically very little on LinkedIn. You had a lot of recommendations. I particularly liked one of them. I'll read it out. Maybe this could offer you some things to comment on a bit later. But what I want you to do is guess who is the person that wrote this first after I'm done. Ok, this recommendation reads I had Philip as a manager at Amazon in PR G two, which I think means Prague. From the moment we started working together, his know-how and technological skills exceeded the rest of the team at the same time. His managerial skills were and still are a great inspiration. His empathy, problem-solving style, and ability to lead is very inspiring. In some ways, Philip is able to see into the future and anticipate the needs and abilities of his team very well and thus lead the whole team very effectively. Very few managers have this skill. So who do you think said that about you?
Time: 00:01:50 Filip Kramer (Amazon): Oh, it can be Ivan, my senior automation engineer in PR G two. Or it can be Abatis, the senior planner in PR G two. One of those.
Time: 00:02:03 JP Picard (Factory AI): Very well. You were right on the second guess.
Time: 00:02:06 Filip Kramer (Amazon): How about this? I'm saying that because normally, when you're working as a manager in engineering, your main target is to hire a team which is better than you. And if the team is not better than you, then you need to develop the team so they are getting better than you. Then you can send them on to the world. Actually, with this, it was a really nice story because he worked in operations and he came to me and I was like, Yeah, I want to work in your team. I want to learn about reliability and maintenance. He had a business background and I said, OK, let's start with planning. I set up a learning process for him. But my management style, in terms of learning, is more to drop them into the water and swim. If you don't know, ask. But if you are swimming in the right direction, we will not touch you. It can be really stressful, but you are learning. Humans learn by mistakes. Back in my career, this was an eye-opening thing when I set my technician or my engineer onto the process, and then the engineer came to me and said I failed and I said, You didn't fail. You learned. That's the point, right? So yeah, that was about this. Actually, he's working at the moment in another environment. We are still in touch with the majority of my team. He used to work in Dematic, who was the main conveyor N MA G supplier for Amazon in Europe. Then he moved to another company, and now he's progressing really well in his career. He's probably moving back to Dematic.
Time: 00:04:05 JP Picard (Factory AI): Well done. Sounds like you passed on some good lessons to him. You can tell that just by reading his tagline on LinkedIn, which reads, Don't let passable or ok be your standard. No matter what you do, do it the best, which seems to have been influenced a little bit by you so well done. I like that you told this story, by the way. It'd be good for you to tell a little bit more about your story. So what I gather is you come from having been a machinist in the Czech Republic, and now you made it all the way to National Reliability Maintenance Engineering Manager for Amazon. I suspect there are a lot of intentional but also unintentional choices and a lot of lessons learned. Do you want to tell us more?
Time: 00:04:52 Filip Kramer (Amazon): Yeah. So completely back to the story. I went to engineering school. In Europe, when you're 15, you have to decide the direction of your career, and then you are going to school which builds your specialization or you are just going to general school, but that means you have to go to university. In that moment, my mom dreamed that I would be a doctor. The teacher from elementary school, the teacher of physics from elementary school, jammed on a bicycle and went to our home to beg my mom to put me into technical school because I loved physics and mathematics. I was just 15 years old, and my mom told me, What about engineering? I said, Yeah, that's a good idea. So I completely changed my direction and went to engineering school. I somehow started to like it, and then during the studies, I built a plan to become a designer, an industrial designer, and do all those mathematical calculations, material calculations, and everything. I really loved the descriptive geometry and drawings. So I started to think about my career in this direction. But when I finished school, I told myself, You can be an industrial designer if you don't know how to make those parts and how to create those parts. We are saying the paper can survive anything, so you can design anything on paper, but in reality, it can be a massive fail. So I started my career as a machinist and as a CNC machinist. As a young boy with an open mind, I learned how to program in hand code and ISO code. It was a really great opportunity because CNC machining didn't just focus on machining, but it was like a medium-sized tool shop. We produced the parts according to the drawing. As a CNC machinist, you receive a drawing, you receive material, and it was on your own to choose the correct tools, program the part, and then run the program. It was full responsibility. You didn't have programmers or technologies in the background. It was a really good school for me, and I learned a lot. After a couple of years of this, I changed companies and started EDM, wire cutting, and EDM machining for precise parts for tool shops. I decided to try the manual part. I did a lot of assembling of the tools and testing of the tools. It was cutting tools and more tools. After that, I became a designer and technologist in one tool shop. Then I figured out how bored I was. It was a repetitive job, designing the tools and just putting modules together. I decided to change my career. I moved to a company called Edwards. At that time, it was BOC. Now it's part of Atlas Copco. It was a really good career move. In that company, when I started, they produced mechanical blowers for gas transportation. After one year, we completely changed the business to vacuum. From a technology point of view, we had to switch our minds and learn everything about the vacuum. For example, if you are putting a flange on a pipe in a pressure system, you have to weld it from the outside. If you're doing a vacuum, you have to weld it from the inside because the direction of pressure is completely different. Same for O rings. You put O rings on the inner part in a pressure system, but for a vacuum, you put O rings on the outer part. It was small details. In that moment, I had a really good manager. He figured out that I had engineering potential. I got projects. He came to me and said, Hey, we have new vacuum blowers, like vacuum pumps, and we need to test them. So design the testing area. I had a free hand to design and play. He told me, You need to cool down the blowers. So I bought radiators from cars and connected them with pipes and pumps. I bought a pump in a shop. Then I called my boss and said, Look, we are cooling down. I did calculations. I learned everything. It was a really good assignment. But my mindset is, I need to build. I have to build and move forward. Then it started to be standardised and in shape. I decided to move and try a management career. I moved to another company, Flowerf. It was a really good career move. I moved there as an engineering manager for a small team of three programmers. They sent me to the Netherlands, and my second nationality is Dutch, so I'm half Dutch. They sent me to the Netherlands for training. The manager of the Dutch team got ill. They told me, We need a manager in the Netherlands because we have a team of 20, a big, massive production plant. We don't need a manager in Czech Republic. You can manage the Czech part remotely. From one week of training, it became 1.5 years in the Netherlands. I grew as a manager a lot because I was still responsible for the team in Czech Republic. I had to manage the team remotely. I traveled to Czech, set the processes, and everything. When my colleague in the Netherlands came back, I moved to Czech. Then I figured out they don't need a manager because I set the processes in a way that everything runs automatically. I moved to the main competitors and worked again in the CNC industry. Then I touched a couple of other industries. One big company, a massive corporate, called me and asked me to be the Reliability Maintenance Engineering Manager there. I rejected the offer because I said, No, I'm not a Reliability Maintenance Engineering Manager. I'm focused on machining. I like creating good things in the machining industry, working with teams, and doing precise things. What am I supposed to do in maintenance? Then the director of the European team called me and said, Listen, I'm going around our sites and want to show you the sites. I said, Yeah, that sounds like a great opportunity. I can see how the company works. We did a trip around seven sites. He showed me the level of automation, how it's working, the opportunities, the gaps, and the issues they face in maintenance. We went to the last site, PR G two in Prague. In that PR G two site, he showed me the problems they are facing, how many callouts from operations they have, the cultural issues, and everything. At the end of the trip, he told me, If you want and sign a contract, this will be your site. I said, Ok, show me the contract. I didn't ask about money, nothing. Just, I'm going to sign because it was fun. As I mentioned, I'm a creative person. I'm not a settler. Back in Europe, in the team, they called me Philip the Firefighter. If you have problems and need to set up everything in order, call me. I love that kind of environment. I love setting a structure, putting it in lines. Once it's running, I'm losing interest. I'm happy to hand it over to my team members because they have already overgrown me. So that's it.
Time: 00:15:26 JP Picard (Factory AI): Well, what a story. There's so much to unpack there. So many things I can comment on. I do like that in many of those moves that you made, you said this was a really good career move. To others, it may not have sounded like it was because you've described something quite problematic and challenging. But you saw it as an opportunity, which I absolutely love. And that's where you earned your title, Philip the Firefighter. I'm pretty sure as well, by the way, I don't know if you know this, but Reid Hoffman, who founded LinkedIn, also earned himself the title of Chief Firefighting Officer of PayPal. Some pretty impressive people have the chief firefighter as a nickname. That's fantastic. This big company you talk about, I wonder if people know about it because a lot of my paycheck goes to this big company. Maybe one question that's on my mind and the minds of others is you've worked in Europe and here. You have quite a bit of international experience now. With that, what are the differences or cultural differences you've observed between these places like Czech Republic and Australia, specifically in manufacturing?
Time: 00:16:59 Filip Kramer (Amazon): When I look at the manufacturing industry in Australia, it's unfortunately not running as well as in Europe because you have automotive. Everyone is producing everything. Most of the stuff in Australia is imported because Australia focuses on mining. You are producing mining equipment in Australia. When I look around for the industry in manufacturing, it's not as massive as in Europe. This comes hand in hand with one thing: you have lots of people. In Europe, we don't have enough engineers, and we have a lack of engineers. European education system prepares you for a job. When you are studying after 15, you are studying one week and working in some factory, learning from experienced workers. It's like an apprenticeship, but also focused on knowledge. In Australia, this is somehow running, but you have a bigger lack of engineers than in Europe. People in Europe are still transferring knowledge. When I hired a team in Australia, I hired around 120 people. The majority of them are imported because there is a gap in knowledge. It's probably because of mining, and many people are in the mining industry. Other industries are booming, and historically, you lost some industries like Holden when they closed factories. You started to lose those skills. Now Australia is coming back to manufacturing. Australian companies are figuring out they have to manufacture in Australia, but the labour is expensive. That's one thing. Also, there is a lack of knowledge and experienced people. It's not just Australia; it's also in Europe. But in Australia, it's more visible because the market is smaller.
Time: 00:19:49 JP Picard (Factory AI): It's such a good point. That's typical of an industry regaining momentum without the skill set to support it. What you talk about is one of the biggest problems that reliability leaders always mention. They have plans but claim they don't have the people to execute them. This leads well to the next thing we wanted to talk about. One of the things I wanted to explore with you, because you would have undoubtedly combated this over your years, is the typical firefighting reactive mode. It sounds like you probably thrive in that environment. Maybe that's why you've been good at solving it. It would be good to explore this because many people we speak with in reliability are stuck in reactive mode. They're open about it. They tell us they're in reactive mode. What has been your experience with this, and what have you learned about transitioning out of reactive mode?
Time: 00:20:59 Filip Kramer (Amazon): I hate reactive maintenance. I hate anything reactive at work. I'm a Six Sigma Black Belt. I'm focused on data collection. The best thing about maintenance is you have heaps of data. You just have to learn how to work with them. You have quick wins you need to do. Reactive maintenance is the most expensive thing in any plant. Maintenance is expensive because you have labour, breakdowns, equipment stoppages, and spare parts. Spare parts are expensive. You are keeping another set of equipment which is doing nothing. Reactive maintenance is the worst. Preventive maintenance is not great either. It tells you to change a bearing every year, but why change it if it's still in good condition? Then you have predictive maintenance. When I first implemented predictive maintenance, it was in PR G two building. I remember telling the general manager we would have a failure in three weeks and needed a 1.5-hour maintenance window to fix it. He looked at me like, What are you talking about? I told him, because I'm an engineer and engineers are wizards. We can see the future. We really can. We just did. The moment changed my mind on where we had to go. Preventive maintenance and shutdowns were the norm. But we started with a meeting, and one of the technicians stopped the meeting and said, The sound of the building changed. Something is wrong. I realized we knew the sound of the building. We know how it normally sounds. I pulled one technician out and created the role of predictive maintenance technician. I removed all tools from that person. My second predictive maintenance technician was female, and she was even more sensitive. They just walked around the building, listening, checking temperatures, vibrations, and touching stuff. They knew the equipment. We created a special work order in our CMMS system with the prefix BN for abnormality. They wrote down those work orders. There is something abnormal in the system. Please check. That was the first stage. I had a good experience with an ultrasound microphone. If you have compressed air or electricity in your building, this toy pays back in one day. Compressed air is the most expensive medium. You spend a lot of money on electricity to generate compressed air. With the ultrasound gun, you walk around the building, listen for leaks, fix them, and save a lot of money. This is the cost of the device. We used this and started to set a baseline, listening to bearings and doing frequency analysis. If you have a standard electric motor spinning at 60 Hz, and you hear noise at 60 Hz, something hits on each spin. If a bearing has ten balls and you hear noise at 600 Hz, it's telling you there is a faulty bearing. You can identify and narrow it down. We set the measuring points and did daily routine walks, measuring. We had the benefit of affording stoppages. I said, Let's run to fail and set a baseline. We got the curves and trends. We knew when failures would occur. We started to measure temperature. The last step was installing small modules. One automation engineer designed a module to measure vibrations and temperature. We started to collect data online. Now we have an app that triggers alarms, sets it to CMMS, and generates work orders. This motor has a vibration or higher temperature. Go and check. It's super easy. The return on investment is massive. You change the bearing when needed, reducing warehouse costs. You see the future. If you know a failure is in two weeks, you have time to bring parts to the site. People call it Industry 4.0. It's nothing complex or complicated. We have this in place. For example, if you have some production line using automation, you have PLCs. PLCs have data. You just need to read the data and start working with them. If you're running a conveyor, measure the voltage on a PLC. If bearings are getting old, the motor load increases, and the voltage rises. The system tells you something is wrong. Fix me.
Time: 00:30:03 JP Picard (Factory AI): That's amazing. What you've described forms a lot of our company's vision, by the way. Your story was absolutely amazing. I took a lot of mental notes from it. The first one is to start simple and small. You described one person rebranded as a predictive maintenance technician, not tons of tools. Sensors may not be necessary. The PLC collects that data anyway. Start small, not tons of tools necessary. Maybe some common pushback is that you need to go through the natural cycles of reactive to preventive. It sounds like you disagree. Do you?
Time: 00:31:06 Filip Kramer (Amazon): You can go from reactive to preventive, but preventive phase. If you feel you have a structure, you can touch the preventive phase. But you don't have to finalise it. You can go straight to predictive. You have lots of low-hanging fruit. But you need to build a roadmap. You can't move without it. You can skip steps. Every company is individual. You can't take one template and copy-paste it everywhere. What helped me are two crucial things: get your team involved and change management in the team. Instead of telling how we are going to do that, I started to ask what we are going to do. We started with a vision. In PR G and in Germany, I sat down with the team and said, Let's build a vision. Everyone said, We want to be the best in maintenance and bring Industry 4.0. That's not a vision. I want one sentence defining us for the next five years. It took us one day, but we ended up with the sentence, Be a trendsetter. Let's not ask people what we are going to do. Let's be revolutionary. Let's build something and show everyone. Let's be the company that is a pioneer. Another part is selling that to your stakeholders. It was complicated. Actually, it was a crazy idea that really helped. Two crazy ideas. First, I negotiated with the general manager, and we put the operations management team at one desk with the maintenance team. They had a maintenance control centre and an operations control centre, and we put them together. They weren't sitting at one desk. Instead of us and them, it changed to us. We are one team. We are talking every day and drinking coffee together. When there was an issue, the maintenance team would say it would be fixed in 15 minutes. Operations would ask after 15 minutes, and the maintenance team would say 15 minutes. When they started to sit together, they communicated better. We figured out that troubleshooting took longer than the repair itself. The repair took seven minutes, but troubleshooting took eight hours. We had a main line down for eight hours because of troubleshooting. We started to ask how critical the timeframe was for operations. They told us 45 minutes. We answered, OK, ask us in 40 minutes. We will be smarter. Leave us for 40 minutes to do our job. We started Engineering 101, a classroom for operations. All operations managers attended. It extended to onboarding for operations managers and directors. We helped operations and other teams understand what we are doing. It improved communication. Then we asked other teams to prepare something for us. We wanted to see under their hands. It helped us understand their terminology and process. We started to run projects together, communicating and moving forward as one team. It wasn't operations, engineering, or IT teams, but one team. It worked really well. Technicians understood operations and IT terminology and could put their hands on the process. We started an initiative where technicians and engineering management went onto the process for one hour weekly. We found small issues workers learned to live with. We started to fix them, and it led to small projects. It was part of the continuous improvement. It opened our eyes to the process and issues.
Time: 00:39:17 JP Picard (Factory AI): That's so cool. It's crazy how small changes, like moving people around, can impact. It led to cross-team project creation, which sounds amazing. I'm sure companies that do this reap a lot of rewards. You gave many examples of innovation. If you narrowed down your innovation process or roadmap to a few simple steps, what would you call these steps?
Time: 00:40:21 Filip Kramer (Amazon): I came from the automotive industry with a Six Sigma and lean background. I'm using PDCS PQ CD: safety first, people, quality, cost, delivery. We started with a vision, then broke it down under SPQ CD. This is a roadmap for every year. How to improve cost in reliability and maintenance engineering: reduce your warehouse, spare parts, increase uptime, reduce maintenance time. Safety means coming to work with 20 fingers and leaving with 20 fingers. Happy technicians are important. They want to be busy and work in a modern environment. These small things narrow down to your roadmap. My strategy and recommendation to the team: if you have one point in a roadmap, close it in three months, max. If you're working on something more than three months, you lose interest. Narrow it down to parts you can resolve in two or three months, max, then move to another.
Time: 00:43:00 JP Picard (Factory AI): I like that. There's this concept from journalism that whatever the deadline, you'll meet it. You're extending it to the time required for innovation. I like that a lot. Last thing I wanted to cover with you. In doing my research on you, I found out that you do some mentoring. You're at a point in your career where you can help a lot of people. There will be some younger leaders listening to this podcast. What would you give as advice to those young engineers looking to climb up the ranks?
Time: 00:44:24 Filip Kramer (Amazon): I hope we are going to have more engineers. Keep your eyes open. Don't be scared to get dirty. Even if you finish university and think you're the best engineering manager, take a spanner. Go to the field. There's a lot to learn. Ask experienced people. Be dumb. Even if you know the answer, pretend you don't. It opens doors. All engineers want to mentor and share information. But if you come off as arrogant, they won't share. When I was a machinist, there was an old guy working on a grinding machine. He was retired, experienced, and always yelled at us. I brought a strategy. I came to him and told him complete nonsense. He got pissed off, yelled at me, but explained in detail what he was doing. He taught me how to measure on a microscope and measure super precise things. It helped me in my future career. I faced an issue about the flatness of the surface and figured out how to measure it based on his knowledge. Have open eyes. Ask questions. Don't be afraid. There's no stupid question. It gives you many more answers. Roll up your sleeves and go to the field. It's not just for young engineers; it's for everyone.
Time: 00:47:10 JP Picard (Factory AI): It sounds like that machinist you talked about took that advice to heart because he was still working on building parts post-retirement. That's commendable. Philip, you've been amazing. You've been generous with your knowledge. This conversation was really fun and valuable. Thank you for having it. If there's an opportunity to reconnect in a year or so and hear what you've been up to, that would be amazing.
Time: 00:47:45 Filip Kramer (Amazon): Absolutely. It was my pleasure. It was a really nice chat today, and I hope it will be useful for people around.
Time: 00:47:56 JP Picard (Factory AI): I'm sure it will be. Thanks very much.
Post Show Notes
In this episode, Filip discusses his experiences with predictive maintenance and the significant benefits it has brought to his career.
If you're interested in learning more about predictive maintenance, we'd love to help.
At Factory AI, we've assisted sites across Australia in transitioning their reliability and maintenance strategies to predictive maintenance with our solution. We've worked closely with our customers to design a predictive maintenance software that is accurate, easy to use, and cost-effective.