Cut Factory Maintenance Costs by 30 Percent의 전체 녹취록

the new version of AI offering assistance in this area? 00:02:25.759 --> 00:02:29.520 And how do you advise business leaders to use it? 00:02:30.560 --> 00:02:32.960 Well, here's where the world is at right now. 00:02:33.120 --> 00:02:35.199 Everyone is trying to mine their data. 00:02:35.280 --> 00:02:41.759 And so there's been a lot of good history out there in particular the asset management tools. 00:02:41.840 --> 00:02:45.599 And so now the challenge is how do I go and tap that information? 00:02:45.680 --> 00:02:55.680 And so, what a lot of our customers are trying to do is uh de-duplicate information, clean the information, grind get the right pattern so that AI can then tell them a couple of things. 00:02:55.840 --> 00:02:58.560 Uh, number one is what should I be doing with that asset? 00:02:58.639 --> 00:03:05.280 So making that decision of I'm getting some signals from different sorts of IoT devices, what does that really mean? 00:03:05.360 --> 00:03:09.360 So the history within the asset management tools will tell you that. 00:03:09.520 --> 00:03:17.599 And then uh going back to the experience gap, as the newer technicians go out and try to fix something, what should they actually be doing? 00:03:17.759 --> 00:03:21.520 Um, and so uh AI can help give that information to them. 00:03:21.680 --> 00:03:27.199 And so this is how AI is is being leveraged and starting to evolve uh to help replace that gap. 00:03:27.680 --> 00:03:37.759 Can you share how that is helping with the AI being connected to the Internet of Things and also some of the risks with that as well? 00:03:38.560 --> 00:03:39.680 Okay, yeah, great. 00:03:39.919 --> 00:03:51.439 So the smartest and the greatest uh technicians out there, I was actually joking with one of them, and uh, and they always say, Oh, I just have to put my hand on that, feel the vibration, and I can tell what I need to do there, right? 00:03:51.599 --> 00:03:56.639 Well, you take someone who's brand new, there's no way they know if it's even vibrating at that at that moment. 00:03:56.800 --> 00:04:00.400 Uh and so what does uh the connected you know uh IoT do? 00:04:00.560 --> 00:04:07.439 Uh well, there's a lot of companies out there that are helping understand vibration, temperature, you know, motor decrease, et cetera, power decrease. 00:04:07.599 --> 00:04:10.400 And all those are just signals. 00:04:10.879 --> 00:04:12.000 What does that mean? 00:04:12.159 --> 00:04:18.319 And so uh what AI can start helping do is understand what those signals could mean. 00:04:18.399 --> 00:04:24.480 Uh and so you're really starting to see, you know, that predictive maintenance get to the place where people want it. 00:04:24.639 --> 00:04:27.920 The biggest drawback here is to blindly trust it. 00:04:28.079 --> 00:04:31.920 And so then one thing is um it's getting really good. 00:04:32.240 --> 00:04:34.800 Everyone should definitely start investing in this way. 00:04:34.959 --> 00:04:42.879 However, we should continue to use the folks who are still with us, uh, still have the experience to pressure test and ensure that we got the right thing. 00:04:43.040 --> 00:04:57.839 So, what it is, what what I recommend for folks out there is is when you get something, take it seriously, understand it, but then go and understand that information, go and do whatever validation you need to, and then go confirm uh that what you are seeing through AI is indeed valid. 00:04:57.920 --> 00:05:00.879 Uh, it's a lot of, you know, the way we would use it to write something today. 00:05:00.959 --> 00:05:05.519 We would, we would, we would ask the prompt to help us think about how to how to be creative. 00:05:05.600 --> 00:05:09.120 And then we would say, okay, now I'm gonna read it and make sure it's what I need it to be. 00:05:09.199 --> 00:05:11.920 Uh similar to when you're doing reliability maintenance. 00:05:12.000 --> 00:05:16.319 Uh, we need to leverage AI uh to make sure that we're getting what we want out of it. 00:05:16.399 --> 00:05:17.920 So that's that's my advice. 00:05:18.240 --> 00:05:18.560 Okay. 00:05:18.879 --> 00:05:23.120 All the legacy technology that manufacturers are having to deal with right now. 00:05:23.279 --> 00:05:26.800 How are they moving to this new A, the new AI tools? 00:05:27.199 --> 00:05:42.079 Technology is going to be a slower, I mean, the these assets, these, these, these huge production lines that our food ab or any manufacturing is dealing with, they're starting to slowly start bringing in uh these manufacturers that will have IoT devices installed in them. 00:05:42.160 --> 00:05:43.920 And they'll come out of it this way. 00:05:44.079 --> 00:05:51.199 Uh, and so uh in one of the tours I was doing, they said, hey, let's go to our smart manufacturing and let's not let's not go to this way, right? 00:05:51.279 --> 00:05:59.040 So the the legacy ones will continue to need those inspections, we'll continue to need you know the monthly rounding, et cetera. 00:05:59.199 --> 00:06:11.279 Uh, but as uh items are being replaced, as you know, 20-year-old pieces of equipment uh come around, those manufacturers will start to embed that intelligent uh assets within them. 00:06:11.439 --> 00:06:18.560 And so then my recommendation is to start investing in those and ensuring that you're slowly building a modern manufacturing plant for yourself. 00:06:19.360 --> 00:06:22.720 Well, AI tools are advancing rapidly, as we mentioned. 00:06:22.959 --> 00:06:31.600 What do you envision as the future of manufacturing and how can business leaders prepare and harness these tools? 00:06:32.480 --> 00:06:37.360 Yeah, well, the future is slowly um being shown to us in different ways. 00:06:37.600 --> 00:06:41.120 And um uh, you know, take those same plants, for example. 00:06:41.360 --> 00:06:46.240 One of them needed a lot of human intervention uh within those manufacturing sites. 00:06:46.399 --> 00:06:50.399 The other ones was a lot of automation already

there. 00:06:50.720 --> 00:06:56.240 And what the future will be is us supervising that automation. 00:06:56.480 --> 00:07:05.120 Again, always having that human element in it because we're gonna get smarter about the way we uh we operate versus be completely replaced. 00:07:05.199 --> 00:07:14.240 And and so then the the future will be um let's let uh technology work for us, but let's always stay in the loop to ensure it's it's producing what we want. 00:07:14.319 --> 00:07:21.519 And so we're more uh in the in that control room, ensuring things are there so that when something pops up, we can then go intervene. 00:07:21.680 --> 00:07:26.480 We can go, we can then go make sure that uh asset doesn't stay downtime too long. 00:07:26.639 --> 00:07:29.439 Uh and then your second question is uh advice, right? 00:07:29.519 --> 00:07:31.120 That I would that I would give here. 00:07:31.199 --> 00:07:37.199 Uh and my advice is that is this is is to hey, you've got a rich set of data within you. 00:07:37.439 --> 00:07:40.959 Start ensuring that you're capturing all your information. 00:07:41.040 --> 00:07:57.519 And so those work orders that you're closing out are it's much more important that you take those notes, that you annotate what happened, that you get that problem cost and remedy in there, uh, and then clean that data and bring it into your data lakes because that information will be what will help you respond in the future much faster. 00:07:57.759 --> 00:08:06.639 So uh I just I just say now more than ever, ensure that uh you're you're getting that information within there, and then you can apply AI to get what you need from it. 00:08:07.759 --> 00:08:14.800 Well, from your experience, do you have a particular use case story that you can share about a company? 00:08:14.959 --> 00:08:25.920 Maybe you don't uh name the company, but is there um a transformation story you can share when they um harnessed the AI technology and IoT and they saw a great result? 00:08:27.040 --> 00:08:30.480 Yeah, I could think of uh automotive uh manufacturing. 00:08:30.560 --> 00:08:35.279 Uh, and of course, there's there's plenty in different industries, but we we we dealt with one of those. 00:08:35.440 --> 00:08:39.919 And they definitely uh saw a huge increase in productivity. 00:08:40.080 --> 00:08:48.159 Uh, but again, it's the reason why I think about the human in the loop is that this uh IoT device got them to the predictive maintenance that they wanted. 00:08:48.320 --> 00:08:58.159 So they were able to reduce their there's reactive and then this they were able to produce their preventive maintenance uh 30% uh because they were able to look at their predictive maintenance. 00:08:58.240 --> 00:09:04.080 And then they had every Tuesday, and this is advice for everyone, every Tuesday they would just look at that telemetry data. 00:09:04.240 --> 00:09:12.240 They would look at what the system was telling them to go do, uh, and then they would uh take out things that they knew weren't an issue and then focus on the things that were. 00:09:12.399 --> 00:09:14.960 And so they were able to see improve their uptime because of that. 00:09:15.200 --> 00:09:17.120 So uh this is why I'm a big believer. 00:09:17.200 --> 00:09:22.879 I was able to see it firsthand and how technology was being used uh for them to get the efficiency that they wanted. 00:09:23.200 --> 00:09:24.080 Wonderful. 00:09:24.320 --> 00:09:29.519 Well, if there was one key takeaway you could share with our audience today, what would that be? 00:09:29.919 --> 00:09:32.320 I said don't wait and start now. 00:09:32.480 --> 00:09:34.240 Uh, start investing in the small ways. 00:09:34.320 --> 00:09:36.320 Like I said, start investing in your information. 00:09:36.559 --> 00:09:46.159 Start investing, making sure you're getting that intelligence out there because when that experienced workforce leaves, all that information will be in your asset management system. 00:09:46.399 --> 00:09:55.120 Then you can plug in all the IoT devices uh and have that harmony that you want to have so that the next person that's more of a novice will have the right information. 00:09:55.279 --> 00:10:00.320 Imagine a world where um all they have to do is scan the asset in front of them. 00:10:00.480 --> 00:10:03.039 And this system can tell them, oh, great. 00:10:03.120 --> 00:10:04.960 Well, here's the history behind that asset. 00:10:05.039 --> 00:10:10.879 Here's what telemetry did is telling me I recommend you do X, Y, and Z and with these sort of tool sets. 00:10:10.960 --> 00:10:16.320 And so the more you start investing in your information now, the more it'll play dividends in the future. 00:10:16.720 --> 00:10:17.360 Awesome. 00:10:17.519 --> 00:10:20.480 Thank you so much for coming on the show and sharing your insights today. 00:10:20.639 --> 00:10:22.080 Of course, Amanda, thank you for having me. 00:10:22.159 --> 00:10:23.279 I'm excited about this. 00:10:23.440 --> 00:10:24.879 Yes, and thank you to our audience. 00:10:24.960 --> 00:10:28.799 If you have any questions or comments, link them below and I'll respond back as soon as possible. 00:10:28.960 --> 00:10:31.279 And meanwhile, have a wonderful week.