Transcript

The final hat I'll put on for today is as a technician on the production floor or in the shipyard. I've just closed my first task here of fitting the bit and now it's time to move on to my. Weld and you'll notice that as soon as we open this page, we actually kicked off this timer in the lower left hand corner. So we can now track cycle time of each individual task and each op card. We can use that to do all kinds of advanced analytics on, you know, quality issues or performance or what's a really clear work instruction versus what's a work instruction. Someone's gotta reread five times. So this is pretty helpful. Again, just building that data asset as we go to keep enriching our operating system with smarter and smarter decisions. Yeah, I like the feedback loop you're talking about here of getting that, getting that straight from the people on the factory floor, helping them accelerate, you know, through their work by having good clear instructions that are up to date. But also that feedback loop of when they're not, you know, people talk about it. My data needs to be perfect before I can do a I absolutely not actually. It should be imperfect with a good process for people to give feedback and correct and. And I think that's like. This is a great example of if you're really out on the factory floor doing this stuff and you're like, man, that was not clear because the person made this is sitting over here. I've done it like giving that feedback loop of whether it's in work instructions or hey, I have something's not compliant, right? I think those are the interesting parts here of like we truly are creating the UDA loop for manufacturing. Yeah, one of the coolest things I saw one of our teams build recently. I won't give away too many details, but basically they saw a repeating pattern in quality issues which they were able to extract with AIP. That a particular screw was getting installed facing the wrong way and it was causing all kinds of disruptions to the line and the screw wasn't getting caught until it had moved further into the assembly process. No one was like checking this one individual screw. And so they were able to just automatically make the recommendation as soon as they ingested all the data of hey, you should go fix your work instructions so that it says install the screw, you know, north-south instead of east. East, West or whatever. And you should also have someone check the screw when it's coming off of this production step a little bit earlier than when you're catching it. And something as small as that, you know, it kind of feels like a no brainer when you say it out loud. But no one would have the time to be parsing through thousands of these OP cards and thousands of these quality issues to actually be drawing those patterns. Yeah. I mean, it's it's really easy for me to sit here and say that in an office, but when you're on the factory floor trying to do that. It is a whole nother ball game of complexity and it's not always clear what people were intending. So I think that's that's cool that we can now give both sides the feedback loop and we can help find issues and correct them sooner. I think that's just the flywheel we're talking about with warp speed. Yeah, yeah. And, you know, really, we want to get folks on the floor pushing less paperwork and doing the stuff they're really uniquely good at, whether that's drilling holes or, you know, cutting these pieces or welding this bit. So that's been also kind of a fun goal for us along the way.