SMT007 Magazine

SMT007-Nov2024

Issue link: https://iconnect007.uberflip.com/i/1528798

Contents of this Issue

Navigation

Page 17 of 85

18 SMT007 MAGAZINE I NOVEMBER 2024 a little guidance. For exam- ple, "You're a process engi- neer; what are you see- ing?" Now, having a per- son in a control room and staring at the dashboard all day isn't needed anymore. Have the AI agent do that. When those dashboards show a problem, open a ticket and file it directly to the process engineer who should resolve it. Oh, and include the synthesis of everything that you learned from all those dashboards so that they know exactly what to do. Interesting that the thinking on how to apply this data has shifted so significantly. The end user for the data analytics may not actually be human. There may be one step in cooking the data before it's ready for analytics. Jennifer Davis: We started talking about this two weeks aer APEX EXPO. It's been incred- ibly well received by all customers and part- ners. is underscores the pace that AI is tak- ing. But it has to be supported by a faithful data foundation. at's the fuel. e better the fuel, the better the AI use case. If the dashboard was a manual transmission, you have just painted the picture for an automatic transmission. Davis: When Tim started to talk about dash- boards, Nolan called it the death of a dash- board. It's true, but yet it's funny because it will change what the customers think they need when they have a solution that is architected by data experts but co-piloted with domain expertise. You can have all the dashboards and data, and have it packaged up the right way. It's still too hard for customers to make use of it as efficiently and quickly as they need to without that domain expertise doing the co-piloting. Now, you can have an AI agent sitting along- side that domain exper- tise, and everything is just incredibly faster. Who do you envision using this AI interface into the data? B u r k e : Fo r t h e p r i m a r y focus I'll highlight two of them. First, it's the front- line supervisors: line super- visors to shi supervisors. ese people are account- able for meeting the produc- tion plan, shi over shi. It's a superpower for them. ey can see exactly where the problems are and take action right now. Second is the offline support technicians and engineers being called to the line when there's a problem. It's always urgent when the line is down or there's a quality issue. How quickly are you able to resolve that emergency? What's your mean time to repair? With the AI tool, the calls to them have more information. It's a much better support tool. Here's everything I need to know in terms of what I actually did, how I fixed it before, guidance on how I might resolve it, and the ability to collect SOPs in a way that people can use them for training and upskilling your more junior technicians. ere are so many knock-on benefits. That makes a lot of sense, especially if you're on the second or third shift and all your process engineers are on the day shift. Burke: Yeah. At the end of the day, if you're working three shis and you fix any one of those shis, that's a third of your day. You can't just throw efficiency away and say, "Oh, I'll only run half as good at night." You need to run well all the time. Tim and Jennifer, thanks so much for the information. Burke: ank you, Nolan. SMT007 Jennifer Davis

Articles in this issue

Archives of this issue

view archives of SMT007 Magazine - SMT007-Nov2024