Podcast: AI best practices for manufacturers – Lessons learned from Ford and Girtz

Sean O’Meara is the chief information officer at Girtz Industries, a company that specializes in engineer-to-order power packaging solutions. At Girtz, he has championed the company's technological transformation and integration of advanced technologies with strategic business processes. Jason Ryska is the director of manufacturing technology development at Ford Motor Company. Sean and Jason recently spoke at IndustryWeek’s Operations Leadership Summit in June. During the panel, “AI's Potential for Manufacturing,” the two discussed how they successfully leveraged AI for their operations. In this episode of Great Question: A Manufacturing Podcast, Scott Achelpohl, managing editor of Smart Industry, shares the audience questions and expert answers from the session.

Below are some key quotes from the podcast:

Your org structure and the interactions within the company and how you manage that space is really, really important and critical. My peer at work every day is my IT group. My peer director and IT, we're on speed dial. We talk five times a day, and so there's a healthy tension there. We're supportive of one another, but what we also do in the projects, I said I have 75 people, but there are 250 plus people that come every day into my environment. They work together, so we collaborate and co-create with team members from the centralized IT department. So, we allow them in an environment to be a little bit unconstrained and creative. We're very conscious of what starts as pilots and a proof of concept, and we monitor it and navigate that space of when do point solutions become platform solutions?

And then me and my partner say, “OK, we need to scale this” or we're conscious about it. Yes, you can do this, but you can do it in these confines, in this plant, in this architecture. So we watch that space, but we manage it together very, very carefully. Again, it's that healthy tension. Yeah, the people feel like we tell them no very often, but we also tell them yes. It can't be yes to everything or you're going to have some real data security breaches and problems, or just sustainment and maintenance of fragmented systems all over the place is a nightmare. So, the org structure, the relationships, in creating an environment where you can co-create new things in the safe environment that's supported and you know when to pluck out things that can be scaled. So that's how we do it for you. It's definitely not perfect, but that tension is there, but be conscious and cognizant and manage that tension. – Jason Ryska

Over a while, I have found what I consider my tinkerers and my champions. So they're the ones that I roll that new technology out to first and foremost, and then they go, “Oh my gosh, that's so cool.” And they instantly start using it in their team to adopt it.

A good case in point is right before the pandemic, we implemented Zoom as a workspace communication platform. And just as we finished up the project, the world shut down, and Zoom was this major new product. Well, now we're rolling out Slack because Slack has all of these great AI features that interface with everything else that we do, and I've got my team saying, “We're not switching. We're doing too much work in Zoom. I'm not going to move over to Slack.” And I said, “No, no, no. You don't have to. Don't worry about it. By the way, do you realize Slack is summarizing our meeting notes? By the way, do you know, Slack AI is looking through your CRM?” And now I've got people pounding on my door saying, “When are you rolling slack out to our department?”

In my role, I do a lot of misdirection and selling value, so I don't want them looking at Microsoft Teams, so I better have a better solution than Microsoft Teams ready for them and selling it to them before I get some guy coming in and saying, “Hey, look what I can do with Microsoft Teams.” And then in the meanwhile, for us, big time is tying that data all together. I want one stream of data, one single pane of view for everybody in my department or everybody in my organization. So they don't want to use another tool that isn't tied in to that data. – Sean O'Meara

I think we over constrained ourselves in the very beginning. I said here's a robotics team, and there's an industrial controls team, and here you’ve got vision expertise, and you are data analytics, and you're an IT person, and you're an OT person. And we had these functional groups. There were five of them at the time. But the problems that I have to solve aren't that clean. It just doesn't work, and we said, “OK, this is mostly robotic. So you're going to get it.” But guess what? It doesn't matter what the problem statement is. What we're going after, it's a collection of all those different skill sets that come together now. It is truly cross functional. There are no mechanical engineers, there are no electrical engineers, there are no computer scientists. It's people that have a broad-based engineering skill set that can go do stuff.

So, we set up agile project teams, and we manage the workload within the team. There's an org structure. If you go on it, it's very neat, and it looks like we know where everybody reports to and who does their performance for you and stuff. But problem statement comes in, and then we look at the skill sets that are required to solve it, and we put people on that team. And we make sure that we don't overload the hell out of one team, but people actively and very much so move around. When you look at that team that was up there five years ago, we would never put a team together like that because they're not all electrical computer scientists with 15 years of industrial controls. These are people, like we heard earlier today, that have passion. They have curiosity. They have an incredible amount of energy. They don't know why they can't do things. They have a solid foundation in engineering principles and problem solving. So we stick them in there and put problem statements in front of them, and we actively move the team on a regular basis. – Jason Ryska

About the Podcast
Great Question: A Manufacturing Podcast offers news and information for the people who make, store and move things and those who manage and maintain the facilities where that work gets done. Manufacturers from chemical producers to automakers to machine shops can listen for critical insights into the technologies, economic conditions and best practices that can influence how to best run facilities to reach operational excellence.