In Stay Sharp Episode 137, CIMdata PLM/PDT Road Map Recap, co-hosts Juliann Grant and Jonathan Scott talk with Andrew Halley, Razorleaf Global Partnership and Alliances leader. Together, they’re recapping the CIMdata PLM Road Map & PDT North America conference that was held in early May outside of Washington D.C. A second conference will also be held later this year in Sweden.
About the Event
Andrew points out that while most content revolved around AI—the event theme was AI in PLM: A Disruptive Opportunity and Challenge—it wasn’t “the traditional AI conversations that we’ve been hearing everywhere. It was the conversations that everybody’s afraid to say out loud.” In keeping with that frank talk, a couple analogies for AI made the rounds. Juliann mentions hearing “AI is like having a husband, meaning that they always have confident answers, but they’re only right half of the time.” And Andrew relates the comparison of AI to a 20-year-old with an amazing memory but no experience—there’s plenty of great data but no intent behind it. Juliann agrees, “The general comment in that area was, there’s a lot of readiness that needs to happen. A lot of data preparation, a lot of conversations around how some companies have done that.”
Andrew Halley, Juliann Grant, Jonathan Scott, Razorleaf
Takeaways
“Fail fast” was a presenter’s take on AI they all appreciated. “We could sit there and wait and try to get everything perfect to move forward with utilizing AI, or we could try it and see what works and work on the process as we go,” Andrew says. Engineering’s typical analysis—often leading to paralysis—process won’t work when it comes to AI. He adds, “It’s moving too fast, and if you try to go that approach, you’re going to sit there for the next fifteen years and never move anywhere.” As Jonathan says, “Data quality is critical, but don’t let it stop you.”
Community and Structure
Jonathan Scott, Razorleaf
The conference spanned two days and included a vendor area in addition to presentations from headliners, industry leaders, and vendors, including Razorleaf. Jonathan described the different types of sessions, starting with those who are observing the industry, “Coming back to, how’s PLM going these days and what are vendors doing with AI and PLM? Connecting back to what’s actually happening for a lot of people and what they see.” Others represented industry players who are involved in process and standards groups or organizations, as well as members of governmental agencies. Those leaders, he says, “have a different angle on it, have different constraints to deal with, but it’s so many different perspectives. What’s so interesting about the event is the presenters aren’t all in the same boat, so you’re hearing all these different angles on the topic.”
One highlight of the event for the group related back to the question of the intent behind AI use. The speaker from MIT Lincoln Laboratory described questioning why she was being asked for specific data, making the point that we shouldn’t be spending resources—even if it’s AI—on work that isn’t focused on the goal we’re trying to achieve. Diego Tambourini from CIMdata underscored that idea, noting that employees are getting lots of pressure from executives and leadership to do something with AI, but many of them are asking what they’re supposed to do with it or what’s the use case. Jonathan agrees, “Give me the specific thing we’re trying to solve, and I can go work on it. But without that, there’s going to be a whole lot of wasted effort.”
Andrew’s biggest takeaway involved the gap between big PLM vendors offering monolithic, do-it-all systems and customers who want systems they can integrate with their existing software and structures. He says, “Our customers want best of breed. That’s what they’re asking for. They want inoperability and they want it as easy as they can get it, but they also know that they have to have a very strong, sound knowledge base of data to use a lot of these best-of-breed vendors.” Jonathan brings up AI and the digital thread as his key takeaway. AI won’t create a digital thread for you or fix a broken one. “It’s going to help,” he says, “but you have to think about how you apply it.”
Learn More About the Event
The full podcast contains more discussion and details on a variety of topics, including why AI isn’t going to give you the perfect digital thread, why Andrew compares the combination of AI and PLM to a messy bowl of spaghetti, what previous podcast guest Dr. Martin Eigner had to say about why companies are struggling to get PLM systems in place, and more.
Be sure to check out Stay Sharp Episode 137: CIMdata PLM/PDT Road Map Recap and join us each week for a new podcast.
Follow the Razorleaf Podcast, Stay Sharp in Digital Engineering on:
