If 2024 was the year when the hype cycle for industrial AI hit its peak, AI and associated manufacturing technologies like machine learning surely will resonate in 2025.
Thomas Wilk of Plant Services, a sister brand to Smart Industry, sat down with Aaron Merkin, who is CTO at Fluke Reliability, to gather his predictions on where manufacturing is headed technologically in the new year.
Merkin brings more than two decades of experience developing enterprise software across a variety of industries and markets, including roles at IBM, Dell, ABB, Aclara (now Hubbell), and Honeywell. His insights also will be featured in Smart Industry's upcoming Crystal Ball Report for 2025.
Below is an excerpt from this podcast:
Tom Wilk: We're heading toward the end of the year, and the start of a new year. My first question is going to focus on trends that you're seeing in industry, perhaps in this year and going into next year. You work with a wide variety of manufacturers in nearly every industry and vertical around the globe. What are some of the key trends that you're seeing right now in manufacturing?
Aaron Merkin: The first is still the continued focus on digital transformation. Customers are looking to more instrument their plans, collect more data and use that stream of data they’re getting from IoT solutions to inform the decision making.
See also: Six ways to incorporate AI into your manufacturing operations
It's been challenging times, and customers are trying to decide where to invest in terms of improving operations and improving reliability. That ability to have data allows them to improve their decision-making, and it’s also accelerating a shift towards predictive maintenance.
We see that customers are recognizing that by having instrumentation in their plant, they're able to better predict the health of assets and they're able to incorporate that into their short-cycle planning for when they're going execute work in the facility. It's really prompting them to be much more aggressive in adopting predictive maintenance.
The second thing we've seen is continued concerns about supply chain resilience. We've had disruptions over the last few years coming from COVID, and we’re still recovering after that. There's also a lot of uncertainty around reshoring and trade headwinds, and whether you’re going to be able to continue to rely on the suppliers that you've had overseas, so you see customers focusing on how to address that.
The mitigation that they're looking at taking is adaptability and transparency into their supply chains—looking for alternate suppliers outside of potentially problematic countries, and looking to dual-source their suppliers for critical equipment.
More important is a look at effective inventory management, and how do they make sure that they have the right parts on hand when they need them, and at the same time balancing and minimizing the amount of inventory that they're holding.
We see a lot of efforts looking at how they can incorporate AI into their supply chain planning, in their demand for parts, as well as for predictive maintenance—that ability to detect faults and plan for when you're going to have outages or predict when you may have an outage, which allows you to optimize the inventory you have on hand rather than being in reactive break-fix mode.
Crystal Ball preview: Top cybersecurity risks in 2025 and beyond
TW: It's been a weird year for manufacturing in the sense that I think a lot of manufacturers have been waiting (in the U.S. at least) for the Fed to cut interest rates, and we're starting to see that happening. As you were saying, supply chain is still weighing heavily in people's minds. We hear various stories where some supply chains are ironed out to where they were approximately pre-COVID, others are still sort of evening themselves out. What do you see specifically as the biggest challenges that are facing manufacturing coming into next year?
AM: I think one of the things that we've seen with manufacturers is the pace of technology change, the amount of information around generative AI, and the idea that we’ve hit an inflection point where if you're not already using Gen-AI over the last five years then you must clearly be behind.
See also: Partners in IoT that made perfect sense
We've seen a ton of pressure on manufacturers, particularly at the plant level, not just to analyze technology but to come up with a plan to adopt it whether it’s clear how they're going to use it or not. Irrespective of the problem that the end customer is trying to solve, we see RFPs and questionnaires asking for AI solutions, whether they're necessarily the appropriate solution or not.
Our customers are really feeling this pressure to not be left behind, and we encourage them to not fall victim to the siren song of being on the bleeding edge of technology, but instead really to focus on making sure that they understand the business outcome that they’re trying to achieve. Then as an organization you identify the business outcome, stick to working on your business priorities, and then work backwards to understand what the technology is that you need to address them.
Part of that is making sure you're not adopting technology for its own sake. Also because we're seeing such an incredibly fast pace of technology change, it's almost inevitable that you're going to define a problem, you're going to adopt technology, you'll run your pilot, and by time you complete that pilot something new or shinier has come along.
You get tempted to keep chasing the latest leading edge technology and never actually quite go into production, versus really understanding the business outcome you're trying to achieve and then, once you're confident that the technology you've chosen accomplishes it, go straight into production usage.
See also: DAPs help keep employees engaged in digital transformation projects
TW: We're also hearing that people feel pressured to adopt “AI something.” With AI as maybe part of your answer, what emerging technologies do you see that are on the horizon that actually will revolutionize or change manufacturing?
AM: Over the last decade or so, there's been an evolution away from enterprise asset management, or the augmentation of EAM with asset performance management solutions. Within those solutions there’s been a shift away from first principle modeling more towards the use of AI and machine learning for modeling.
We're seeing that going to hit the inflection point where customers are seeing real value in adopting these AI models versus first principle models for modeling the health of their assets. That includes not just modeling the health of individual assets, but the full adoption of digital twins and trying and using a digital twin philosophy or approach to model the entire flow of production through their plant floor.
I think those are the two biggest things: the evolution away from first principles specifically for asset health, but also the technology being at a point where you can comfortably use utilize it to model an entire production process.
PS: You know what's taking me by surprise, too, on a tool level, is the extremely rapid adoption of acoustic imaging. Fluke has helped lead the way with one of their devices. Have you seen that trend too for condition monitoring?
AM: We are seeing significant adoption of acoustic imaging, particularly for customers who are doing occasional spot checking, or if they’re still heavily into route-based maintenance and not quite ready to deploy a continuous monitoring technology like a wireless sensor.
See also: New report reveals ‘grind’ of digital transformation, slow road to success
It’s also used as a good red-yellow-green initial analysis for those customers who are bringing in a service company to do intermediate inspection, or who themselves are starting to work in some kind of condition monitoring program into their daily work.
TW: We've been talking about the skills shortage for a while in manufacturing and everyone I've talked to in 2024 has said the same thing: The shortage hasn't gone away, but we're trying to do our best to cover up for the lack of technicians in various ways. What are you hearing from your customers, related to how they're using technology to combat whatever skills gaps they might have on their teams?
AM: One of the areas we've seen is that adoption of cloud-based technologies allows a centralized workforce to more effectively monitor operations across multiple sites.
The second area is on a factory floor, within a plant, the deployment of wireless technologies allows broader coverage of assets, so it allows a smaller group of technicians and resources to monitor the balance of plant without having to conduct walk-around route based inspections. Also there's very much a challenge with skills shortages, and I think there was a lot of hope that, post-COVID, the skill shortage would go away. We're seeing that's not the case and that manufacturers are going to have to build a longer-term plan.
Webinar rewind: 'Next Year’s AI' and more 2025 industrial technology insights
From the technology front, we see adoption of VR and AR capabilities to perform training. I think there's still a question of whether AR is really safe to be used on the factory floor, but we see it as an opportunity to provide off-floor simulation and get people familiar with operations before they actually move on to the floor.
Beyond just technology in this area, and more related to business operation and practices, is that there’s much more proactive engagement. There's been investment continuous learning for the employees that companies have, and making sure the organization keeps upskilling them, both so teams can keep up with emerging technology trends, but also as a way of retention so that the teams feel like they're being invested in and valued. The second is through working with partnerships with trade schools and universities, so that you're kind of on the forefront of establishing that pipeline of skilled employees as they're coming out of training in these organizations.
Episode 4 of (R)Evolutionizing Manufacturing: The Art of the Possible and The Art of the Practical
For example, Fluke participated in the WorldSkills Program that took place in Lyon this year, and we've done that as a way of trying to be on the forefront of the education of the next generation of electricians, both from enabling them to be more productive in their daily work, but also obviously there's a little bit of getting them used to using our tools to perform their jobs, so when they go out into the workforce there’s a preference for having us as their partner.
TW: I'm feeling an enthusiasm among 45-and-unders this year especially. There's a lot of videos being uploaded to either Instagram or TikTok where under-30s are showcasing their technical skills, and they're excited to share too.
AM: It's been really interesting to see the amount of influencers in the trade industry space, whether it's plumbing or electricians, who are sort of making it cool again to be in the trades and bring in the generation behind them. I think there's been a shift away from the idea of that a university or college education is the only path to economic success in United States. The influencers on Instagram are able to communicate that the work is interesting, it's challenging, and it can be economically rewarding for an individual.
