What’s In It For Me When a Foundry Goes ‘Smart’
Beginning in 2021, AFS Corporate Member Grede, headquartered in Southfield, Michigan, drafted the blueprints for an all-encompassing Industry 4.0 infrastructure for two years. They began laying software and hardware foundations throughout 2023, and also went live with IoT software in sand operations for half of their plants. Full implementation will be completed in 2026.
This year, the metalcasting giant will cross important milestones in its five-year journey to become AI driven and monitored throughout the sand, melt, and finishing departments at all nine of its foundry locations. Its substantial investment in the Norican Monitizer product is the foundation of the entire smart manufacturing upgrade that will ultimately solve labor knowledge-gaps and effect gains for customers that include cost reduction, improved casting quality, as well as product traceability and transparent product history through barcoding.
Grede is the first foundry in North America to implement Norican’s Monitizer Industry 4.0 solution across all the primary processes of metalcasting.
Casting Source interviewed Grede Chief Technology Officer Susan Bear to get a firsthand look at what goes into a technology shift of this proportion and discover the real “why” behind the massive undertaking. Following are excerpts from that conversation. (Editor’s note: Bear will deliver a presentation about the company’s Industry 4.0 journey and achievements during the AFS Metal Casting Congress held in Milwaukee this April.)
Casting Source: You’ve researched, you’ve planned, and you’ve begun to execute an elaborate strategy to become an Industry 4.0, smart foundry. But what’s in it for the casting customer?
Bear: We invested in Norican Group’s Monitizer as the foundation of this “home” we’re building. It is also the foundation of a new commercial advantage for customers. We are creating the ability to have traceability and barcoding on 100% of our products, meaning we will verify that every one of our established IoT process parameters have been measured along the way. We can send our customer a product that we know meets all internal quality and operations requirements––the history of that part is all in the barcode, and that will be there forever. Because we’ll have this traceability, which we feel is going to be key in the future, we’re providing them transparency into the product itself.
Because of this data on key casting characteristics, the customer will then be able to report to their customers. We’re not only helping sell our business to the casting customer, but we’re helping them to sell to their customer by giving them something to offer that they've never had before in the supply chain: a complete product knowledge of that exact part––so this becomes a multi-layer advantage.
CS: Is there a cost impact for casting customers?
Bear: At the end of the day, it comes down to the P&L. The way you affect your P&L is by pulling cost out of your processes. The way you do that is you reduce scrap and you improve efficiencies. And how do you do that? Well, you put technology in. To do that, you have to have the know-how on all your processes. Ultimately, this is going to improve your cost position, and what happens when you do that is, you become competitively more of an advantage to your customers. So, from a price perspective, the smart foundry has an advantage over its competition, because their price can be lower.
Customers are extremely demanding on quality. Technology enables a foundry to produce a highest quality product at a price that's now more competitive. Then, there's nothing left––that's the end game.
CS: How did Grede arrive at the decision to begin the journey; that the time was right?
Bear: In 2019, when we developed our 2026 strategic direction for the company, systems and innovation was one of the strategic imperatives that we outlined as a means to achieve our goals. The 2020 pandemic became an accelerator for attrition, which heightened our need for smart manufacturing. With the mass exodus of the industry, there was this knowledge loss happening. Everyone in the industry was hit hard, not only with attrition, but attrition of employees with 20 years of experience or more. When that happens, you lose more than a worker––you lose the know-how. So, we needed to find a way to quickly compensate in order to remain a top supplier to the industry and in order to continue to grow. By replacing know-how with data, we are enabling newer employees to be as effective as industry veterans.
CS: Obviously, Industry 4.0 technology isn’t something you just plug in one morning and say, voila, here we are. What does this process actually look like?
Bear: We always knew the technology was going to play a role in the future, but COVID really accelerated our planning and execution.
We recognized that software is an enabler of a process; technology is not the process itself. We couldn't just start in throwing Industry 4.0 technologies into our environment––we had to first identify and document the processes. So, we spent from 2020 to 2022 putting in dozens of processes into project management software, a sand standard, product engineering, technical excellence standards, quality system, metal standards, preventive maintenance processes, and more. Before you can move forward with technology, you have to have the process know-how to push into the technology. As attrition continues to happen in this industry, technology will compensate for that loss of knowledge.
CS: What’s the actual status as of Jan. 1?
Bear: You have to really break up your IoT implementation by process. We've taken it and broken it up into sand, melt, and then finishing. Process control at the source is key to superior quality, which is why we are focused on the sub-processes. The two heartbeats of a foundry are sand and melt, so that’s where we’re starting.
Today, we have five of out of nine foundries now running Monitizer for our sand systems in Phase 1.
All of our plants will be utilizing sand Phase 1 by May of 2024. This is what you call the discovery: real time monitoring and notifications when control limits or operating zones aren't met.
Then melt starts right around May of 2024 and goes till May 2025. It's about a year to get all your plants through one phase of one process. Now, that doesn't mean you aren't implementing AI in parallel. The way AI works is, it needs time to collect data––it needs to watch what you're doing. You can't just click on AI; it needs to learn. We'll start implementing AI in probably October of ’24 for our plants’ sand systems.
So, 2024 is a huge year for us. We will implement Monitizer for melting for most of our foundries–– and AI will start to come in for the sand as well.
CS: What does Phase 2 consist of?
Bear: Phase 2 is the AI portion. And that is where it takes out that loss of know-how and replaces it with technology that can make recommendations and provide guidance.
That is why we picked Norican––they’re the only one that has the AI piece. A lot of other companies don't feel that AI will actually ever have a role in the foundry as it does in other industries, but I beg to differ there. At Grede, we think it has a huge role in taking us from the art to the science, in terms of operational knowledge. This AI step is going to be critical in the future.
I spent about a year benchmarking different companies, and it takes an extremely long time to do the due diligence. You have to do it right from the start or it's going to cost you a lot of money and time.
Another advantage of Norican is (1) the tie-in with Simpson sand system––all of our plants use Simpson, and (2) all our plants use DISA and Wheelabrator. Norican is still compatible even if you’re not using one of their brands, but we felt it’s better if you choose a supplier that knows your equipment and business inside and out.
CS: When people hear Industry 4.0, they often think predictive maintenance. How does PM factor into Grede’s big ‘smart’ picture?
Bear: Maintenance is extremely important to us for two reasons: (1) Foundry equipment is really hard on itself so maintenance and uptime is critical. But (2), maintenance also happens to be where a lot of the human know-how is, which means there’s a potential to lose it.
So, the other piece of IoT that we've kicked off in Phase 1––starting in January and it's on a very accelerated curve––is a system called FIIX from Rockwell. This is a predictive maintenance IoT software that's giving you PMs that need to be completed. But where it gets kind of fun, from an Industry 4.0 standpoint, is that it links into Monitizer, the smart AI software that we're calling the foundation of our home. And it starts doing things like monitoring a temperature of a coil and a furnace. And if it notices a spike over a certain temperature, it will flag you: “Hey, you need to go do maintenance on that coil.” This is before you have downtime. It takes us from being reactive and puts us in a proactive position.
So, we're documenting everything––all the specs on all the equipment, all the parts parameters and limits for temperatures and cycles. You push all of that knowledge into your FIIX and your Monitizer, and it starts telling you before you hit a downtime event. This is an enabler to help run your foundry more efficiently.
CS: What does the employee adoption of the technology look like?
Bear: Culture is just as important as the technology. Culturally, we have been bringing the organization along since 2020 as we developed our strategy.
One thing I will say is, you have to prepared to get uncomfortable to get comfortable again. Our culture has been developed over the years to understand a certain expertise and things are done on the operations side. And now, you need a whole other skill set. Your team is either going to want to train to learn and use it, or it's going to be so foreign to them that they don't want to learn. So, you have to have a culture that wants to embrace change and embrace learning the innovative solutions that are going to put us at the top in terms of being the best casting supplier there can be.
You have to have a culture that wants to get uncomfortable to get comfortable. They have to be risk friendly. Obviously, you have to have executive buy-in, and our executive team is certainly driving that this is where we want to go as a company. But the culture is so extremely important. You can put technology into any plant, but if they don't want to use it to their benefit, then it's just going to get dusty on the proverbial shelf.
Our biggest challenge, though, from the plant floor perspective, is that we can’t get it to them fast enough. I can’t tell you how many times a plant manager will call me and ask when they’re going to get it. They all want to be great.
CS: For the plants where Monitizer has been installed, how is the technology impacting the everyday life of the frontline worker?
Bear: They’re still there doing their job––this software doesn't take their place at all. However, they now have the ability to watch what they're doing live on a screen in real time–– monitoring sand those control limits.
And if something's going astray, the notifications take some pressure off the team, especially those who don't have that 20-year know-how. The last thing you want to do is put any employee in a situation where they have to react but they don't know what to do. It's not fair to them. The software has automated notifications if something has stepped outside of a control limit, and it automatically goes to the quality manager or the supervisor on the line. Then they can quickly react instead of waiting to run an entire shift and now we have all this scrap. Now, the frontline worker doesn’t feel like they let somebody down. The automation takes away that pressure of needing to know everything, especially as a new employee.