Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How one factory cut lead times by 70 percent with Cellular Manufacturing
Transcript
- Lucas: Picture a factory floor where machines are grouped by function — all the milling machines in one corner, all the drill presses in another, and parts travel in big batches from one department to the next. That layout is called 'batch and queue,' and it's been the default for a century. Luna: And it's probably what most of us picture when we think 'factory.' But you're about to tell me there's a better way. Lucas: There is, and it's called cellular manufacturing. Instead of grouping by machine type, you group the different machines needed to make a specific family of parts into a small U-shaped 'cell.' One operator might run three different machines. Parts move one piece at a time. Lead times collapse. Luna: Before we go deep — Lucas, these deep dives into real operations fixes are exactly what make this show useful for people running actual businesses. If that resonates with you, listener, the reason we can keep this ad-free is listener support. You can buy us a coffee at buy me a coffee dot com slash fexingo. Every little bit helps us keep doing episodes like this. Lucas: Yeah, it really does. And today's episode is a perfect example — we're looking at a mid-sized aerospace components supplier I'll call AeroCell for confidentiality. They were drowning in lead times of 12 weeks, on-time delivery around 62 percent, and work-in-process inventory stacked to the ceiling. Luna: So they tried cellular manufacturing. What was the first step? Lucas: They picked one product family — a set of 40 different bracket and clamp parts that accounted for about 30 percent of revenue. They physically moved all the machines needed to make those parts — CNC mills, lathes, deburring stations — into a single U-shaped cell. That took a weekend and a lot of floor marking tape. Luna: I imagine the machine operators had some strong opinions about that. Lucas: Oh, absolutely. The traditional layout had specialists — one person only ran mills, another only ran lathes. In the cell, they had to learn multiple machines. That cross-training was the biggest people challenge. But the plant manager gave them eight weeks of training and said, 'We're not going back.' Luna: And the numbers? You said lead times dropped. Lucas: From 12 weeks down to 3.5 weeks — that's a 70 percent reduction. Work-in-process dropped 80 percent, because parts weren't sitting in queues between departments. Floor space freed up by 40 percent. And on-time delivery climbed from 62 percent to 94 percent within eight months. Luna: Okay, but I've heard that cellular manufacturing can cause an initial dip in productivity. Did they see that? Lucas: They did. The first month was brutal. Throughput actually dropped about 15 percent. Operators were still learning the new machine sequences, and balancing the cell — making sure each station had the right amount of work — took trial and error. They had to rebalance the cell three times in the first two weeks. Luna: So how do you rebalance a cell? Lucas: You time each operation — say, milling takes 40 seconds, drilling 20 seconds, deburring 30 seconds. Then you distribute the work so that no station is the bottleneck. In their case, they ended up having one operator run the mill and the drill, while a second operator handled deburring and inspection. That evened out the cycle time to about 45 seconds per part. Luna: So it's not just rearranging machines — it's rethinking the entire workflow and who does what. Lucas: Exactly. And the beauty is that once the cell is balanced, you get continuous flow. One piece moves from operation to operation with no waiting. Defects get caught immediately because the next operator sees the previous step's work right away. In the old layout, a bad batch might not be discovered until days later. Luna: That sounds like a version of the Toyota Production System's 'andon cord' philosophy — stop the line when there's a problem. Lucas: It's directly descended from that. Cellular manufacturing is one of the pillars of lean. Toyota's been doing it since the 1960s. But what's interesting is how many mid-sized Western manufacturers still use batch and queue, even though the evidence for cells is overwhelming. Luna: Why do you think that is? Lucas: Two reasons. First, the upfront disruption. Moving machines, retraining people, and the initial productivity dip — that scares plant managers who are evaluated on quarterly output. Second, the accounting systems don't capture the benefits. Traditional cost accounting allocates overhead based on machine hours, so a cell that reduces inventory and lead time actually looks worse on paper because inventory carrying costs aren't visible. Luna: So the finance department might actually resist the change because their reports show higher unit costs. Lucas: Exactly. But if you measure what matters — cash to cash cycle time, customer satisfaction, defect rates — the cell wins every time. AeroCell's CFO eventually saw that the inventory reduction freed up $1.2 million in working capital. That got everyone's attention. Luna: Were there any product families that didn't fit the cell model? Lucas: Yes. They had a handful of high-volume, low-variety parts that were better served by a dedicated transfer line — basically an automated assembly line. Cells work best for medium to high variety, low to medium volume. If you're making the exact same part a million times a year, a dedicated line is more efficient. Luna: So the key is to classify your product portfolio and match the process to the product. Lucas: That's the essence of operations strategy. And it's not just for aerospace. I've seen cellular manufacturing work in electronics assembly, medical devices, furniture making — even in a bakery that organized its equipment into a pastry cell and a bread cell. Luna: I want to go back to the cross-training piece for a second. How did the operators adapt? I imagine some loved learning new skills, others hated it. Lucas: You're right. In AeroCell's case, about 20 percent of the operators struggled and eventually self-selected into different roles. But most embraced it. The cell operators ended up with more varied work, more autonomy, and higher pay because they became multi-skilled. Turnover in the cell dropped to almost zero. Luna: There's a human dimension to operations that doesn't show up on the process map. Lucas: The best operations managers understand that. You can have the perfect layout, but if people aren't engaged, it won't sustain. AeroCell's plant manager spent half his time on the floor the first month, coaching, answering questions, celebrating small wins. Luna: What about maintenance? In a cell, if one machine goes down, the whole line stops. Lucas: That's a real risk. But because the cell is simpler and the operators run multiple machines, they become more attuned to early signs of trouble — unusual vibrations, temperature changes. They implemented total productive maintenance, where operators do basic cleaning and inspection daily. Downtime actually decreased compared to the old layout. Luna: So the cell model forces a culture shift from 'I just run my machine' to 'I own this product.' Lucas: That's exactly it. And that ownership mentality is what drives continuous improvement. After six months, the AeroCell team had implemented 40 kaizen improvements — things like moving a tool rack closer to reduce walking, adding a visual signal when bins are full. Those small changes added up to another 10 percent throughput gain. Luna: Are there any common mistakes companies make when implementing cells? Lucas: The biggest mistake is trying to go too fast. One company I know tried to convert the entire factory to cells in one month — chaos. AeroCell started with one cell for one product family, proved the concept, then expanded. Second mistake is not involving the operators in the design. The people who do the work know the workflow better than any engineer. Luna: So start small, involve the team, and expect a rough first month. Lucas: Exactly. And measure what matters. Don't just look at output per hour. Track lead time, inventory turns, and first-pass yield. Those are the metrics that tell you whether the cell is really working. Luna: I think the takeaway for our listeners is that cellular manufacturing isn't just a layout change — it's a system change that touches people, processes, and even accounting. Lucas: Right. And the results, when done right, are transformative. AeroCell went from a struggling supplier to one of the most reliable in their aerospace customer's supply chain. They're now planning a second cell for a different product family. Luna: Thanks for walking us through that case. I think a lot of operations folks will be looking at their factory layout a little differently tomorrow. Lucas: That's the hope. And if you try it, let us know how it goes.