Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How the US Army Transformed Global Supply Chains
Transcript
- Lucas: When you think of cutting-edge supply chain operations, the US military probably isn't the first thing that comes to mind. But in the last decade, the Army has quietly pulled off one of the most impressive logistics transformations in recent history. Luna: I've heard about this — the Global Combat Support System. It sounded like a massive ERP rollout, but with tanks. Lucas: Exactly. It's essentially a military-grade enterprise resource planning system. But the mindset shift is what's really interesting. For decades, the Army operated on a 'push' model — they'd stockpile inventory at forward bases, predicting what units might need, and just ship it forward. Luna: That sounds incredibly wasteful. And it must have been hard to adapt when actual needs differed from predictions. Lucas: That's exactly the problem. By the early 2000s, the Army was sitting on over $70 billion in inventory, much of it obsolete. The system was designed for a Cold War scenario with fixed supply depots and predictable front lines. But in asymmetric warfare in Iraq and Afghanistan, units needed parts for MRAPs, up-armored Humvees, and drones — not M1 tank treads. Luna: So what did they do? Just implement SAP and hope for the best? Lucas: More or less, but with a critical operational twist. In 2015, the Army launched gcss army — Global Combat Support System — which is built on SAP's platform. But the real change was shifting from push to pull. Units now submit requests through the system, and the supply chain responds based on actual demand data, not forecasting. Luna: That's the lean manufacturing principle — like Toyota's kanban system, but for the battlefield. Lucas: Precisely. In fact, the Army studied Toyota's production system extensively. They even sent logistics officers to visit Toyota plants in Kentucky. The result is a system where every spare part has an RFID tag, and the supply chain is visible in near real-time. At Fort Hood, a pilot program in 2024 reduced the average wait time for a critical repair part from 14 days to under 6 days. Luna: Six days is still a long time if you're waiting for a tank transmission. But compared to two weeks, that's huge. Lucas: And they're still improving. The goal is to get to 24-hour turnaround for most parts. But the bigger picture is cost: since implementing gcss army, the Army has reduced its total inventory by about 30 percent, freeing up billions of dollars for other priorities. Luna: I imagine the cultural resistance was immense. The Army is not exactly known for embracing change quickly. Lucas: You're right. The biggest challenge wasn't the technology — it was changing the mindset of supply sergeants who had been taught for decades to 'hoard' parts. In the old system, if you had a spare engine, you didn't report it, because you might need it later. The new system requires transparency and trust. Luna: How did they overcome that? Lucas: Through a combination of top-down mandates and bottom-up incentives. The Army made unit readiness scores partially dependent on how well units managed their supply data. If a unit had excess inventory, it hurt their rating. They also created 'logistics command centers' — essentially war rooms where real-time data is displayed on giant screens. Luna: So it's like a commercial operations center, but with camouflage. Lucas: Exactly. And the results are measurable. In 2025, the Army reported that its aviation fleet readiness — the percentage of helicopters available for mission — had increased by 12 percentage points since the full rollout of gcss army. That's a massive operational improvement. Luna: I read that Walmart implemented something similar with its Retail Link system — sharing point of sale data with suppliers to drive replenishment. The military seems to be doing the same thing with its defense contractors. Lucas: That's a perfect comparison. In fact, the Army now requires its major suppliers — like Lockheed Martin and BAE Systems — to integrate their inventory systems with gcss army. So when a part ships from the factory, it's already visible in the Army's supply chain system. Luna: That level of integration is hard even in the private sector. I wonder how they handle cybersecurity. Lucas: With good reason. The system is built on a classified network, and every transaction is encrypted. But the biggest risk isn't hacking — it's data quality. If someone enters the wrong part number, the entire supply chain can get misdirected. So the Army invested heavily in training and in barcode scanning at every step. Luna: Sounds like the same challenges any company faces with an ERP implementation. But the stakes are higher — lives depend on it. Lucas: Absolutely. In a combat situation, a missing part can mean a vehicle is down for days, putting soldiers at risk. That urgency is what drove the Army to push this through. And the lessons are directly applicable to commercial operations. Luna: What's the biggest takeaway for a business leader listening? Lucas: I'd say two things. First, the importance of real-time data over forecasting. The Army learned that predicting demand is inherently inaccurate — it's far better to create a system that responds quickly to actual demand. Second, the cultural change is harder than the technology. You can buy SAP, but you can't buy the trust needed to make a pull system work. Luna: That's a good point. And I think a lot of companies still rely on spreadsheets and gut feelings for inventory management. Lucas: Right. The military is often seen as bureaucratic and slow, but in this case, they're ahead of many corporations. And if you're interested in how these ideas apply to your own operations, it's worth diving into the gcss army case study. Luna: On that note, if these conversations have moved your work forward in any way, we'd love your support. Listener contributions are what keep this podcast ad-free and focused on real operational stories. Lucas: Yeah, we don't run ads — it's just us digging into topics like this. If you find value in that, you can support the show at buy me a coffee dot com slash fexingo. Luna: And no pressure — truly. But every contribution helps us keep doing deep dives like today's. Lucas: So speaking of deep dives — I want to look at one more aspect: how the Army is now applying machine learning to its supply chain data to predict failures before they happen. Luna: Predictive maintenance — that's the holy grail. How far along are they? Lucas: They've started piloting algorithms that analyze engine vibration data and oil samples to forecast when a part will fail. At Fort Bliss, they reduced unplanned maintenance on M2 Bradley fighting vehicles by 25 percent in the first year. Luna: That's impressive. And it shifts the supply chain from reactive to proactive — ordering parts before the breakdown happens. Lucas: Exactly. It's the next evolution of the pull system. Instead of waiting for a unit to request a part, the system anticipates the need. That's where commercial supply chains are heading too — companies like GE and Rolls-Royce already do this with jet engines. Luna: So the military is essentially catching up to, and in some cases leapfrogging, private industry. Lucas: In logistics, I'd argue they're now ahead of most sectors. The combination of RF ID, real-time data, and predictive analytics is creating a supply chain that's more resilient and efficient than anything most companies have. Luna: It makes you wonder — if the Army can do this, why can't every company? Lucas: Good question. The answer is probably culture and investment. The Army spent over a decade and billions of dollars on this transformation. Most companies don't have that patience or budget. But the principles scale down — even a small business can start with basic demand-driven replenishment. Luna: That's reassuring. So the lesson is: start with the data, change the culture, and the technology follows. Lucas: Exactly. And that's what we'll be exploring more in future episodes — how operational transformations happen in different contexts. Thanks for joining us today.