Latest / The Operations Podcast with Fexingo: Process, People, and Profit in Modern Business / How UPS Built a Package Operations Empire
Transcript
- Lucas: Twenty-five million packages a day. That is the number UPS moves — every single day, Monday through Saturday, Christmas Eve included. And the remarkable thing is not the volume itself, it is the fact that they route each one through a hub and spoke network designed in the 1950s and still running on industrial engineering principles from that era. Luna: You say 'still running' like it is a bug, but I think it is a feature. UPS has been perfecting the same core operating model for seventy years. Lucas: Exactly. And the model is brutally simple. All packages go from origin to a regional hub, then to a central super-hub, then back down. It adds miles — a package from Brooklyn to Manhattan might fly through Louisville, Kentucky — but it creates density. And density is what makes the economics work. Luna: I want to come back to the Louisville hub, Worldport. But first — how do they even sort twenty-five million packages? That is a staggering number. Lucas: They sort them with a combination of conveyor belts, scanners, and human hands. About 60% of packages are sorted automatically now, but the rest still require a person to read a label and toss it onto the right belt. The key metric is 'sort rate' — packages per hour per employee. UPS tracks it obsessively. Luna: And that is where the industrial engineering culture comes in. UPS famously time-studies everything. Drivers have a specific method for how to hold their keys, how to step out of the truck, how many steps to take to the door. Lucas: Right. The '340 Methods' — a training manual that dictates everything. But it is not micromanagement for its own sake. It is a systematic attempt to remove wasted motion. Every second saved across 500,000 employees compounds into millions of dollars. And that culture is why UPS has historically had the lowest cost per package in the industry. Luna: But that cost advantage has gotten squeezed. Labor costs are up — the 2023 Teamster contract raised wages significantly. And Amazon is building its own delivery network, which pulls volume out of UPS's system. Lucas: Both are real pressures. Amazon was UPS's largest customer for years, and losing that volume forces UPS to run its network at lower density. That directly hurts margins. But here is where the operations playbook kicks in. UPS has been investing in automation to offset labor cost increases. They have spent over $1 billion on new sortation systems at hubs like the one in Atlanta and in Parsippany, New Jersey. Luna: And the automation is not just about sorting. It is about rethinking the layout of the hub itself. At Worldport, they installed a new system called 'Smart Sort' that uses machine vision to route packages more efficiently within the hub. Lucas: That is a great example. Smart Sort is not a flashy AI — it is a practical application of computer vision to read addresses and barcodes faster than a human can, and then dynamically assign the package to a specific chute. It shaves seconds per package, but on seventy million packages a week, that is massive. Luna: And then there is ORION. I want to talk about ORION because it is one of the most famous operations algorithms in logistics. UPS spent a decade and hundreds of millions building it. Lucas: ORION stands for On-Road Integrated Optimization and Navigation. It is a route optimization algorithm that tells drivers the most efficient sequence of stops. Before ORION, drivers used their own knowledge of their routes, which was often very good but not mathematically optimal. ORION saved UPS an estimated 100 million miles of driving per year — that is about 10 million gallons of fuel. Luna: But the rollout was famously rocky. Drivers hated it initially because it would tell them to make a left turn that felt wrong, or to skip a stop and come back to it later. It took years of tweaking and trust-building. Lucas: That is a crucial point. Operations improvements are not just about the algorithm. They are about change management. UPS had to retrain 55,000 drivers, and they had to prove that ORION's suggestions actually worked. They ran pilot programs where they compared driver-selected routes against ORION routes and measured the time and fuel difference. Eventually the data won people over. Luna: And now they are layering on even more data. UPS trucks have telematics that track speed, braking, idling, even the direction the driver looks. It feeds back into the system to refine the model. Lucas: That is the next frontier. They call it the 'Network of the Future' — and it is not just about routing. It is about dynamically adjusting the hub and spoke structure itself. Traditionally, UPS has fixed hubs that sort packages the same way day after day. But with real-time data, they can reroute packages around a congestion point or a weather delay before it causes a cascade. Luna: That sounds like a fundamentally different operating model. More like a real-time adaptive network than a static hub and spoke. Lucas: Exactly. And they are testing it at a new facility in Texas called the 'Smart Hub'. It uses autonomous guided vehicles to move packages between sorting areas, instead of the traditional conveyor belts. If a particular destination has a surge of volume, the system can allocate more belt capacity to that destination on the fly. Luna: Let me push back a little. All this automation and data — doesn't it conflict with the union culture? The Teamsters have historically resisted technology that reduces headcount. Lucas: It does create tension. The 2023 contract had provisions that limit the use of autonomous vehicles and require that new technology be bargained. But UPS has been careful to frame automation as augmenting workers, not replacing them. At the Parsippany hub, for example, automation reduced the need for temporary holiday workers, but the union full-time employees kept their jobs and got higher wages. Luna: And there is a strong argument that the union actually forces UPS to be more operationally disciplined. When you cannot squeeze labor arbitrarily, you have to invest in better processes. That is a healthy constraint. Lucas: I think that is right. UPS's operating margin before the 2023 contract was around 13%, which is very high for logistics. The contract added about $500 million in annual labor costs, but UPS expects to offset that through automation and efficiency gains. They have a clear plan. Luna: Now, I want to talk about the air network. UPS has one of the largest cargo airlines in the world — over 290 aircraft. And Worldport in Louisville is the heart of it. That hub processes about 2 million packages per day, mostly at night. Lucas: Worldport is staggering. It is a 5.2 million square foot facility — that is about 90 football fields under one roof. Packages arrive by air, get sorted, and then depart on outbound flights within about four hours. The entire operation is timed to a nightly cycle that starts around 11 PM and ends by 4 AM. Luna: And the sorting technology there is mind-blowing. They have these 'crossbelt sorters' that can route 60,000 packages per hour. Bags tumble down chutes, packages spin around on a loop, and each one gets diverted to the correct belt for its destination. Lucas: The crossbelt sorter is a classic piece of industrial engineering. It is a series of carriages on a loop, each with a moving belt. A package lands on a carriage, the carriage moves along, and when it reaches the correct destination chute, the belt spins and pushes the package off. It is mechanical, precise, and incredibly reliable. Luna: And the reliability matters because UPS has commitments. If you pay for Next Day Air, the package is guaranteed by 10:30 AM. That promise depends on Worldport delivering every single night. Lucas: That is the operational challenge in a nutshell. The entire system is designed for one outcome: packages arrive on time, every time. And they measure it down to the minute. UPS tracks 'on-time performance' at every step — from pickup to sort to delivery. If a flight is delayed by 15 minutes, it ripples through the entire network. Luna: And they have backup plans for backup plans. If a sorter goes down, they have spare parts on site. If a flight is delayed, they have standby aircraft. The culture is built on redundancy and contingency. Lucas: Speaking of contingency — let's talk about the 2020 holiday season. During the pandemic, volume surged 30% year-over-year. UPS was handling 30 million packages a day at peak. And they did it with the same number of drivers and sorters, just running more hours and using temporary workers. Luna: But that also created quality problems. On-time performance dipped. Customers complained. That experience is part of why they are accelerating automation now — to handle surges without degrading service. Lucas: Exactly. And that points to the next big frontier: last-mile delivery. The hub and spoke system works brilliantly for long-distance, but the last mile to the customer's door is the most expensive part. UPS has been testing drones for rural delivery and electric delivery vans for urban areas. Luna: The drone program is interesting because it is not about speed — it is about cost. In rural areas, a driver might spend 20 minutes driving between stops. A drone can fly directly and drop a package. UPS has a subsidiary called UPS Flight Forward that has a Part 135 air carrier certificate from the FAA, allowing it to operate drones commercially. Lucas: And they are starting with healthcare deliveries — transporting blood samples and medications between hospitals. That is a controlled use case where the economics work. It is not about replacing all drivers; it is about targeting the highest-cost, lowest-density routes. Luna: I think that is the hallmark of great operations thinking. You do not try to change everything at once. You find the specific point where a new technology or process delivers the most leverage, and you start there. Lucas: That is the UPS playbook. Incremental, data-driven, relentlessly focused on cost per package. They are not trying to be the flashiest logistics company. They are trying to be the most efficient. Luna: And that efficiency has created a remarkable moat. FedEx competes on speed and reliability, Amazon competes on integration, but UPS competes on having the lowest cost structure in ground delivery. That is hard to replicate. Lucas: It is very hard to replicate, because it is based on decades of accumulated operational data and a culture that cannot be bought. You can buy the same sorters, the same trucks, the same software. But you cannot buy the institutional knowledge of how to run a network at that scale. Luna: You know, it strikes me that this is really an underappreciated form of competitive advantage. When people think of UPS, they think of brown trucks and uniforms. But the real asset is the operations system underneath. Lucas: Absolutely. And that system is being challenged right now by changes in the market — e-commerce growth, labor costs, Amazon's encroachment. But if any company can adapt through operational excellence, it is UPS. Luna: I think we have covered a lot of ground. Let me ask you one final question: what is the one operational metric you will watch over the next year to see if UPS's strategy is working? Lucas: I would watch the cost per package. UPS reports it in their quarterly filings. If automation and network redesign are working, that number should start to decline, even as wages rise. If it stays flat or rises, the strategy is not offsetting labor costs. Luna: That is a good answer. And it is remarkable that a company that moves 25 million packages a day can still find ways to improve. That is the beauty of operations — there is always another 1% to squeeze out. Lucas: And we have to give credit where credit is due. This episode was made possible by listeners who support us directly. We do not run ads, and a small group of listeners chips in monthly through buy me a coffee dot com slash fexingo. That is literally what funds making this many episodes. So if these conversations have moved your work forward in some small way, that is the place. Luna: Yeah, it makes a real difference. Keeps the show independent and allows us to dive deep into topics like this. Lucas: So back to that cost per package — we will be watching it. And we have links in the show notes to UPS's investor presentations if you want to dig into the numbers yourself. Thanks for listening.