How Sumitomo Drive Technologies Is Fitting 30,000 Square Feet of Inventory Into 7,500

Open on YouTube ↗
Overview

Sumitomo Drive Technologies is rebuilding its distribution center around automation. The goal is to fit a growing business into the building it already has. On an episode of Manufacturing Happy Hour, host Chris Luecke spoke with two leaders of the project, Tony and Shawn. The central question was how a manufacturer that still runs its warehouse largely on paper can move to a high-density, robot-served system. Tony and Shawn share one position: automation lets the company grow without adding floor space or many more people. They also say that the workforce, the data infrastructure, and the long-term benefits still have to be managed and learned along the way. The project was not finished at the time of the conversation.

16 min read

The Walmart Analogy

Tony described the project through a shopping trip. Normally you get a list, drive to Walmart, and walk the store to find each item: toothpaste, shampoo, something from electronics, something from sporting goods. The automation project reverses that. You stand at the checkout counter, say "toothpaste," and it appears. The same happens for items from sporting goods or the garden department. Everything arrives organized. You can then take it home, or send parts of it to other people, such as your wife, sister, aunt, or uncle, without going anywhere yourself.

In the warehouse, the system brings goods to the material handlers. Automated guided vehicles (AGVs) then carry the goods to their destination. That might be a manufacturing location, an assembly cell, a parts order, or a shipping truck.

What the System Actually Does

Shawn gave the more literal version, starting with why the project exists: "We're a growing company, and we're running out of space." The system they chose lets them take almost 30,000 square feet of material and put it into a 7,500-square-foot footprint. (He misspoke several times as "75,000" before landing on the correct figure, to some laughter.) That compression frees up inventory space across the building. It lets the company grow without physically expanding.

He walked through the flow. Inventory comes in the back door and is scanned. The system says exactly where to put it, so material is not handled several times. It goes directly from receiving into either the AutoStore system or the high racks. For orders, small parts come out of the AutoStore. An operator scans an order that might include four or five SKUs, and the system brings all of those parts to the operator. The operator places them in a bucket, which moves down the line to consolidators. Heavier material still comes from the high racks. It is combined with the small-parts order, loaded onto an AGV, and sent to assembly or manufacturing areas to complete the order.

Shawn also described the current setup. Today, all those small parts sit in racks about 30 feet tall. Operators ride order pickers 30 feet into the air to pull parts off shelves, bring them down, and then consolidate. So besides efficiency and space, the new system adds a safety benefit.

Origins: COVID Demand and a Warehouse Running Out of Room

Tony traced the project back to 2021. Sumitomo Drive Technologies is heavily involved in the material handling industry, especially parcel handling. During COVID, people ordered everything online and shipped through UPS, FedEx, and DHL, and the company's business grew significantly. Because it makes parts and units, that growth pushed inventory to levels it had not seen before, even though the company tries to keep inventory low.

The company had already expanded its headquarters by 100,000 square feet in 2018. By 2021 it was again facing expansion just to hold inventory. Its warehouse is 100,000 square feet of high-bay racking reaching 35 feet. Tony said it did not make much sense to keep expanding the plant for inventory, and the team figured "there had to be a better way."

The idea came from a customer who showed them an AutoStore installation. It was a similar application, with industrial and heavy products, and the team asked, "Why not us?" They partnered with Swisslog as the integrator, which Tony called a major asset. The plan takes 30,000 square feet of high-bay racking holding what Tony described as thousands, even hundreds of thousands, of boxes and moves it into a 7,500-square-foot cube. Fourteen robots run on top of the cube and deliver product to material handlers. From there, AGVs carry it to the assembly line.

The decision was made in 2021, but actual execution began around 2025. Tony explained the gap as deliberate. The team needed time to get comfortable with the project's economics and to plan for the process change. They also needed to communicate with employees so that people would not see the project as "you're taking my job from me." He said the team was diligent and measured, which may explain the five-year gap. It was important to have a solid business case and to bring employees along.

Measuring Success: Automation and Density

Luecke summarized Tony's framing of the project as having two parts: automation and density. Turning 30,000 square feet into 7,500 means the storage uses 25% of its former space. He asked how each part is measured.

On automation, Tony pointed to labor. He said the shortage of skilled labor is real, but the shortage of entry-level material handlers is "very real" for the company. Applicants tell them they don't want to go 30 feet in the air to pick a 75-pound box. Under the new system, a new employee instead stands at a terminal. Tony called it "almost video game-like." The employee clicks a button, performs quality checks, and places items in a bin that goes onto an AGV. The company expects picking accuracy and output to rise. It also expects to need fewer people in material handling, with those who are no longer needed there redeployed elsewhere in the business.

On density, Tony said the benefit "just can't be overlooked." About 25% of the warehouse space becomes available for growth. That could mean more inventory to support a larger business, or more value-added work such as assembly lines or manufacturing.

The Workforce Question

Shawn added that the company does not plan, and never has planned, to eliminate its workforce. Once the system is running at full speed, he estimated it will need about 65 to 75% of the current warehouse workforce. He said the facility has enough openings to more than employ the rest in other roles. Reducing headcount for profit is "not the goal at all," in his words. The bigger benefit, he said, is that the company can scale without adding many people. It will still hire as it grows, just not as many as a fully manual process would need.

Tony tied this to a concrete target. The company wants to grow 50% over the next five years. Under its plan, the same number of people it has today would support that larger business through 2029. Tony called this the "light bulb moment." Historically, the company added people every time it grew. Efficiency improvements helped only incrementally, and this was a chance to do something different.

Luecke noted that Tony had previously named attrition as one of the business's top challenges. He added that the company has not been able to hire all the people it would like.

From Pen and Paper to a Single Interface

Asked what it's like to introduce technology into an environment with little of it, Shawn described the present as almost entirely manual. Orders come out of the ERP system, but very few warehouse transactions happen electronically. Everything is printed. Orders are matched on paper, labels are printed to match the paperwork, and the paperwork is handed from operator to operator through the whole process. The ERP exists, but it is not easy for floor operators to use at a terminal, so they print everything.

In the new setup, Swisslog's software system connects to the ERP and becomes a single point of access for all warehouse services. Workers will have handheld computers, wrist-mounted tablets, and forklift-mounted tablets, so everything is electronic. An operator logs in and the screen shows exactly what to do. When the task is finished, it assigns the next task by priority. Shawn described it as one interface from the time material comes in the door to the time it goes out.

Recruiting and Onboarding a New Generation

Luecke asked how the company plans to attract people who would rather work in a "high-tech logistics environment" than ride a lift 30 feet up. Shawn spoke bluntly about younger workers. In his view, they are not the workforce of 20 years ago, which was willing to "sweat it out" to provide for their families. Younger workers, he said, want to do things as easily as possible with the least effort and still get the same or more pay. Since the company still needs these positions filled, he said, it has to make the work as easy and interesting as possible for new employees.

Tony framed it around turnover. The company would like entry-level hires to stay for life, and he believes its environment and value proposition allow that. But "we're not naive enough to know that everybody that we bring in is going to be an employee for life." The system shortens onboarding and makes training easier and more visual. "It's not like playing a video game, but it's like playing a video game," he said. He compared it to the difference between digging a hole and working at a computer. He stressed that operators don't need to be computer experts. He called the system intuitive and visual, and said most entry-level hires are younger and comfortable with technology.

In the paper-based process, a trainer has to explain which part is which, and a worker who doesn't know a part number can pick the wrong item. Tony said those problems largely go away when the process is automated and screen-driven. Training becomes more on-the-job. When someone leaves, a replacement can be brought up to speed, by his estimate, within the first or second shift.

Luecke added his own view. He sees manufacturing and logistics finally catching up to how people interact with technology in daily life. Generations raised on smartphones expect easy digital interfaces. However long an employee stays, the business benefits from getting them effective quickly.

The Longer Arc: From Distribution to Manufacturing to Automation

Asked for a story showing how more technical work changes a workplace, Tony gave some company history. Around 2013, Sumitomo Drive Technologies was mainly an assembly and distribution company in North America. It decided that model was not sustainable long term and began bringing manufacturing back to the US. That meant moving some operations to satellite locations to make room for manufacturing. It also meant retraining many material handlers and assemblers for more skilled work, including CNC manufacturing. Tony said this got employees excited about a company that was committing to manufacturing for the long term.

When the first round of tariffs hit in 2019, the company accelerated. It built a factory of almost 200,000 square feet in Mexico to support nearshoring and more manufacturing. It is also continuing to invest in large-scale manufacturing in Chesapeake, Virginia, for the larger industrial gear drive units it sells into mining, steelmaking, power generation, and other industries. Tony said the shift created both technological opportunity and general excitement about a growing company. He described it as a company recognizing that a purely global supply chain "maybe was a thing of the past." He sees the AutoStore project as the next step: from manufacturing to a more automated warehouse and a more automated company.

Where AI Fits and Where It Doesn't Yet

Since the project started in 2021 and ran through what Luecke called the first AI boom, he asked whether AI had changed it. Shawn said it had not changed the scope much. The underlying technology has existed for a while. Apart from helping the team get answers to questions faster, the newest AI developments have not changed the physical scope. He added that he "almost guarantee[s]" AI will affect the project in the very near future.

Luecke suggested the project is building the infrastructure for AI later, and Shawn agreed. "AI doesn't do anything for you when you're dealing with pen and paper." With the operation moved onto software systems, decisions that are now manual or fixed by programming could later become dynamic decisions supported by some kind of AI.

Tony agreed. He said the system will produce more data, which should help the company manage the business better, and AI will build on that. But he stressed how much is unknown. If they spoke again 12 or 18 months later, he expected the team would find benefits nobody can foresee today, because "we just aren't that smart" yet about what will happen. He described a flood of new AI tools, mentioning Claude and Gamma as examples he sees on LinkedIn. He called the AutoStore effort "very traditional automation." As for how AI plugs into it, he said, "I just am not sure that we know quite yet." Luecke suggested a second conversation in two or three years to cover lessons from a running facility.

Shawn's Lesson: Do the Homework Early and in Detail

For Shawn, the biggest lesson is education. The team started in 2021 and spent more than a year deciding which technology to use before choosing AutoStore. It then had to select a vendor and begin the actual process. Despite that preparation, he said he and his team are still learning every day. There were things they did not know and assumptions they made.

His advice is to start studying early and plan beyond installing the automation itself. Teams should work out how the system connects to the existing ERP, how processes communicate with each other and with the AGVs, and how the system will be used in the future. A clearly structured plan at the start leads to a better result. He credited Swisslog with steering the team and said the project would have been impossible with internal resources alone. He also urged leaders to get their teams "into the weeds." In his view, the decisions that matter are made at that level, and a small detail there can affect much larger systems.

Tony's Advice: Problem First, Communication, Scalability, and Action

Tony framed his answer as advice for anyone considering automation, in several points.

First, start with the problem. In his view the technology is easy and the solution is generally simple, so the real question is what you're trying to solve. For Sumitomo Drive Technologies, the problems were scaling without adding a lot of fixed cost and dealing with labor shortages that the company faces now and expects to continue.

Second, communicate early and often. Automation is still a frightening word for people whose jobs it affects. Employees need to understand that the aim is not to displace them but to make their jobs easier and more efficient. If someone needs to move to another role or function, the company has the ability to do that.

Third, plan for scale. Tony described AutoStore as modular, "kind of like a Lego." Capacity can be added on top or to the side. If the business grows faster than expected, or in five to eight years, the company could extend the system rather than start over.

Finally, "just do it." Teams can spend a long time on spreadsheets, meetings, economic cases, and vendor pitches, but at some point they have to commit. In his view, companies not pursuing automation are "missing the boat," because it is the wave of the future. He ended on the labor situation that drove the project. He does not expect labor shortages and the difficulty of bringing people into the workforce to go away, and said the problem is "probably only going to get more difficult."