Units Needed for Autonomous Mobile Robots at a Distribution Center

Should a distribution center start with autonomous mobile robots or forklifts?

A distribution center moving 2,000 totes a day across two shifts usually has two separate jobs: moving totes between zones, and picking them into orders. Autonomous mobile robots fit the first job when the moves repeat on mapped routes, such as receiving to reserve storage, reserve to pick faces, or pick areas to pack-out. They do not replace the forklifts that lift pallets into high racking or load trailers, so most sites run both and decide which moves to automate first.

Adoption data shows the category is in wide use, but it does not size your building. A January 2025 survey of 217 professionals by Peerless Research Group found that about 48% of companies reported using robots in warehouses and distribution centers. That figure covers robotics broadly, not autonomous mobile robots alone, so it shows the market is active without telling you the units needed for your floor.

A Peerless Research Group 2025 automation study of 139 respondents, reported in Modern Materials Handling, found that 41% were using shuttle systems and/or mobile robotic storage and retrieval systems, and 59% planned to upgrade or implement that equipment within two years. Those shares include storage systems as well as mobile robots, so they describe the market, not a fleet size for your site.

Think of the choice as a sequence rather than a contest. Begin with the moves that repeat most, prove the robots on those routes, and then decide whether the next phase is more transport, a picking change or more forklifts. A site that starts with pallets and trailers gets little tote relief, while a site that starts with repeat tote runs can measure the result within weeks.

How many units would 2,000 totes a day need?

Start from moves, not totes. A tote may be touched several times on its way through the building, including receiving, put-away, picking, packing and staging. Count every move the robots would take over, because a fleet works on moves, and 2,000 totes can become a much larger number of moves.

The sizing method is simple to state. Divide the daily moves by the productive hours each shift really delivers, then divide by the moves one unit completes per hour on your own routes. Add margin for charging, traffic, faults and maintenance. Every input is local, so a unit count from another building does not transfer to yours.

Build the margin on purpose. Faults, traffic at pick faces and staff standing in a route all reduce the effective rate of a unit. Ask for these losses to be shown in the sizing rather than assuming a clean run for every unit through every hour of the shift.

As a rough orientation only, 2,000 totes split evenly across two shifts is about 1,000 per shift. Real waves rarely split evenly, since inbound trailers arrive in bunches and order releases spike. A flat average hides the peak hour that causes backlogs, so size the fleet to the peak and then check how many hours the average day actually needs.

• Tote moves per hour by lane, for each shift
• Route length and number of stops on each lane
• Peak-hour volume, not the daily average
• Charging time and battery plan for each shift

Where do autonomous mobile robots beat forklifts, and where do they lose?

Autonomous mobile robots do best on level floors, consistent routes, predictable traffic and standard loads. They can run the same loop through a shift without breaks, which matters on the second shift, where backlogs often build. Forklifts keep the advantage on pallet work, high reach, trailer loading and irregular loads that need an operator's judgment.

Labor context is broad rather than site-specific. A news summary of Bureau of Labor Statistics data reports a 3.3% hires rate and a 3.3% total separations rate across all industries for 2025, and says transportation and warehousing employment fell by 105,000 jobs from its February 2025 peak. According to FRED, warehousing and storage employment stood at 1,836.2 thousand workers in December 2025, seasonally adjusted.

The Bureau of Labor Statistics Occupational Outlook Handbook projects 4% growth for hand laborers and material movers from 2024 to 2034, from about 7.0 million jobs in 2024. A wider labor pool does not solve a shift that cannot be staffed today, so the hiring picture is one input among several rather than a reason to automate on its own.

Safety belongs in the comparison too. A GAO figure, as reported by Public Now, estimates 3.8 cases per 100 workers in transportation and warehousing in 2022, the highest serious injury and illness rate of 19 sectors. That figure covers the whole sector, not robots, but it argues for a safety review of any route where robots and people share aisles.

Exceptions are where the two options differ most. A robot can stop and wait for a clear path, but it does not decide whether a damaged tote should be set aside or a mislabeled tote sent back to receiving. Plan a staffed exception point so those totes do not sit in the robot path.

What do throughput figures from vendors and case studies actually show?

Throughput numbers are the most quoted and the least portable. A third-party logistics provider reported more than 500 units per hour per associate on a goods-to-person system in a vendor-sponsored case study video, and the facility is designed for 1.6 million picking units per day. These are reported and design rates for one site, so they cannot carry over to your floor without the same layout and volume.

A separate vendor case study for a beverage distributor reported that a redesigned person-to-goods layout with nine autonomous mobile robots raised units picked per hour by 93% and cut labor by 21% within six months. That is a relative gain with no absolute rates, so it shows direction, not a figure to copy into a plan.

Compare any vendor figure against your own manual baseline. If your pickers already move a high volume of units per hour, a relative gain says less than it appears to say. The baseline measured on your own shifts sets the reference point, not a number from another building.

Use these figures to write questions for vendors, not to set your unit count. Ask for measured rates under conditions like yours, the layout assumptions behind them, and references you can call. Any unit count that cannot be traced to a measured rate should stay out of the plan.

Which building conditions change the mix?

Floor area is the first thing people quote, but 60,000 square feet alone rarely settles the mix. Layout matters more: aisle widths, dock positions, the number of pick faces, and whether zones are separated by doors, ramps or floor changes. A long, clear main aisle suits transport units. A building with many cross-aisles and tight pick zones may need a different combination.

Picking method is the next question. A station that brings totes to a picker removes walking time, which is why picking is often the first place to look when a backlog persists after transport is covered. Goods-to-person equipment adds fixed installation work, so weigh it against the same peak-hour data you would use for transport units.

Charging is the constraint most often missed on two shifts. Every unit needs time on a charger or a battery swap plan, and that time must fit between tote waves without leaving a gap in coverage. Forklift count rarely drops to zero either, so keep enough counterbalance trucks for putaway, replenishment and trailers.

Shift patterns matter too. If the second shift has fewer supervisors or a different mix of experienced staff, the fleet may need different route rules or a different exception process on that shift. Write the shift differences into the site assessment so the plan does not assume one crew all day.

How should a site run the first pilot?

Start with one or two leased units on the highest-volume tote route and run them across both shifts for several weeks, including a normal peak day. A pilot that runs only on day shift says nothing about the second shift, where charging gaps and staffing differences tend to show up.

Track the same measures before and after: tote moves per hour by lane, peak-hour backlog, faults or stops per shift, and time spent waiting for a unit or a forklift. Compare them against the manual baseline from the same weeks, because volumes move with seasonality and a short window can mislead.

Set the success test before the pilot starts. Agree on the peak-hour backlog target, the acceptable number of faults per shift and the minimum share of tote moves the pilot route must carry. A pilot without an agreed test tends to end with an opinion instead of a decision.

Earn the next units with data. If the measured rate covers the peak, the fleet plan follows the numbers. If it does not, the same data points to a layout, charging or picking fix, and that fix belongs in the plan before more units are added.

What should you check before deciding?

The answer changes with a few site facts. Walk the floor with the questions below before any fleet size is discussed, and record the answers for each shift separately, since the two shifts often differ in volume, staffing and charging time.

If most tote moves follow a handful of repeat routes, autonomous mobile robots are a strong candidate for the first phase. If routes change daily, or picking is the real bottleneck, the first step may be a picking change or more forklift coverage, with robots following later.

Ask how the floor changes over the year. Seasonal stock moves, new racking and changed dock doors all alter routes. A fleet mapped once and never updated loses value fast, so write the remapping process into the plan rather than treating it as an afterthought.

Ask the same questions of any vendor, including us. A useful answer names the assumptions behind a unit count and states what a pilot must prove before the fleet grows.

How does getting a leased unit work with us?

We are an OEM-neutral integrator. We choose robots across manufacturers that fit your floor, lease them through financing, deploy and integrate them, train your staff, and service every unit under our service plan. One partner handles the whole lifecycle, backed by regional service engineers in every state we service.

The first step is a free site assessment, including site mapping and a review of routes, charging space and systems links. The lease quote, with the payment and terms, comes after that assessment, so it reflects your floor rather than a generic setup. If the assessment shows that forklifts or a picking change fit better, we will say so.

Training is part of the start, not an extra. Supervisors need to know how units behave in traffic, how to clear a stop and when to call for service, and the floor crew needs a clear rule for exceptions. The quote lists what training and service cover, so you can see the scope before you commit.

Many operations start with one leased unit on a pilot route and add units in steps. That keeps the first commitment small and lets measured data set the rest of the fleet plan.