When distributors start looking at autonomous mobile robots (AMRs), one question comes up almost immediately: “How many AMRs do I need?”
It’s a fair question. It’s also the wrong place to start.
Most AMR vendors answer it by sizing a fleet. They look at order volume, estimate trips, and recommend large fleets of small tote-carrying robots, each handling tiny payloads, to keep things moving.
At first, that approach sounds logical. In practice, it often leads to more robots than necessary, higher aisle congestion, increased charging infrastructure, more system complexity, and limited impact on actual picking productivity.
The Numina Group takes a fundamentally different approach. Instead of asking how many robots are required, we ask, “How do we design the smartest, highest-throughput picking workflow so you need fewer AMRs to get more done?”
That shift changes everything, including how many robots you actually need.
In this article, we break down why many AMR deployments end up oversized, what actually determines the right robot count, and how our RDS™ Batchbot™ approach delivers higher throughput with significantly fewer robots. You’ll see how payload capacity, batch density, workflow design, and real-time orchestration work together to reduce congestion, cut labor, and accelerate ROI.
Sections in this article:
Autonomous mobile robots, or AMRs, are self-navigating robots used in warehouses and distribution centers to transport goods without fixed paths or manual control. In most operations, AMRs are used to:
Unlike traditional automation, AMRs use sensors, mapping, and software to move dynamically through a facility, adjusting routes in real time based on obstacles, traffic, and task priorities. They are often positioned as a way to increase flexibility and scale operations without major infrastructure changes.
But how AMRs are deployed—including what they carry, how work is structured, and how tasks are orchestrated—determines whether they actually improve throughput or simply add more moving parts.
Most AMR systems on the market are built around low-weight, tote-based robots. Each robot carries a small payload, typically a single tote or limited set of items.
However, to achieve meaningful throughput, these systems often require larger robot fleets, more charging stations, increased aisle traffic, more mapping and machine learning training, and greater fleet management complexity.
And even with all of that, throughput tends to plateau because tote density is limited. When each robot carries so little, the system depends on volume of movement rather than efficiency of movement.
More robots become the workaround for a workflow that is not optimized. It’s why many operations end up with crowded aisles, idle time, and marginal gains in pick productivity.
The number of AMRs required in a facility is driven not by fleet size alone but by how much productive work each trip can carry and how efficiently the operation is orchestrated. Several factors shape that outcome.
Payload capacity has a direct impact on robot count. Small payload AMRs move limited work per trip, which increases the number of trips required. More trips mean more robots.
For example, our Batchbot™ AMRs are designed for high-capacity movement. They can transport:
Because each mission carries significantly more work, one Batchbot™ can often replace 5–10 small AMRs.
Efficiency improves when more work is grouped into each movement.
Batch cart picking allows multiple orders to be processed within a single trip. Each batch cart can hold 30–40 orders, increasing the amount of productive work completed at every stop. This higher batch density, in turn, leads to fewer trips, fewer touches, less consolidation, and faster order completion, which directly reduces the number of AMRs required to support the same throughput.
Robot count is heavily influenced by how the picking process is structured. Key workflow decisions include the following:
If operators are still walking long distances or orders are poorly grouped, adding more robots does not solve the underlying inefficiency.
The most important factor is not the robot itself. It’s actually the system coordinating the work.
RDS™ performs real-time optimization decisions like:
Robot count is directly tied to how effectively this orchestration reduces idle time, travel, and congestion.
Slotting plays a critical role in overall efficiency. Our RDS™ system integrates slotting intelligence into order release and workflow decisions, including the following:
Better slotting and smarter order release reduce wasted motion and allow fewer AMRs to handle the same workload.
Our approach is built around a simple operating model: AMRs transport, people pick and RDS™ orchestrates. This structure shifts the role of the robot from decision-maker to high-efficiency transport resource.
AMRs are not the brain of the operation. RDS™ is.
RDS™ acts as the real-time orchestration layer across labor, robots, order release, batching, and slotting. Instead of robots making isolated decisions, RDS™ delivers deterministic, rules-based coordination across the entire workflow.
Operators use voice-directed picking to execute tasks with speed and precision. RDS™ directs:
This approach turns each operator into an extension of the system’s optimization logic.
Batchbot™ AMRs move high-density batch carts and full pallets rather than individual totes, which reduces aisle congestion, minimizes forklift travel, and supports both each-picking and case-pick-to-pallet workflows.
In each-picking operations, RDS™ can direct picking directly into the shipping container, eliminating totes and reducing repacking labor. In case-pick-to-pallet workflows, Batchbot™ helps automate pallet movement and staging.
The Numina Group’s system centers on high-density batch cart picking, refined through decades of RDS™ WES-WCS development. Each batch cart can hold 30–40 orders. At the same time, RDS™ Voice directs operators to pick multiple orders at each SKU location, improving pick density, travel reduction, fulfillment speed, and consolidation efficiency.
This approach increases the amount of productive work completed at each stop, rather than spreading it across multiple low-value movements. As a result, operators spend more time picking and less time traveling, which is a primary reason fewer AMRs are needed to support higher throughput.
Batch quality is not static. As order profiles, SKU velocity, and operator locations change throughout the day, the system must keep rebuilding better combinations in real time.
RDS™ continuously evaluates a variety of factors to build the most productive batches in real time:
These inputs allow RDS™ to assemble tighter, higher-density batches on the fly, rather than relying on fixed grouping logic or delayed optimization. The result is more work completed per stop, fewer wasted movements, and a steady reduction in the number of AMRs required to maintain throughput compared to systems constrained by low-capacity tote movement.
When payload capacity, batch density, workflow design, and orchestration are aligned, the impact shows up across the entire operation. The gains extend to throughput, labor performance, accuracy, congestion, and the speed at which the investment pays back.
The most immediate result is improved picking output. By moving more work per mission and increasing productive activity at each stop, the operation can process more orders without relying on oversized robot fleets. You can expect:
These gains come from combining high-capacity transport with better batch construction and less wasted motion throughout the picking process.
Better workflow design does more than reduce robot count. It also cuts the amount of labor tied up in travel, staging, forklift movement, and rehandling.
With our RDS™ Batchbot™, you’ll see:
Instead of spending time on low-value movement, operators can stay focused on productive picking tasks, which improves output without adding unnecessary labor.
High throughput means little if execution becomes harder to manage. The Numina Group’s approach keeps operators and AMRs working in sync, which supports both speed and consistency.
Our system consistently delivers 99.98% accuracy, minimal idle time, and continuous synchronization between operators and AMRs. That level of coordination helps reduce missed picks, delays, and task confusion while keeping work flowing across the floor.
Fewer robots only matter if they still produce the required output. Following our model, higher-capacity movement and tighter orchestration reduce the need for excess fleet size and the burden that comes with it. As a result:
That translates into cleaner traffic flow, less infrastructure overhead, and a more manageable operating environment as the system scales.
When throughput rises and unnecessary labor and equipment movement drop, payback tends to accelerate. That’s why many operations see a relatively short path to return on investment.
Most operations see ROI in under two years, driven by forklift elimination, labor savings, increased throughput, and reduced infrastructure requirements. Additionally, in case-pick-to-pallet environments, Batchbot™ can replace dedicated forklift drivers, manual pallet staging labor, and multiple tote-based AMRs per SKU zone. That combination of lower labor demand, fewer supporting assets, and higher productivity is what makes the numbers work.
The system scales quickly. Additional AMRs or batch carts can be added in days without reprogramming. Because RDS™ manages orchestration centrally, adding capacity doesn’t require reworking workflows or retraining the system, and you can increase throughput incrementally without bloating the fleet.
Yes. RDS™ dynamically updates routes and workflows without requiring machine learning retraining. As slotting, SKU velocity, or facility layouts change, the system adjusts task assignments and movement paths in real time to keep performance stable without long reconfiguration cycles.
Instead of moving single totes, Batchbot™ moves high-density carts and pallets, which increases work per mission and reduces congestion. Higher payload capacity, higher batch density, and centralized orchestration allow fewer robots to complete more work.
Most customers see ROI in under two years, driven by labor savings, forklift reduction, and higher throughput. The biggest contributors are reduced travel, fewer handling steps, and lower infrastructure requirements. When those gains compound, the payback period shortens quickly.
The real question is not how many AMRs you need but how to achieve the highest throughput with the fewest robots. But if you want to understand how many AMRs your operation actually needs, the answer starts with workflow design.
Our answer is clear: RDS™ real-time optimization with voice-directed picking and high-capacity Batchbot™ AMRs. This model adapts to changing order profiles, scales without unnecessary fleet expansion, and delivers consistent performance across both each-picking and case-pick-to-pallet operations. The result is higher productivity per operator, per robot, and per square foot.
Contact us today to learn more about modeling the right AMR count based on your throughput, order profile, and picking strategy.
About Numina Group
Numina Group’s RDS™ is a modular and scalable software suite for warehouse automation software solutions optimized to streamline the entire pick, pack, and ship order fulfillment operation.
