What is the best order picking automation system?

There is no single best order picking automation system for every warehouse. The right choice depends on throughput requirements, SKU complexity, order profile, labor availability, facility layout, storage density needs, budget, and how much flexibility the operation requires. Common options include shuttle systems, cube storage, goods-to-person AMRs, directed picking AMRs, and mobile manipulators.

How do I choose the right picking automation for my warehouse?

Start by evaluating your order profile, including lines per order, units per SKU, SKU variability, peak volume, required throughput, available labor, and facility constraints. Then compare automation systems based on throughput, density, flexibility, installation time, labor savings, risk, and total cost of ownership. The best system is the one that fits your operational profile, not necessarily the most advanced or most expensive option.

Which picking automation system provides the highest labor savings?

Mobile manipulators typically provide the highest labor savings because they can automate storage, transport, item picking, buffering, replenishment, and sortation. Traditional goods-to-person systems such as shuttles, cube storage, and AMRs automate inventory movement, but item picking usually still happens manually at workstations unless separate robotic picking cells are added.

What is the difference between goods-to-person automation and mobile manipulation?

Goods-to-person systems bring inventory to a human picker at a workstation. Mobile manipulation systems go further by using robots that move through the storage area and pick items directly into order totes. This can reduce travel, reduce congestion, lower the need for fixed picking stations, and automate more of the fulfillment process.

Are shuttle systems good for order picking automation?

Shuttle systems can be a good fit for large warehouses that need high throughput, high vertical storage density, and have the budget and time for a major infrastructure project. They have a long track record and can deliver fast tote retrieval, but they usually require high upfront investment, long installation times, conveyors, lifts, engineered racking, and manual picking stations.

What are the disadvantages of shuttle systems?

The main disadvantages of shuttle systems are high capital cost, long installation timelines, limited flexibility after installation, complex maintenance, and possible single points of failure such as lifts or conveyors. Shuttle systems also usually automate storage and retrieval, but not the item picking itself.

Are cube storage systems good for order fulfillment?

Cube storage systems are a strong fit when very high storage density is the main priority and the order profile works well with tote retrieval from a dense grid. They can be more modular than shuttle systems and do not rely on a single robot, but throughput can depend heavily on SKU slotting, bin accessibility, robot traffic, and port capacity.

What are the limitations of cube storage systems?

Cube storage systems can face throughput limitations when robots need to dig through stacked totes to access the right inventory. They also require fixed grid infrastructure, very flat floors, fire safety considerations, and external picking ports. Like most goods-to-person systems, cube storage usually still relies on human pickers or separate robotic picking cells.

When are goods-to-person AMRs a good choice?

Goods-to-person AMRs can be a good choice for warehouses that need more flexibility and lower upfront cost than shuttle or cube systems. They usually require less fixed infrastructure and can be easier to implement, but they may deliver lower throughput because floor-traveling robots move more slowly and can create congestion around picking stations.

What are directed picking AMRs best for?

Directed picking AMRs are best for operations that want to improve manual picking productivity without redesigning the warehouse. They guide human pickers through the warehouse and reduce walking, but they do not automate the actual picking task. This makes them flexible and relatively low cost, but labor savings are more limited than with goods-to-person or mobile manipulation systems.

What are mobile manipulators best suited for?

Mobile manipulators are best suited for warehouses that want to automate more of the fulfillment process, including item picking. They are especially useful when the operation needs strong labor savings, high throughput, flexible workflows, and reduced dependency on fixed picking stations. Systems like Brightpick Autopicker are examples of mobile manipulation used for robotic order picking, buffering, replenishment, and dispatch workflows.

What are the limitations of mobile manipulators?

The main limitation of mobile manipulators is robotic pickability. Some items may not be suitable for autonomous robotic picking and may require human-in-the-loop fallback. Mobile manipulators may also be less suitable for order profiles where many units of the same SKU need to be picked at once, because human pickers at goods-to-person stations can sometimes pick multiple units more efficiently.

Which picking automation system is best for high throughput?

Shuttle systems, cube storage, and mobile manipulators can all support high-throughput operations, but they achieve it differently. Shuttle systems use fast tote retrieval and conveyors, cube systems rely on dense grid storage and robot traffic, while mobile manipulators reduce travel by picking directly inside the storage area. The best option depends on order profile, SKU distribution, pick density, and whether the operation wants to automate item picking or only inventory movement.

Which picking automation system is best for storage density?

Cube storage systems usually offer the highest storage density because they eliminate aisles and store totes in a compact grid. Shuttle systems and some mobile manipulation systems can also provide high density by using vertical storage. Directed picking AMRs generally do not improve density because humans still need to access the storage area directly.

Which picking automation system is best for brownfield warehouses?

For brownfield warehouses, flexible systems with less fixed infrastructure are usually easier to implement. Goods-to-person AMRs, directed picking AMRs, and some mobile manipulation systems can be better suited than large shuttle or cube systems because they typically require fewer structural changes. However, the best choice depends on floor quality, ceiling height, fire safety requirements, available space, and integration needs.

Do automated picking systems still need human workers?

Many automated picking systems still need human workers. Shuttle systems, cube storage, and goods-to-person AMRs usually bring inventory to human pickers at workstations. Directed picking AMRs still rely almost entirely on human picking. Mobile manipulators can reduce this dependency because they pick items robotically, although some products may still require human fallback.

Can warehouse picking be fully automated?

Warehouse picking can be highly automated, but full automation depends on SKU characteristics, order profile, exception rate, and system design. Mobile manipulators can automate more of the process than traditional goods-to-person systems because they can pick items directly. However, most operations still need some human support for exceptions, non-pickable items, replenishment, maintenance, and process supervision.

What is the most flexible picking automation system?

AMR-based systems are generally among the most flexible because they require less fixed infrastructure than shuttles or cube storage. Directed picking AMRs are especially easy to deploy in existing manual warehouses, while goods-to-person AMRs and mobile manipulators can offer more automation. Flexibility should be evaluated not only by robot movement, but also by how easily the system can scale, change workflows, handle new SKUs, and adapt to future demand.

What is the biggest mistake when choosing warehouse picking automation?

The biggest mistake is choosing a system based on technology category alone instead of matching it to the operation’s real requirements. Throughput, order profile, SKU mix, item pickability, facility constraints, labor costs, growth plans, and integration complexity all matter. A system that works well for one warehouse can be a poor fit for another.

April 29, 2026

Pros and cons of different robotics solutions for order fulfillment

How to choose the right picking automation for your warehouse

Order picking is one of the most labor-intensive and error-prone parts of warehouse operations, accounting for up to 50% of total warehouse costs. That is why more and more warehouses are turning to automation to solve labor shortages, growing order volumes, handle SKU complexity, and meet customer expectations.

Choosing the right system is not just about replacing manual work with robots. It is about selecting the automation architecture that best fits your operation. A fast-growing retailer, an established medical distributor, and a multinational 3PL will each have different requirements, and those differences will determine which type of automation is best suited to the business.

Today’s most common automation options for order picking include shuttle systems, cube storage, AMRs, and mobile manipulators. There is no one-size-fits-all solution. Each come with different trade-offs in throughput, density, flexibility, installation time, labor savings, and cost.

Let’s look at how these most common automated fulfillment systems compare.

Shuttle Systems

In shuttle systems, small robots called shuttles travel through aisles on rails, retrieve totes from storage and bring them to a lift. The lift takes the totes down to a conveyor, which then transports them to picking stations. After all the items for the order are picked, the totes are returned via the same process back into storage.

Pros:

High throughput: Shuttles travel at high speeds and can typically access each tote directly from the aisle, resulting in high throughput. Similarly, each goods-to-person picking station is fed by conveyors, enabling fast cycle times and minimizing congestion.
High density: Shuttle systems can reach heights of up to 60 feet, which makes them space efficient for large warehouses with lots of vertical space.
Long track record: These systems have been on the market for 30 years, with lots of established vendors able to supply and service them.

Cons:

High costs: Shuttle systems require significant capital investment to pay for engineered racking, conveyors and lifts. The cost of a shuttle system typically starts at $5 million and can easily run into the tens of millions. In addition, the complex machinery is costly to maintain and each installation requires extensive spare parts inventory to minimize downtime in case of failure.
Long installation times: It typically takes 1-2 years to install a shuttle system given the extensive infrastructure required and long lead times from suppliers.
Limited flexibility after installation: Shuttle systems can be configured for different warehouse layouts, but once installed, they are not easy to move, reconfigure, or scale without significant infrastructure rework.
Single points of failure: A broken lift or conveyor belt can cause the whole system to stop for an extended period of time.
Picking is still manual: Shuttle systems retrieve goods efficiently, but item picking is still done by humans at the picking stations. While these can be replaced by fixed robotic cells, these add additional cost and complexity to the system.

Mobile manipulators

Mobile manipulators are robots that not only store and transport inventory, but also pick items directly. They can be either floor-traveling AMRs or grid-based robots that move on a fixed grid above the shelving.

Each robot carries one or more order totes and picks items into them within the storage area. Most mobile manipulators also include human-in-the-loop fallbacks for items they cannot pick autonomously.

Pros:

High throughput: Mobile manipulators pick orders directly inside the storage area, significantly reducing travel distances as the robots do not need to travel back and forth to fixed picking stations for each pick. This speeds up picking, reduces congestion, and increases throughput compared to traditional goods-to-person systems such as shuttles or cube storage.
Lower cost: Thanks to the speed and efficiency gains from mobile manipulation, fewer robots are typically needed to achieve a given throughput. This reduces costs and makes these systems more economical, with a higher ROI.
Maximum labor savings: Because each robot handles order totes directly, mobile manipulators can automate the entire fulfillment journey, from storage and picking to buffering and sortation. This eliminates the need for downstream automation, maximizes labor savings, and can even enable lights-out operations.
High density: Mobile manipulation systems can reach heights of up to 40 feet, or 12 meters, and do not require conveyors or large areas of empty space around the system for robot travel.
No single point of failure: Each mobile manipulator operates independently and handles workflows end to end. This means that if one robot breaks down, it does not affect the others, and its work can easily be reassigned to another robot.

Cons:

Limitations of robotic picking: Some SKUs may not be suitable for robotic picking, which requires human-in-the-loop fallbacks where mobile manipulators bring these items to human picking stations for processing.
Limitations around order profiles: Because mobile manipulators pick items one by one, they are less suited for operations where each order requires many units from the same SKU. These profiles are often better served by traditional goods-to-person automation, where a human picker can pick multiple items at once. As a rule of thumb, the transition point is typically around an average of more than five units picked per SKU.

Cube storage

Cube storage systems store totes in a dense, cube-like grid structure with robots traveling on top of the grid. The robots retrieve storage totes from inside the cube and deliver them to external picking ports. After the required items are picked, the totes are returned back into the cube.

Pros:

High density: Cube storage offers the highest density because it eliminates the need for aisles or conveyors. However, most cube systems are limited to a height of only 20 feet / 6 meters.
No single point of failure: Each robot operates independently and one breaking down does not affect the others.
Modular scalability: Cube systems are generally more modular than shuttle systems. Capacity and throughput can often be increased by adding more robots, ports, bins, or grid sections, though doing so often requires adding fixed infrastructure and temporarily pausing the system.

Cons:

Throughput limitations: The throughput a cube system can deliver depends heavily on the operation’s order profile. If a small number of SKUs account for most of the volume, those items can be stored near the top of the cube and accessed quickly. However, when demand is spread more evenly across a wider range of SKUs, robots often need to dig through totes to retrieve the right items, which can significantly constrain throughput. Cube systems can also suffer from congestion, particularly around picking stations and ports, when too many robots are operating on the grid.
Fixed infrastructure: Cube systems are more modular than shuttle systems, but still rely on a fixed grid structure. Moving, expanding, or reconfiguring them requires extensive planning, space, and installation work.
Strict facility requirements: Cube storage requires extremely flat floors and upgraded fire safety systems, which add costs especially for brownfield warehouses.
Picking is still manual: Cube systems retrieve goods efficiently, but item picking is still done by humans at the picking stations. While these can be replaced by fixed robotic cells, these add additional cost and complexity to the system.

Goods-to-person (G2P) AMRs

Autonomous mobile robots (AMRs) retrieve totes or racks from storage and transport them to goods-to-person picking stations. After all items for the order are picked, the AMRs return the remaining inventory back into storage.

Pros:

Flexibility: AMRs do not require heavy fixed infrastructure such as conveyors or engineered racking, which makes them simpler to implement and scale.
Lower upfront cost: AMRs typically cost less to implement and maintain than shuttle or cube storage, primarily because they do not need as much supporting infrastructure.
No single point of failure: Each robot operates independently and one breaking down does not affect the others.
Density: Some AMRs can reach heights of up to 40 feet, or 12 meters, enabling high storage density. However, overall density is still limited by wider aisles and the empty space required for robot navigation, particularly around goods-to-person stations.

Cons:

Low throughput: Goods-to-person AMRs typically deliver lower throughput because floor-traveling robots move more slowly and often face congestion around picking stations, which further reduces speed and efficiency.
Navigation maintenance: Many AMRs rely on fixed floor-based guidance systems, such as QR codes, to navigate the warehouse. This adds maintenance complexity and cost, as navigation markers must be kept clean, visible, and properly labeled at all times.
Safety: Goods-to-person AMRs are typically not safe around humans given their heavy loads, which means the automation area needs to be fenced off.
Picking is still manual: G2P AMRs automate the movement of goods to workers, but item picking is still done by humans at the picking stations. While these can be replaced by fixed robotic cells, these add additional cost and complexity to the system.

Directed Picking AMRs

Mobile robots guide human pickers to remove items directly from pallets or storage shelves and place them onto autonomous mobile robots. The AMRs then move through the warehouse from one pick location to the next until the order is complete. This reduces travel distances for human pickers and increases labor productivity.

Pros:

Flexibility: Directed picking AMRs require minimal changes to the warehouse because human pickers continue picking directly from pallets or storage shelves, unlike other automated fulfillment systems that require totes for storage. This makes directed picking robots simple to install and easy to reconfigure when warehouse layouts or workflows change.
Low cost: These types of AMRs typically cost less than other forms of automation because they do not require totes or specialized shelving. They also require fewer robots, since they move directly from one pick location to the next rather than traveling back and forth to picking stations like goods-to-person systems.
No single point of failure: Each robot operates independently and one breaking down does not affect the others.

Cons:

Low labor and cost savings: Directed picking solutions deliver lower labor savings than other types of robotic automation because human pickers still need to retrieve items from storage and walk between picks. In directed picking systems, human pickers typically pick 80 to 120 order lines per hour, compared with 300 or more lines per hour at a single goods-to-person station.
Replenishment is still manual: Unlike other automated fulfillment systems, directed picking AMRs do not automate putaway and still require humans to replenish inventory into storage.
No improvement in density: Directed picking AMRs do not improve storage density compared with manual picking. Human pickers still need to operate within the storage area and pick directly from shelves, which limits how high shelves can go and how narrow each aisle can be.
Picking is still fully manual: Directed picking improves the picking process by reducing walking and guiding workers through tasks, but it does not automate the actual item picking. The core labor remains human, so labor savings are more limited than with mobile manipulation or goods-to-person systems.

About Brightpick

Brightpick is a leader in AI-powered robotic solutions for warehouses. The company’s multi-purpose AI robots enable warehouses of any size to fully automate order picking, buffering, consolidation, dispatch, and stock replenishment. The award-winning Brightpick solution takes just weeks to deploy and allows companies to keep their warehouse labor to a minimum. With offices in the US and Europe, Brightpick has more than 250 employees and hundreds of AI robots deployed with customers.

Get started with Brightpick

Pros and cons of different robotics solutions for order fulfillment

Other resources

Most Commonly Asked Questions About Gridpicker

How Good Data Makes Running a Warehouse Easier

Brightpick Enters Automotive Market in Strategic Partnership with NAPA