Can Rack-Climbing Robots Use Your Existing Warehouse Racking?

Can Rack Climbing Robots Use Existing Racks?

Some rack climbing robots can use suitable existing warehouse racking after an approved retrofit. Compatibility must be established for the specific rack, robot, load carrier and layout. A supplier's reference to standard racking does not establish that every rack already installed in your building can remain unchanged. Climbing channels, container supports, clearances, rack condition and the engineering of the modified structure all affect the answer.

This distinction matters when comparing a warehouse automation retrofit with a new storage installation. Keeping the building, keeping the rack frames and keeping the complete storage arrangement are three different project outcomes. Each preserves a different portion of the original investment.

This guide concerns robots that climb storage structures to retrieve containers or cases and deliver them to workstations. It does not assume that the robot carries a complete pallet simply because the original structure was sold as pallet racking. Buyers should establish the handled load unit before discussing storage density, robot numbers or installation speed.

Why the Retrofit Question Matters in 2026

Recent announcements make this a timely procurement question. On September 8, 2026, Hai Robotics announced its selection for a European fashion fulfillment project involving more than 1,500 HaiPick Climb robots. The announcement describes a planned deployment; its scale and throughput figures should not be treated as independently verified operating results.

Earlier, on August 27, the company reported more than 10,000 HaiClimber robots contracted for customer projects worldwide. Contracted equipment is a different measure from equipment delivered, commissioned or running in production.

There is also operating evidence from another supplier. Exotec's September 15 announcement with GXO describes a deployed system serving Guess in the Netherlands, including 127 robots and eight goods-to-person stations. Its October 6 First Supply update discusses operations after more than a year. These supplier reports support commercial relevance, but neither establishes that a reader's existing racks can be reused.

The useful response to this momentum is a more specific buying question: which existing assets can remain while still delivering the required storage and order service? That question applies to a spare-parts distributor, a fashion fulfillment center and a multi-client warehouse, even when their preferred robotic architectures differ.

Define What You Want to Reuse

When assessing robots for existing racks, ask the supplier to identify retained assets individually. Request a drawing that marks every rack row and its proposed treatment. A sentence promising compatibility with the existing facility is too broad to price or verify.

There are three practical options to investigate: adapt suitable existing rack components, install a new robotic storage zone inside the existing building, or automate selected rows while retaining manual storage elsewhere. A hybrid arrangement can preserve useful capacity for goods that do not fit the robotic system.

The technology examples also need careful interpretation. Hai Robotics' original 2025 launch description explains that climbing channels are attached to one side of compatible industry-standard racking. Its current solution page specifies standard cross-bar racking. Those descriptions establish an interface to be designed and installed, not permission to attach equipment to an arbitrary old frame.

Exotec describes modular racks as part of its Skypod storage system. A modular system can fit an existing building without preserving the original rack components. Procurement comparisons should therefore record exactly what is supplied, retained and removed for each offer.

There is historical evidence of another approach: Maersk's 2023 BionicHIVE pilot announcement specifically discussed climbing existing warehouse racking. That dated pilot is useful evidence of the retrofit concept, not proof of current commercial availability or universal compatibility.

Examine the Rack-to-Robot Interface Before Counting Locations

Start the warehouse racking retrofit review with an asset record: original drawings where available, rack manufacturer and component series, current configuration, load notices, inspection history and repairs. Mixed components or undocumented alterations need resolution before a supplier can reliably define what remains.

The Rack Manufacturers Institute's discussion of automation design highlights alignment, dynamic behavior, slab loading and fire protection as important retrofit considerations. The applicable requirements depend on the chosen system and site. A floor or rack tolerance quoted for one automated installation should not become a universal specification for all climbing robots.

Separate interface geometry from structural capacity

A robot may have a suitable mechanical interface while the proposed rack configuration still requires engineering changes. Conversely, a structurally adequate frame may lack the geometry needed for a particular climbing channel or container transfer. Both questions need an answer.

Ask the robot supplier to provide its interface requirements and the responsible rack engineer to assess their application to the retained components. The review should identify the loads introduced by the robot, its carrier and the stored inventory, including relevant operating and stopped conditions. Manufacturer-approved connection details belong in the design package; they should not be improvised during installation.

Clearances should be checked with the installed accessories included. Beam connectors, guards, labels, base plates and bracing can occupy space absent from a simplified sales drawing. A useful layout shows the complete robot and carrier envelope at storage locations, transitions and service positions.

Keep maintenance access in the usable layout

A dense storage block must still support the supplier's approved method for inspecting and recovering equipment. Ask how a robot stopped at height is secured and serviced, which neighboring locations become unavailable, and what access equipment is required. This is a design and service-scope question for the supplier and qualified site team.

The cost of those provisions belongs in the comparison from the beginning. Removing a service aisle from a drawing may improve the displayed location count while making the proposed operating arrangement unacceptable. For the wider building assessment, use the site's AMR site-survey guide; this article's focus is the decision about retaining the storage assets.

Confirm the Load Unit: Pallets, Totes and Cartons Are Different

Existing pallet racking often contains a mixture of reserve pallets, open cartons, shelf bins and oversized products. Converting its structure into robotic storage and retrieval does not mean all those inventory formats become robot-compatible.

Build the initial inventory assessment around the actual unit presented to the machine. Record container dimensions and total weight, including packaging and any tray. Also record bottom condition, rigidity, protrusions, open lids and label location. A carton with suitable outer dimensions may still need support if its base deforms during transfer.

For tote handling robots, determine whether the quoted carrier is supplied by the automation vendor, retained from the warehouse or approved from a third party. If a tray supports the original carton, include the tray's weight, space and return movement in the process. Keeping product packaging can reduce repacking work while still creating a new carrier-management task.

Separate inventory into four working groups:

• Directly compatible: the existing container and contents meet the specified handling conditions.
• Compatible with a carrier: the goods need an approved tray, tote or other support.
• Compatible after a process change: receiving, packing or replenishment must prepare the load differently.
• Outside the selected system's scope: the goods remain in another storage and picking process.

Evaluate these groups by order activity as well as SKU count. A system that accommodates most catalog numbers may cover little daily work if the incompatible items dominate orders. Conversely, a small automated assortment may handle a large share of routine picks.

Request a representative sample that includes normal packaging variation. Testing only newly purchased, perfectly rigid containers misses the condition of stock already circulating through an existing warehouse. The agreed acceptance range should describe what receiving staff can recognize and route consistently.

Compare Usable Capacity and Order Service Together

In goods-to-person automation, the storage layout must both hold inventory and supply the required order mix. Maximum locations, container deliveries and completed order lines measure different things. They should not appear in one comparison column as interchangeable measures of output.

Double-deep storage introduces another planning question: how is a rear container made available when a front container occupies the access path? The answer depends on the selected mechanism and storage policy. Request a demonstration of that specific condition and include any necessary movements in the performance model.

Also ask how replenishment competes with retrieval. A layout that meets the outbound target during a demonstration with no inbound work may behave differently while receiving fresh stock, processing returns and presenting empty carriers. The relevant comparison uses the same order profile and simultaneous activities for every option.

A worked example: retained racks may not provide enough usable capacity

The following is an illustrative planning example created for this guide. The numbers are assumptions, not supplier specifications or measured customer results.

A distributor is evaluating two layouts in the same available area. Retaining and modifying selected racks provides 12,000 nominal tote locations. Replacing the storage block provides 16,000. The operation needs space for a peak inventory of 11,000 totes. For the exercise, both designs use an 85% planning occupancy ceiling to leave room for the agreed operating movements.

Both location counts already exclude positions permanently lost to climbing interfaces, maintenance access and other fixed restrictions. The 85% ceiling is an additional operating assumption applied to the remaining positions.

The retained-rack layout does not satisfy the assumed peak requirement. It could still work if the business deliberately reduces the automated assortment or provides another storage area, but that changes the scope and cost of the proposal. The new layout meets the inventory condition; its order-service capability still needs separate validation.

Now assume the target is 1,200 completed order lines per hour. Four staffed stations demonstrated at a sustained 250 lines per hour each would provide 1,000 lines per hour. A fifth equivalent station would yield an arithmetic total of 1,250, but only if the robots, storage access, replenishment and downstream packing can sustain the supply. Adding locations alone does not close the service gap.

The 85% ceiling and station rates are not general recommendations. Replace them with values supported by the chosen system, actual assortment and representative tests. The example illustrates why retaining steel and meeting the operating requirement are separate decisions.

Price the Conversion Work That Rack Reuse Creates

An ASRS retrofit comparison should separate the cost of retained assets from the cost of making them usable. The purchase price of new racks is visible. Engineering, selective removal, installation around old components, carrier changes and temporary stock handling are easier to omit.

Ask suppliers to price the same functional scope under each layout. The quotation should state what inventory and order demand the system supports, which parts of the building are occupied, and what remains manual. A lower price for a smaller automated assortment is not a saving on an equivalent system.

Existing racks can have significant value where their configuration aligns with the proposed process. That value should appear as avoided work or retained capability. It should not be assumed simply because the steel is already paid for.

For leased facilities, clarify ownership and end-of-lease obligations for the retained racks and added equipment. A system that can be expanded is not automatically economical to relocate. Ask what must be dismantled, re-engineered or replaced at another building, and keep those assumptions separate from normal maintenance.

Plan the Retrofit Around Inventory Migration

For brownfield warehouse automation, the most difficult installation period may occur before the robots start serving orders. Existing stock must be moved, identified and made available to the business while its storage area changes.

Divide the transition into zones with explicit inventory ownership. Each SKU-location combination should have a clear source of truth during the move. If the warehouse management system still directs people to a location under conversion, physical progress can create picking failures before automation has delivered any benefit.

A practical first phase converts a defined inventory group and its associated workflow. Select enough variation to test the proposed scope, rather than choosing only the easiest containers. Include replenishment and returns if those activities belong to the purchased process.

Before moving stock, settle these responsibilities:

1. Who confirms that a zone is empty and available for installation?
2. Who records each container's identity and new location?
3. How are open orders handled when inventory moves between manual and automated areas?
4. Where are incompatible or damaged carriers routed?
5. What constitutes permission to begin normal order release for the converted zone?

Where automated storage exchanges containers with conveyors or other equipment, material identity and transfer completion must agree across systems. The site's guide to material-handoff confirmation covers that interface in more detail.

After the first zone is operating, measure the activities that could invalidate the reuse decision: inaccessible locations, carrier rejection, inventory discrepancies, time spent on manual support and capacity lost to service restrictions. These measures connect the installed rack arrangement to the business outcome.

Expansion should use the findings from that phase. If retained rack geometry creates a recurring restriction, copying the same layout across the building repeats the restriction. Conversely, a successful, documented conversion gives the next zone a much firmer design basis.

What to Request Before Approving Rack Reuse

The final proposal should make the retained-asset decision reviewable. Ask for a marked-up layout, a container compatibility schedule, the approved engineering scope, a migration plan and an operating-capacity model tied to your demand.

The layout should distinguish usable storage locations from nominal positions. The compatibility schedule should identify exceptions and their manual process. The engineering scope should name who accepts responsibility for the robot interface and modified rack configuration. The migration plan should identify the stock movements and service interruptions needed to build it.

Finally, request a comparison against a new storage zone using the same inventory, order and growth assumptions. The outcome may favor retaining most racks, replacing them, or keeping a manual reserve area beside the robotic installation. The right choice is the arrangement that meets the agreed operating requirement at an understood conversion cost.