Automated Storage System Planning: Put Workflow Before Equipment

An automated storage system can increase usable storage and reduce handling, but only if it matches the goods, order profile and routes through the warehouse. Start by mapping how inventory arrives, moves into storage, gets replenished and is picked; then define the capacity and retrieval performance the operation needs. That evidence helps you compare pallet cranes, shuttle systems, mini-load systems and vertical lift modules without selecting equipment on automation level alone. It also exposes constraints such as building structure, software interfaces and peak demand before they become expensive design changes.

Map the workflow before comparing equipment

Begin with the whole process, not just the storage aisle. Trace goods from receiving and any inspection or labelling through putaway, storage, replenishment, picking, packing and dispatch. Include returns and damaged or unidentified stock if these pass through the same operation. An automated store may handle only one part of this chain, so identify where work will start and finish at its interfaces.

Record where products wait, where staff or equipment travel, and where work is handed between processes. For example, a system that delivers items quickly to a picking station may not improve order completion if packing capacity is already the constraint. Likewise, fast retrieval is of limited value if replenishment cannot keep the pick locations supplied.

Write down the intended operating pattern. Is the system supporting full-pallet movements, case picking, each-item picking, production supply, or several of these? Will orders be released continuously, in waves, or in batches? The answers affect the storage technology, the number and location of workstations, and the conveyor or other handling equipment needed around it.

Use inventory and demand data to set requirements

Build the design brief from actual warehouse data where possible. Averages alone can conceal the conditions that determine whether a system performs well. Review demand by season, day and shift, and distinguish routine activity from peaks and unusual order mixes.

• Inventory profile: Record each SKU’s dimensions, weight, packaging, handling restrictions and storage conditions. Note whether products are stable in standard containers or vary in shape and size.
• Movement profile: Identify which SKUs move frequently, which are slow-moving, and how often items are replenished. Consider whether demand is predictable or changes substantially.
• Order profile: Measure order lines, units per line, orders per release and the mix of single-item and multi-item orders. Include returns or urgent orders if they affect daily operations.
• Operating pattern: Document shifts, working hours, staffing assumptions and the desired service level during typical and peak periods.
• Existing constraints: Note current storage methods, travel distances, bottlenecks, manual workarounds and inventory accuracy issues.

Check data quality before using it to size the system. Missing product dimensions, inconsistent unit-of-measure records or outdated stock data can lead to unsuitable containers, incorrect capacity assumptions or unreliable replenishment plans. If data is incomplete, identify the uncertainty and decide how it will be resolved before approving a final design.

Separate storage capacity from retrieval performance

Storage capacity describes how much inventory the system can hold. Retrieval performance describes how reliably it can deliver the required items to the next process. These are different design questions. A dense installation may provide many storage positions but still create queues at its access points or picking stations.

Set requirements for both. Define the number and type of storage locations needed, including space for replenishment, temporary buffers and exceptions. Then specify the required flow in terms that reflect the operation, such as pallets moved, order lines supplied or containers delivered over a defined operating period. Use demand data to set these targets rather than relying on a supplier’s headline rate alone.

Ask how the proposed system performs under the expected product and order mix, not only in an idealised scenario. The design should account for replenishment activity, simultaneous tasks, station capacity, equipment interruptions and recovery after a stoppage. Request a clear explanation of the assumptions behind any performance estimate, including the number of operators and workstations it requires.

Compare automated storage system types by the job they do

Automated storage and retrieval systems (AS/RS) cover several technologies. Names and configurations vary, so compare the actual load unit, movement pattern, access method and operating limits offered in each proposal.

The table is a starting point, not a substitute for a layout and process design. A unit-load system is relevant when pallet movements dominate; it is not automatically the right choice simply because pallets are present. A vertical lift module may fit a controlled range of small items, but it may not suit a broad mix of oversized goods or a process needing many items at once. Shuttle configurations differ, so verify the exact load and movement method rather than relying on the label.

Check the building and the interfaces around the system

An automated storage system has to fit the site as well as the stock. Confirm available floor area, clear height, column positions, access routes and the locations of receiving, picking and dispatch. The equipment layout should preserve the routes needed by people and other material-handling equipment, including access for installation and maintenance.

Have qualified professionals assess the building and equipment design together. Structural capacity, floor conditions, fire protection, utilities and maintenance access can affect what is feasible. For a New Zealand site, confirm the applicable building, structural, fire and workplace safety obligations with the relevant design professionals and authorities before committing to the installation.

Plan the handoffs into and out of storage. Pallet conveyors, lifts, workstations, packing areas and manual handling points can all influence the system’s real output. If one station is expected to serve several processes, model how work will be prioritised and what happens when that station is unavailable.

Define software responsibilities and exception handling

Automation depends on dependable inventory and task information. Establish how the warehouse management system (WMS), equipment control system and any other business software will exchange orders, stock movements and status updates. In some designs, the WMS manages inventory and work allocation while a warehouse control system coordinates equipment; the division of responsibilities varies by solution and should be explicit.

Agree how the system will handle short picks, damaged goods, missing labels, blocked locations, order changes and inventory discrepancies. These events need a defined path rather than an improvised workaround. Check that operators can identify the cause of an exception, correct it safely and return inventory or equipment to normal operation.

Ask for practical demonstrations or documented scenarios covering routine work and exceptions. Confirm who owns interface testing, how data errors are diagnosed, and how work continues if a software link or part of the equipment is unavailable. A system that performs well in normal conditions can still disrupt the warehouse if its recovery process is unclear.

Compare total operating cost, not equipment price alone

Assess the full cost of the proposal across its working life. In addition to storage equipment, account for design and engineering, software and integration, building changes, utilities, installation, commissioning, training, maintenance and replacement parts. Include the cost of planned downtime and the operational impact of an unplanned interruption.

Set expected benefits against the same workflow assumptions used to design the system. Potential gains might include reduced travel, more usable storage space or steadier order processing, but they depend on the site and operating model. Test whether the business case still works if demand grows more slowly than expected, the product mix changes or the system achieves less throughput than the initial estimate.

Ask suppliers to identify the assumptions, exclusions and service requirements in their proposals. Compare like with like: storage positions, supported load units, workstations, interfaces, performance conditions, training and ongoing support. A lower initial equipment cost may not represent lower total cost if essential integration or site work is excluded.

Use a staged decision process

1. Document the current process. Map inventory movement and identify the operational constraint the project is intended to address.
2. Validate the data. Confirm dimensions, weights, order patterns, replenishment needs and peak demand before fixing system requirements.
3. Set measurable requirements. Define storage capacity, retrieval performance, operating hours, service expectations and exception processes.
4. Develop the site and system design together. Check building constraints, access, interfaces, workstations and material flow as one layout.
5. Compare proposals on consistent assumptions. Review performance conditions, included scope, software responsibilities, support and total cost.
6. Test before handover. Agree how normal operation, peak scenarios, exceptions, inventory accuracy and recovery will be checked during commissioning.

Before signing off, make sure the design team can explain how the system meets the defined workflow, not just how much stock it can store. Ask which assumptions would change the design and what operating limits staff need to understand.

Common planning mistakes to avoid

• Choosing technology before measuring demand: A familiar or highly automated option may not match the loads, order mix or access pattern.
• Sizing only for average activity: Peak periods, replenishment and uneven demand can create queues that average figures hide.
• Ignoring surrounding processes: Faster retrieval will not resolve a constraint at packing, dispatch or replenishment.
• Assuming all inventory fits a standard load unit: Confirm container and product compatibility, including unusual dimensions and handling requirements.
• Leaving exceptions until commissioning: Decide how errors, damaged stock and equipment downtime will be managed during design.
• Comparing proposals with different boundaries: Check whether each supplier includes the same interfaces, site work, training and support.

Choose an automated storage system only after the workflow, inventory profile and performance targets are clear. Then compare technologies and supplier proposals against those requirements, including building fit, software, exceptions and total operating cost. That sequence makes it easier to identify a solution that supports the warehouse rather than creating a new bottleneck.