Executive Summary
Every warehouse has one of a number of order picking methods, whether it has been formally designed or simply evolved over time. The question is not whether there is a picking strategy, but whether it is appropriate for your warehouse layout, order volume, SKU profile and customer requirements.
The wrong way can increase travel time, picking errors, labour costs, order delays, without making the underlying problem obvious.
In this guide, we will cover the five most common warehouse picking strategies used in manufacturing, distribution, and fulfillment operations.
The 5 Warehouse Picking Strategies: When Each Makes Sense
The right method depends on volume, layout, SKU profile and system capability
Discrete picking
A single picker completes an entire order before moving on to the next order. This is the most simple way and is good for low order volumes.
Batch picking
A single picker visits the warehouse once to pick items for several orders. This decreases travel time when a lot of orders contain the same SKUs.
Zone picking
The warehouse is divided into zones and each picker works in a particular zone. Orders are picked by zone until all required items are picked.
Wave picking
Work is released to the floor in scheduled waves (based on carrier pickup times, labor available, shipping priorities, or production schedules) rather than released immediately for picking.
This is one of the most efficient ways to work in a high-volume warehouse.
Cluster picking
A picker takes multiple totes or containers and picks multiple orders at one time on one pass through the warehouse, keeping each order separate during the pick.
| Method | How it works | Best for | Main downside | WMS required? |
|---|---|---|---|---|
| Discrete Picking | One picker completes one order at a time. | Low order volume and simple operations. | High travel time as volume increases. | No |
| Batch Picking | One picker gathers items for several orders in one trip. | Orders with many common SKUs. | Less effective when orders vary significantly. | Helpful but not essential |
| Zone Picking | Pickers work only within assigned warehouse zones. | Large warehouses with many SKUs. | Requires coordination between zones. | Usually |
| Wave Picking | Orders are released in planned groups based on shipping priorities and warehouse capacity. | High-volume operations with scheduled shipping windows. | Requires careful planning and wave management. | Yes |
| Cluster Picking | Multiple orders are picked simultaneously using separate totes or containers. | E-commerce and high order volumes. | Can become complex without good picking technology. | Usually |
This is a good comparison to help evaluate wave picking vs batch picking and the wider variety of warehouse picking strategies available today. The ideal solution for you will depend on the order volume, warehouse layout, SKU characteristics, labor availability and the capabilities of your warehouse management system.
Let’s
Talk
Book your manufacturing AI strategy session
Discrete and Batch Picking: The Starting Point for Most Operations
For many warehouses, batch picking is not the starting point in warehouse operations. Most begin with discrete picking because it is easy to implement and requires very little coordination. As order volumes increase, however, the limitations of that approach become apparent and many operations move to warehouse batch picking.
Discrete picking, the default approach: a single order picker is responsible for one sales order and picks all of the items required for that order before returning to the staging or packing area. It is the simplest picking method and does not require a warehouse management system, coordination between pickers or sorting after picking.
Where discrete works: facilities processing fewer than 50 orders a day, warehouses with high-value or fragile products requiring individual attention, or companies fulfilling complex orders with many different SKUs that are difficult to separate later.
The travel penalty: the most important disadvantage is the time of travel. For each order, each picker walks the entire warehouse and often passes the same storage locations multiple times. As the order volume increases, travel is the biggest source of wasted labor, a picker walking four miles in a shift and spending a lot of that distance retracing the same routes.
Travel, not picking, is where discrete picking loses the shift
Batch Picking: First Step to Efficiency
Warehouse batch picking allows one picker to pick for several orders at once, reducing unnecessary travel. The picker will collect all the required items for several orders on one trip through the warehouse and sort them into individual customer orders at the staging area.
Where it fits
This approach works well in high volume operations with multiple orders with the same or overlapping SKUs, especially in e-commerce and B2C fulfillment settings.
A common comparison is batch picking vs wave picking, and batch picking is usually the first improvement that warehouses make because it reduces travel quite a bit without the complexity of wave management.
The sort burden
The hardest part is after the picking has been completed. Items picked for several orders need to be correctly sorted either at a dedicated sort station or on a multi-tote picking cart. As the diversity of SKUs increases this sorting process is more time consuming and can offset much of the travel time saved.
Getting batch size right
The batch size is also important. In many operations, a batch of 5-15 orders is a good compromise between efficiency and accuracy.
Sorting errors increase in frequency after the batch is larger than about 20 orders. Ultimately, the ideal batch size is a function of the number of order lines and the amount of SKU overlap between customer orders.
Zone Picking: When the Warehouse Is Too Large for One Picker per Order
Zone picking involves dividing the warehouse into physical zones and assigning each picker to a zone. An order passes through several zones; each picker picks only items for their zone, then hands the order off (physically or electronically) to the next zone.
A zone picking system typically divides the warehouse into 3 to 8 zones by product type, inventory speed, storage requirements or temperature conditions such as frozen, ambient and cold chain zones.
Each order only passes through the zones that have the items it needs. The picker in Zone A finishes their assigned picks and then the tote or pallet moves to Zone B and the next picker adds their items. The process continues until all the order lines have been fulfilled. The last zone or a central sort area brings together multi-zone orders prior to being taken to packing.
Zones cut travel, but an uncoordinated release just moves the bottleneck
Two Popular Methods for Zone Order Picking
- The tote physically moves in sequence from zone to zone.
- The management of this approach is simple, but bottlenecks can be created since the throughput of the whole operation depends on the slowest zone.
- All zones pick simultaneously, with items from each zone brought together at a merge point or consolidation station.
- This increases throughput, but needs more coordination, such as sortation conveyors or automated consolidation.
Zone picking is most effective in large warehouses where it would be inefficient to walk the entire space for each order. It also works well when product categories require specialization such as frozen goods, hazmat, high value or bulky inventory. Warehouse zone picking is often used in multi-client 3PL operations, as different product areas may have different access requirements.
Zone picking, however, does not do away with all bottlenecks. It reduces the travel time for pickers. However, if the releases of waves are not coordinated, and zones get their work in irregular bursts, pickers may wait around. Good wave management, coupled with zone picking, results in more consistent throughput than zone picking alone.
Not Sure Which Picking Strategy Your Volume Justifies?
20 minutes. Tell us your daily pick volume, warehouse size, SKU profile and carrier cutoff structure, and we will tell you whether the answer is batch, zone, wave or a combination.
Wave Picking: Coordinating Order Release for Maximum Labour Efficiency
Wave picking is a warehouse order fulfillment strategy in which pick work is released to the floor in scheduled batches, called waves, instead of as orders arrive. Each wave is timed to co-ordinate pick activity between zones, carrier pick up cut-offs and available labour capacity.
Unlike batch picking, which combines orders to be picked by one person to reduce travel time, wave picking controls the timing of the work being released to the picking floor. It is a coordination strategy and not a mere consolidation strategy.
A wave can be a discrete pick, batch pick or zone pick. Wave management manages the timing and flow of work, and the picking method within the wave manages how each individual picker completes their work.
Zone picking is not necessary for wave picking in a warehouse. A single picker can receive one batch and complete it as a wave of orders. However, most high-volume operations combine warehouse wave picking with zone picking to maximize throughput in large facilities.
Wave picking controls when work reaches the floor, not how it is picked
Wave Release Criteria
Wave release criteria generally consist of three categories:
Carrier cutoff-driven waves
The WMS groups orders by carrier departure times and releases each wave sufficiently early so that picking and packing is completed before pickup. Most operations run 3-5 waves per shift, scheduled backwards from carrier cutoffs.
Labour balanced waves
The WMS calculates workload per wave and sizes releases based on available picker capacity. This avoids overloading the team at the beginning of the shift and minimizes idle time at the end.
Zone-balanced waves
The system schedules waves so that each zone gets similar pick density, avoiding a situation where one zone becomes a bottleneck while other parts of the system are under-utilized.
Effective WMS wave picking requires three capabilities:
- Configurable wave rules based on cutoff times, carrier priority and zone load
- Sufficient order volume to justify scheduled releases
- Visibility into carrier schedules, order priorities, and zone capacity
Smaller operations running less than 50 to 75 picks per shift often find that wave management creates more coordination overhead than throughput gains.
Let’s
Talk
Book your manufacturing AI strategy session
Cluster Picking, and How the Five Methods Compare
Cluster picking is a batch picking method that uses multi-tote sorting at the picking location. Rather than picking items for multiple orders and then sorting them, a picker uses a multi-tote cart that he or she carries or pushes or a scan-directed device that identifies the right order tote during the pick. This means that sorting happens on the fly during the pick walk rather than as a separate downstream step.
Cluster picking moves the sort into the pick walk instead of after it
Cluster picking is effective for medium volume operations with high SKU variation where traditional batch picking presents sort accuracy challenges. Cluster picking removes the need for end of line sorting as each item is assigned directly to its destination tote, increasing order accuracy.
This typically means multi-tote carts, scan-enabled devices and WMS support to pick multiple orders simultaneously. Most operations cap clusters at between 3 and 12 orders, since higher clusters increase the chances of pick confirmation errors.
Mid-market manufacturers looking for cluster picking Odoo and similar ERP-specific configurations often want to add cluster picking capabilities to their existing systems versus deploying a separate WMS. These workflows can be supported by properly configured ERP platforms with warehouse extensions.
| Method | What it optimizes | Best operation profile | Main limitation | WMS/ERP requirement |
|---|---|---|---|---|
| Discrete | Order integrity | Low volume, complex orders | Travel time scales with volume | None; paper possible |
| Batch | Travel efficiency | High volume, high SKU overlap | Sort burden at end | Basic WMS or paper |
| Zone | Floor coverage and specialization | Large warehouse, category separation | Bottleneck zone limits throughput | Zone configuration in WMS |
| Wave | Labour and carrier timing | Multi-zone, high volume, carrier-critical operations | Requires wave management logic | Wave management WMS feature |
| Cluster | Sort accuracy during pick | Medium volume, high SKU diversity | Cart management complexity | Multi-order scan device |
Understanding the differences between order picking methods allows operations teams to select warehouse picking methods that are appropriate for their order profiles, labour and system capabilities.
AI-Driven Wave Management: What Dynamic Wave Picking Looks Like in Practice
The standard wave picking operation is based on fixed rules, that is, waves are released according to a schedule and sized by rule. Wave management based on AI replaces static rules with real-time optimization, changing wave size, sequencing and zone assignment as conditions change during the shift.
Dynamic sizing replaces a fixed wave that either overloads or starves the floor
What AI Changes About Wave Release
- Dynamic wave sizing: machine learning models are used to predict pick completion times using picker performance, zone density, and remaining order volume. The system adjusts releases to ensure consistent throughput, instead of releasing fixed-size waves that either overload or underutilize the floor.
- Carrier cutoff optimization modifies the wave timing to changing conditions. The system can automatically change the wave sequencing when a carrier delay of 30 minutes happens, removing the need for manual intervention.
- Zone load balancing distributes the pick density over zones and avoids bottlenecks by shifting work around when a zone falls behind.
- Pick path and anomaly detection: AI can also optimize pick paths within each wave based on aisle congestion, inventory locations and individual picker speed. Continuous anomaly detection identifies waves that are deviating from predicted completion rates, enabling supervisors to intervene before carrier deadlines are missed.
A Layer On Top of the ERP or WMS You Already Run
- GrayCyan adds a layer on top of your current ERP or WMS in operation and develops AI wave management layers. Linking wave release logic with scanner data, carrier ETAs and labour performance data results in a wave program that adapts to the operation rather than forcing the operation to a set schedule.
- The same applies to broader operational data integration. For example, GrayCyan helped Access Industrial reduce engineering proposal time from 8 hours to 30 minutes by integrating workflows and automating manual processes.
- Similarly, when incorporated into systems, data-driven optimization can boost warehouse throughput.
The wave program adapts to the operation instead of the operation adapting to a schedule
Let’s
Talk
Book your manufacturing AI strategy session
Frequently Asked Questions: Wave Picking and Warehouse Picking Strategies
Need to Optimize Picking Throughput Without Replacing Your ERP or WMS?
Tell us about your current picking method, daily pick volume and carrier cutoff structure. Built on the ERP or WMS you already run, GrayCyan will show you what wave management, zone picking optimization or AI-driven pick path routing looks like, and increase warehouse throughput without disrupting the systems and processes your team already uses.
