Picking methods: matching the retrieval pattern to the order profile

The enemy is distance, not dexterity

In a conventional facility, the time spent reaching for goods is a small fraction of a picker's shift; the rest is spent getting from one location to the next and dealing with paperwork or devices. Every pattern described below is an attempt to spread that travel across more picked lines. This is also why layout and location assignment set the ceiling: no method rescues an arrangement that scatters popular items across the building.

One order at a time

Working a single order from start to finish is simple to supervise, needs no sorting afterwards and makes any error obvious and attributable. It is also the most expensive way to move, because a journey around the aisles yields only the lines of one order. It remains the right answer for large orders that fill a trolley or pallet on their own, and for operations where orders arrive continuously and must leave immediately.

Several orders in one journey

Collecting the same item for many orders in a single visit, or carrying several containers on a cart and filling them as the picker moves, spreads the walk across far more lines. What it creates is a sorting obligation: either the picker sorts into containers at the moment of picking, or the goods are sorted afterwards at a bench or by machine. Where orders are small and repetitive, this trade is heavily favourable; where orders are large and varied, the sorting can cost more than the walking saved.

Releasing work in waves against the clock

Rather than picking as orders arrive, work can be grouped and released in blocks aligned to vehicle departures, carrier collections or shift boundaries. Waves make the workload predictable, allow the fast zone to be replenished between them, and give supervisors a natural checkpoint. Their weakness is rigidity: an order arriving just after a wave closes waits for the next one, which is why many sites run scheduled waves for the bulk of volume and a continuous trickle for urgent work.

How the pick is confirmed changes the error profile

Paper lists are cheap and silent about mistakes. Scanning a barcode at the location and on the item verifies both, at the cost of a device in the hand. Voice systems keep hands and eyes free and suit cold or bulky work, while light-directed displays are fast in dense, stable fast-pick areas but expensive to reconfigure. The decision is less about raw speed than about which errors the operation can least afford and where operators need their hands.

Dividing the floor rather than the order

Large buildings often split picking geographically, so each operator works a defined area and orders are assembled from several contributions. That approach and its batching variants have their own mechanics, sequencing rules and consolidation requirements, covered separately alongside the sortation methods they depend on.

Frequently asked questions

Which method gives the highest picking rate?

Rates quoted for different methods are not comparable, because they count different work. Batched approaches report high lines per hour while deferring sorting into another department, so the honest comparison is total labour from release to a packed, verified order.

Does voice picking reduce errors compared with scanning?

They fail differently. Scanning verifies identity directly and catches the wrong item; voice relies on a spoken check digit at the location and keeps hands free, which suits heavy or cold work. Many sites combine them, using voice for movement and a scan for verification.