Less walking per order,
not faster walking.

BinLogic is a warehouse management system for brands and retailers that run their own fulfillment. What follows is how to choose a pick method for your order profile, where each product should live, and why the walk between picks is often the biggest share of the time.

Get a demo Read the method

Methods and rules

Single-order picking Batch picking Zone picking Wave picking S-shape routing Return routing Velocity slotting Cube-per-order index

The problem

For most pickers, the walk between picks is the biggest part of the day.

Pickers spend the shift walking

The pick list is in the order the orders arrived, not the order the shelves are in, so the same aisle gets walked again and again. Order-picking research has long found travel to be often the largest share of a picker's time, and no standard report shows aisle-metres walked per order.

The best sellers live at the back

Locations were assigned when the products arrived, and nothing has moved since. Last year's slow line sits at waist height by the pack bench while this season's fastest mover is on the top level of the far aisle, reached by ladder, many times a day.

We changed how we pick and nothing improved

Someone read about wave picking and switched. The waves were sized for a peak day, so carts overflowed on busy days and pickers waited on quiet ones, and the team drifted back to the old list. A pick method is a fit with your order profile, not an upgrade you install.

The method

Choose the method, slot the stock, sequence the walk.

Choose the method from the order profile

Picking one order at a time suits orders that are fairly large. Batching groups orders so one trip serves several of them. Zoning gives each picker one area and the part of every order that lives there. Waves release orders for a common destination together across areas. Waves are usually picked in batches, and many floors run more than one method by order type and time of day.

Put stock where the picks are

Start from the cube-per-order index (Heskett, 1963): divide the space an item needs by the number of trips it generates, and put the lowest ratios nearest the point where pick tours start and end. Then adjust for the body: heavy, fast lines near knuckle height, which is where the NIOSH lifting equation sets its vertical multiplier at 1.0, not on the top level or the floor.

Sequence the walk, then check the record

Sort each pick list into walking order through the aisles. Routing is a solved problem for standard aisle layouts; what remains is putting the frequent picks close to the start and end of the walk. Confirm every pick by scan at the location, because a slot is only as good as the record that says what is in it.

The shape of it

The pick task is only as good as the location record.

BinLogic is built mobile-first: floor work such as receiving, putaway, picking and cycle counts runs on a handheld device.

On any floor, a pick confirmed by scan at the location is the evidence that the slot holds what the record says, and re-slotting only pays when every move is recorded the same way.

Ways to release the work

Single-order picking Batch picking Zone picking Wave picking + whatever mix the order profile needs

Sorted by location · Walked once

Pick list

In walking order

Confirmed by scan

Location record

What the slot holds

Questions

What teams ask before they change how they pick.

Which pick method is right for us: batch, zone or wave?

It depends on how many lines an order has, how many orders arrive per hour and how big the building is. In the research literature, batching groups orders "into a number of sub-sets, each of which can then be retrieved by a single picking tour"; zoning assigns each picker "the part of the order that is in his assigned zone"; and wave picking releases orders "for a common destination" simultaneously across areas. Our rule of thumb: many small orders favour batching, many SKUs or specialised storage favour zones, and fixed carrier collection times favour waves. In the literature, waves are usually picked in batches.

How much does picking really cost?

More than most teams assume. De Koster, Le-Duc and Roodbergen's 2007 review of order-picking research reports that order picking is estimated to cost "as much as 55% of the total warehouse operating expense", and reproduces a typical breakdown of a picker's time (Tompkins et al., 2003) in which travel is the largest share; the review notes that travel is often the dominant component. That is why slotting and sequencing matter: they attack the time that adds no value, rather than asking people to move faster.

What is the best way to plan the shortest pick route?

For the common layout, parallel aisles with a cross aisle at the front and back and none in the middle, it is a solved problem: Ratliff and Rosenthal showed in 1983 that the shortest route can be found exactly, with computing effort "linear in the number of aisles". Simple rules such as S-shape, where any aisle with a pick is walked end to end, or return, where the picker enters and leaves each aisle from the same end, are easy for pickers to follow, at a cost in distance: de Koster and colleagues' 2007 review reports S-shape routes averaging between 7% and 33% longer than optimal. Routing and slotting work together; neither replaces the other.

Where should heavy, fast-moving items go?

Near knuckle height, about 75 cm (30 in) from the floor for an average-height worker, and in practice near the front of the shelf. NIOSH's revised lifting equation starts from a load constant of 23 kg (51 lb) and reduces it as the hands move away from the ideal; the vertical multiplier is 1.0 when the load is at 30 inches (75 cm), and the horizontal multiplier is 1.0 when the hands are within 10 inches (25 cm) of the mid-point between the ankles. The equation assesses two-handed lifts; it is a guide to the lifts that strain people least, not a slotting rule. OSHA's 29 CFR 1910.176(b) adds: "Storage of material shall not create a hazard," and tiered stock must be stable.

Less walking per order, not faster walking.