March 18th, 2022 | 3PL, eCommerce, Order processing

Effective Picking

Effective Picking

The basic requirement for all fulfilment operations is accuracy and efficiency, nowhere is this more important than in order picking. Fast and accurate picking is made possible by two distinct processes working effectively together. The first is the slotting logic that determines how stock is distributed in the picking face, the second is the picking strategy that determines how orders are picked.

Product slotting is covered in a related blog post here, the rest of this article describes a few of the different approaches that can be used to ensure your picking is fast, accurate and cost effective. It assumes that you’ve already optimised your pick face slotting to distribute stock across the pick face in the way that works best for you.

Think about your ‘order pool’

The starting point when thinking about how to organise your picking is to think about the sort of representative ‘order pool’ of pickable orders that you might be faced with at the start of a busy day. Unless your orders are remarkably uniform it’s unlikely that a single picking strategy will work across the whole order pool. There will usually be a mix of single-line and multi-line orders, some unusually large and perhaps a significant proportion are for a particular SKU or combination of SKUs. The order pool could include orders for SKUs that are especially heavy, large or fragile. Some might need to be despatched using a carrier that will collect at three that afternoon, others that will go at seven in the evening.

These are all characteristics that relate to an order. The other side of the coin is the relative locations of the items that need to be picked. Your order processing system must be able to assess all the orders in the pick pool and the locations of the items that need to be picked and assign each order to the optimal picking process.

The optimal groupings that make sense in your operation will be specific to your own environment, below are some typical examples:

Multiline orders:
Multiline orders might be picked into a subdivided carts, with the pickers being prompted to scan the items into the appropriate subdivisions so that items are already sorted into customer orders when the cart is presented to the packers.

Single SKU orders:
Barcoded SKUs required for single item and single line orders can be picked in location sequence, guiding the picker through the picking face in a simple route that avoids them having to double back on themselves. The required items can be picked into a mixed tote.

Bulk orders:
Trade orders or large consumer orders can picked directly to pallets that are identified with a temporary licence plate, if required multiple pickers can work on large orders simultaneously, stopping and restarting the pick as required. Large orders might be picked directly from bulk locations rather than from the picking face to avoid triggering an immediate replenishment.

Priority orders:
Orders that need to be prioritised in order to meet their target despatch time or need to be pushed through a more complex despatch process might be picked individually.

The picking process will typically require the picker to scan each item into a cart, tote or other mobile location that’s uniquely identified with a barcode. These are then queued to await packing.

As complicated as necessary, as simple as possible

There is a tension between trying to ensure each order is picked as effectively as possible and subdividing the order pool so much that it becomes unnecessarily complex. However in some environments it can make sense to refine the groupings further. Below are more focussed groupings that might significantly reduce the distanced walked by your pickers or improve the efficiency of the packing process:

Multiline orders in specific aisles:
Multiline orders might be grouped together so that they can all be picked from a single aisle. This approach can be very powerful if used in combination with slotting logic that ensures your most popular SKUs are present in each picking aisle, significantly increasing the proportion of multiline orders that can be picked within a single aisle.

Orders that are all for the same SKU or combination of SKUs:
If your order pool often contains a significant number of orders for the same SKUs or combination of SKUs it can be far more efficient to identify these patterns, group the orders together and complete a single pick for the items required. pick a single SThe picking strategy can pick from bulk?

Cross docked orders:
In environments where stock is frequently received after the customer order has been received OrderFlow can be configured to support the cross-docking of incoming stock. This allows stock received in incoming deliveries to be matched with pending orders and immediately packed. Incoming stock items for which there is not an immediate requirement are put away into warehouse locations in the usual way.

Cross docking of orders can significantly improve the efficiency of the warehouse by reducing the proportion of items that need to be handled twice.

An ongoing process of refinement

Just like slotting, finding the picking strategies that are right for your operation is an ongoing process. In order to be able to monitor and tweak it on a regular basis your WMS must be able to capture the key metrics that you want to use to measure the efficiency of your picking. These are likely to include:

  • average time taken to pick an order
  • average distance between pick locations
  • % of orders that miss their target despatch time
  • number of orders picked by each picker
  • number of items picked by each picker

These figures will need to be tracked over time and analysed in combination with the metrics that measure your slotting efficiency, which is the subject of the blogpost here.