In this section, we present some of the basic concepts for improving
order picking systems:
Usually the items in a warehouse exhibit varying characteristics
with respect to dimensions, weight, demand, and other properties.
It is natural to apply certain storage and retrieval strategies
depending upon the product families or individual products within
families.
A basic rule in assigning products to storage locations is storing
"better" products in the "better" locations
in the order picking system.
A "better location" is a location which provides faster
and more ergonomic access to the product stored. For example,
in a pallet flow rack, the pick-face that corresponds to
the direct front of the order-picker (also called the golden-zone)
is a better position than a pick-face which the order-picker has
to bend down to reach. (figure)
A measure of "goodness" of an item could have been the
frequency that it is requested. If an item is requested frequently,
it is logical to keep that item in an easily-accessible location.
But if the item is too heavy, it may be too much time consuming
to replenish that item to that favored location.
Another measure of "goodness" for an item is occupying
smaller space. On the other hand if an item is requested very
infrequently, it is not necessary at all to assign it to a favored
position, just because it occupies little space. If that practice
were followed, the "best" locations could be filled
with lots of small products that are not really requested much.
Another basic rule in assigning products to storage locations is taking into consideration the dimensions. Cube matching of the items with the storage locations is essential to eliminate space inefficiencies. Shelf dimensions should be spacious enough to allow easy picking, but tight enough to avoid unused space. Here is a bad usage of shelf space versus good usage:
Orders contain one or more items. It is important to know about
the structure of the orders (such as number of items per order,
correlation among items in an order), since it affects storage/retrieval
strategies and design of order picking systems. Simple plot of
percentage of orders vs. number of items in an order would give
considerable insights.
A very common strategy is batch-picking of small orders, that is picking several orders in one travel. This can bring significant reductions in travel time, which is the most significant part of total time spent in order picking. Here is a simple example that illustrates the concept (red points are locations of single-item orders, and blue ellipse represents the picker):
A detailed discussion of order picking principles is available (Frazelle, 1988).
With thousands of items being stored and retrieved in a system,
it is crucial to maintain control and track the movement of items.
Barcodes are an important technology for identification of items
when they are received, stored, retrieved, assembled into orders,
and shipped.
