The Infosys global supply chain management blog enables leaner supply chains through process and IT related interventions. Discuss the latest trends and solutions across the supply chain management landscape.

« Postponement strategy: what are the key drivers and design elements | Main | PAS 55: The New Standard for Asset Management »

Virtual Locations in a Warehouse

Have you ever thought of a location in a warehouse which has a physical pallet, but actually the location does not exist? Strange, isn't it? Well, let's dig into this.

A  few years back I had been to a country in the Far East to carry out business process study for multiple warehouses for a marine food processing plant. Here the raw material was mainly fish which were of all kinds of varieties; these would arrive in trucks and unload the lot in a sorting area where they would be segregated based on the weight and size. They were then placed in large steel containers and then putaway into a large cold storage at sub zero temperatures around under -40 degree Celsius.

The cold storage had a peculiar way of storing these steel containers. The cold storage did not have a racking system, any aisles or bays; it was just a large room. Location numbers were on the floor and the same was painted exactly opposite to it on the ceiling above. This was done so that once a container was placed on the floor, its position in the cold storage would be not visible, therefore the location number on the ceiling was used as an alternative to identify which position the pallet was placed. These steel containers were placed in rows with upto 7 pallets height, placed one over the other, forming a 'bay'. In the same way multiple bays were created for additional pallets, thus forming an 'Aisle' between these bays.

A set up of this kind would be a challenge for any WMS to handle such a composition of containers placed in such a manner. One would generally treat this as a bulk location, but again, stock was placed based on FEFO, fast moving items etc, which meant containers having fish expiring early had to be placed at the 7th stack (on top of the 6th container) and the ones expiring next on the 5th level and so on.

A similar situation prevailed for a ceramic manufacturer in the Middle East. In this case it was about finished goods placed in a massive yard, where ceramic goods were placed in pallets, shrink wrapped and then placed one top of the other upto a maximum of 3 levels since these were ceramics and each pallet could hold the weight of 2 pallets on top of it.

The challenges in these two scenarios are as follows:

1.Maintain stack sequence for each location. Most WMS packages do not maintain this aspect. The location code along with the stack sequence will identify each position. The system will generate a stack number each time it putaway a pallet to the location and stamp it to the pallet number to locate its location. Say for Location L102, when the first pallet is putaway, the location will be L102/S0, meaning the pallet is placed in location L102 and ground level S0, when the next pallet is placed on top of this, the location will be L102/S1 which means location L102 and stack sequence S2, i.e. its placed above L102/S0 and so on.

Most WMS systems in this case have to be customized to store max stack heights for each location upto which it can putaway pallets and also identify at which stack height locations have reached, before carrying out further putaway to that location. This solution would be required in case there is a need for identifying the exact location of a particular pallet in the warehouse.

2.Since there is no racking system, it is not possible to provide any location address for each location. It may be as a single location but with a maximum capacity of n pallets that can be placed one over the other.

3.LPN may help to identify a pallet, but again, any WMS would not be able instruct which pallet will be on top of which one, since stack height sequence may or may not be maintained in any WMS, unless it is considered as a customization.

4.Pallets with different SKUs placed in the same location would reduce efficiency, i.e. if the pallet to be picked is right below three other pallets having different SKUs, it would lead to removing the top three ones, then take the required pallet and placing the other 3 back on the same position, which would consume time and also needs additional space to carry out such a task. So it would be advisable to have the same SKU pallets to be placed in the same location.

5.FEFO will work only if pallets of the same expiry date are stacked up in a single location. This would not be applicable for the ceramics company, but for the cold storage, definitely yes. Other way round would be to a have a logic to putaway pallets expiring earlier on the top and the ones much later in the bottom of the stack.

These are some of my thoughts around 'Virtual locations' in a warehouse, which exists only due to the presence of a pallet and non-existent when the pallet gets picked. I presume there would be other ways in which these requirements have been met, but this is the closest I have come in resolving them!


Nicely put up areticle which gives a great insight into the unfamiliar challenges that a consultant can face in these food processing industries.

As you have said customizations to the WMS package may be unavoidable in these areas of implementation.

Why introduce a separate stack # field? The putaway timestamps of the pallets relative to each other in a particular location should suffice.

Also, FIFO can be practised here only if putaway is restricted with a rule like: Do not allow putaway into a location where there is already a pallet that has expiry date earlier than that of the pallet you are about to put.

If pallets of different items are kept in the same location (in the marine food example)...well...that could get tricky. The system can still have a convoluted way of handling it. But, I suspect too many human errors in picking would be made.

A time stamp would be a good idea, Soumya but this cannot be printed on a pick list. The time stamp can be internally used to sequence the stack fields, which can be used in the pick list. FIFO as a condition to putaway is nice to have but again it would be heavy on system resources when this check is done. You are right; it will be quite tricky to if multiple SKUs are kept in the same location, it would make picking and putaway rather more complicated. Thanks for writing into my blog.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter