Logistics News: Optimizing the Number of Sortation System Diverts and Pallet Positions

The specific question was around an idea one reader at a wholesale distributor had for potentially improving distribution productivity. That led us at Distribution Digest to also ponder the broader question of how a company should consider the number of pallet positions to use at the bottom of each divert in an automated sortation system.

I wish there was a simple formula, but the reality is that there is no right answer, in a sense, and coming up with the best answer for each company involves a whole lot of variables. In my career, I have seen as few as 1 pallet position at each divert to as many as 10. This particular reader was pondering an increase to as many as 20 pallet positions per divert.

At a high level, of course, the first question is one of volumes. In the end, the system needs to be capable of supporting a given number of pallets per day. That can be achieved by having more diverts and fewer pallet positions per divert, or fewer diverts and more positions at the end of each one.

The productivity in pallet building generally declines as the number of pallet positions increases. It takes more time to determine which pallet the carton goes on (even when supported by auto ID technologies), and beyond just a couple of positions adding more will increase travel time walking to each pallet.

However, each divert adds to initial and on-going system cost, so companies need to find the right balance between initial system investment and on-going distribution center efficiency.

In most cases, of course, the analysis is usually constrained as well by physical or budget realities. An existing or even new building will have limits as to the length of a sorter. Available budget may also constrain the theoretic optimal length of the sorter and number of diverts.

All that said the right number of diverts is also based on such factors:

• Number of orders per batch
• Number of batches per day
• Number of pallets per order
• Number of cases per order

Another factor for some DCs occurs when orders are batched and picked by carrier.  This was the case for this particular reader and adds an additional level of granularity to the equation.

First - the cost of the physical sorter is a onetime capital investment, whereas pallet build labor is incremental and on-going.

Second - it is logical that for every unique DC operation there is a point on a curve where, based on what we said above, additional diverts do not yield higher pallet build productivity.  That would be the theoretical optimum number of diverts.

That being said, based on my experience spread over hundreds of sorting systems, I would be inclined to keep the number of pallet build locations per divert in a range of 1 to 6 with 3 being the design target.

Relative to the number of people working on each divert, I would go with no more than 1 per pair of side-by-side diverts.  More people can be assigned temporally if/when the lines become backed up due to a surge in volume.

As mentioned in the original response, I would again caution about the inherent danger of having forklift traffic in and around where people are working.

If you have some thoughts on this we’d love to hear them.

Agree or disgree with Holste's perspective? What would you add? Let us know your thoughts for publication in the SCDigest newsletter Feedback section, and on the website. Upon request, comments will be posted with the respondent's name or company withheld.