The popular batch-order picking and sorting system was first deployed in DCs in the mid 1970s. While they have evolved technologically over the past three and half decades the basic model is pretty much unchanged. However, the DC environment for which they were originally intended has changed considerably.

Today’s distribution centers are providing an ever increasing array of individualized customer specific services that require special “off-line” handling, assembly, packaging, customization, and store-ready shipments that disrupt the continuity of orders flowing from picking to shipping. While these services are generally recognized to be beneficial downstream of the DC, they increase operational complexity and labor cost at the DC.

Much of today’s VAS requirements were unheard of in the mid 70’s when the concept of batch-order picking and sorting was conceived. Back then the order fulfillment process was much more simplistic and predictable when compared to today’s multi-functional, demand driven DC operation.

So, it’s not surprising that many of the logistics executives, who provided insight and opinion for the upcoming SCD Automated Case Picking (ACP) 2011 report, tell us that they prefer discrete order picking over mechanized batch-order picking and sorting. Interestingly, these same executives are expressing high to very high interest in ACP. The key driver appears to be the need for more flexible and intelligent material handling solutions.

In a nutshell - here’s what they are saying:

There is an explosion of VAS in the DC. Once or twice a week they have to react, often with little or no prior notice, to new never seen before requirements or specifications. These highly tailored customized services usually require setting up a separate standalone manual processing area. This adds a whole new level of operational complexity and cost. Failure to comply can result in an entire shipment being rejected and/or substantial charge backs.

Many claim that the batch order picking and sorting model does not or cannot, provide the operationally flexibility and agility they now require, making it even more difficult to provide VAS at a competitive and quick response level. They worry that control of the individual customer order could be compromised creating errors, service delays, and more charge backs.

The most egregious example, and the one most often cited, relates to the use and placement of bar code labels on cases. Here the growing trend is away from uniform standardization of label design and location on the case (known in the industry as the UCC-128 shipping compliance label specification). An even more troubling emerging trend among some major retailers is the elimination of all bar code labels (including the I2 of 5 UPC SKU identifier) from the case. These bar coded labels are what have been driving mechanized picking and shipping systems for the last 30 years. They are of critical importance to enable system operations - see “Worrisome Trend in Retail Label Requirements”.

The Batch Order Picking Model – Historical Perspective

A little known fact is that the motivation behind the concept of batch-order picking was originally quite simple - to sell more case sorting equipment.

Going back to the mid 70’s, one of my responsibilities as Manager of Sales & Engineering for ACCO IHS Div. (located in Frederick, MD, but no longer in business), was to expand the market for ACCO’s proprietary sorting equipment.

ACCO had a line of high end tilt-tray and tilting slat sorters sold primarily to large chain retailers like Sears, Montgomery Ward, and JC Penney who were deploying high volume mechanized handling systems in their regional order fulfillment centers. These systems were controlled with relay logic, early stage programmable logic controllers (PLC), and a room full of computers. Needless to say the market for these high-end sorters was limited.

ACCO also had a line of pushers that were capable of diverting cases and totes off of a belt conveyor at the rate of 40 to 60 per minute. To control the activation of the pushers in a sorting application, ACCO’s engineers developed an inexpensive solid state tracking memory. An operator, located at the beginning of the sort conveyor (referred to as the induction station) would read a two-digit sort code on the case and enter it into the tracking memory using a standard twelve key numerical keypad consol.

The first such system was called Sortrac and it consisted of up to (31) pneumatic pushers, designed to be mounted on the side of a belt conveyor controlled by the tracking memory. The first Sortrac installation was on the shipping dock at a Lionel Corp. (the toy manufacturer) warehouse where workers would manually push “sort” cases off of a live roller conveyor onto gravity lines flowing directly into trailers. The live roller was converted to a belt conveyor. Sortrac pushers, along with the solid state tracking control, were installed over a weekend. As a result one person at the keypad consol replaced several manual sorters. ROI was about 3 months.

At that time manual discrete order picking was the normal and most logical method of picking and filling orders in the DC. Typically, workers tour the warehouse (usually on foot) several times a day picking cases and items for one large order, or perhaps 2 to 3 smaller orders. The cases and/or totes were stacked on push carts, which when full were brought to a pre-assigned area for consolidation onto shipping pallets.

It was in this environment that ACCO’s engineers saw an opportunity for a low cost, pre-engineered, mechanized order consolidation sorter. By having the pickers (assigned to a specific conveyorized picking zone) batch pick SKUs for several orders at a time, ACCO could sell its standard Sortrac sorting system to sort and consolidate the SKUs into orders for palletizing or direct trailer loading. The cost of the system would be justified by reducing picking labor and increasing throughput capacity while improving order accuracy.

It may be interesting to note that the Feb. 23rd edition of “Sorting it Out” “Can a Company’s Product Mix Stymie Adoption of Automation” describes a hard goods company that installed just such a system in the mid 70s. It was surprising to hear that the system is still operating daily, however, limited to processing small volumes of seasonal SKUs.

In 1976 the material handling industries first batch order picking system with automatic induction, laser bar code scanning and sorting was installed at Kobacker Shoes in Columbus, OH. The sorting belt conveyor was equipped with (30) Sortrac pushers that were specially designed to divert shoe pair boxes into gravity skate wheel lanes at the rate of 60 cases per minute, where they were packed into shipping cases. In addition to being the shoe industries first batch-order picking/sorting system, it was the first installation of a bar code scanner in a distribution center application. The system was in service for about 15 years before being upgraded and expanded.

The Kobacker batch-order picking and automated sorting system, was featured in all of the material handling trade publications at the time, and gave legitimacy to the batch order picking and sorting model.

Throughout the years, the question that skeptical DC executives frequently asked was – ‘Why take a bunch of individual customer orders, summarize them by SKU, have multi pickers pick all the SKUs at one time (batch pick), then deploy sorting equipment to put them back together in individual orders again?’ The simple answer has been that the higher productivity, throughput, and operational efficiency gained from batch order picking trumps hands-on operational control of picking orders individually.

This answer proved to be sufficient to get many batch order picking systems approved over the years. However, as stated above, the proliferation of VAS poses new challenges for that model. Going forward, an alternative is to reconsider discrete order picking based on a new generation of automation technologies.

What’s Old is New Again

Recently, in some trade publication, you may have seen the headline “The Robots are Coming!” While this is certainly true, in fact what is really coming is the rebirth of discrete order picking methods enhanced by automation technologies. Although, this time it will be supported by highly flexible, adaptable, and agile automated solutions based on discrete order picking processes.

Discrete order picking has always been the most logical way to manage the picking process. Batch order picking simply cannot compete with the flexibility, adaptability, and agility inherent in the discrete order picking model.

The axiom that - the more automated a process is, the less flexible it is - has been debunked by intelligent control systems that can look ahead and make adjustments in real-time. That is exactly what drives mixed case robotic palletizing systems and is made possible by relatively cheap computing power with blazingly fast processing speed.

Final Thoughts

I’m pretty sure that the batch-order picking model will continue to serve some distributors who are heavily invested in it. But now more than ever, skeptical DC executives should ask – “What are the alternatives?” This question is especially appropriate in light of the multi-functional role now being played in the typical DC.

The upcoming Supply Chain Digest 2011 report on ACP technologies, will provide real world examples of who, what, where, and how alternative solutions are been deployed and justified.

Agree or disagree 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.

You can also contact Holste directly to discuss your material handling or distribution challenges at the Feedback button below.