Supply Chain Digest: The Two Paths for DC Automation

By Dan Gilmore, Editor-in-Chief, Supply Chain Digest

Through a special arrangement, what follows is an excerpt of a current article from Supply Chain Digest, presented here for discussion.

Very few companies have yet taken the highly automated path, but I believe the interest in doing so is growing.

Here is a provocative comment: distribution operations in the U.S. will morph into two extremes when it comes to automation. One group will invest in highly automated systems to get rid of the costs and headaches of DC labor; another group will, in large measure, eschew automation altogether, and use more manual systems for their flexibility and in some cases better results. The mid-level approaches – moderate levels of automation – will see their “market share” decline.

Distribution centers in Europe are known to have, in general, much higher levels of automation than those in North America. Why? In general, higher labor and land costs make automation easier to justify. You need to be able to get more throughput out of less space, and each employee adds a lot of cost and is simply a lot harder to downsize or fire if needed in many European countries. In contrast, we don’t yet really have the high cost of land here in the U.S.

In talking with logistics executives, however, I find an increasing number that would love to automate their perceived labor costs and headaches away.

“We have never been big on automation in the past,” a director of distribution for a well-known CPG company told me this spring. “Now, we are looking first to automate everything until we show we just can’t justify it. It’s a real change of mindset.” The catalyst for this change in attitude – labor headaches and turnover. Despite paying above average wages, DC employee turnover for the company has been running at 50 percent.

A few years before, the VP of supply chain at a major brewer shared his vision of a “lights out” DC operation with me. Laser-guided vehicles would perform almost all of the movements, with very few employees required.

The lack until now, however, of any highly automated alternatives for case picking has been a big barrier to the “lights out” vision. As these technologies mature, I believe they will be a real driver of increased DC automation.

Let’s now swing back the other way. I also hear lots of folks tell me they’ve had enough of automation. Factors often cited: big capital expense, inflexibility down the road as things inevitably change, and too often less than stellar results. One manager recently told me, in all its new DCs, his company was moving away from automation after finding voice picking on pallet jacks delivered superior results over more automated approaches.

Very few, I’ll admit, have yet taken the first path toward highly automated DCs, but I believe the interest in doing so is growing.

Discussion Questions: Do you also see a trend of companies moving towards a highly automated distribution vision, or one that has relatively little automation? Will the middle approach get squeezed? Is the “lights out” vision a potential reality – or just mostly wishful thinking? What, if anything, can we learn from the Europeans?

BrainTrust

Discussion Questions

Poll

7 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Mark Lilien
Mark Lilien
16 years ago

Physical distribution executives have been debating increased automation paybacks for decades. Many distribution centers haven’t tried sustained significant productivity incentives for their work forces, however. When productivity incentive programs are done skillfully, the paybacks often beat the risk and expense of major automation. Another alternative: sooner or later, more physical distribution will be done from Mexico, cutting labor costs considerably.

Nikki Baird
Nikki Baird
16 years ago

I have heard more lately from eCommerce distribution execs, so my perspective here is skewed: they tell a story of building distribution capacity as fast as they possibly could by throwing as many people at it as they possibly could, during the big rush to eCommerce. But now they’re at capacity in their existing facilities, or are missing fill rates because they don’t have real-time visibility into their inventory position, and they’re faced with either investing in automation to increase efficiency and raise the bar on capacity in their existing sites, or run out and buy/build a new DC. Automation is winning out in the short term.

But it’s fascinating to me that the messages these guys key in on are the basic value propositions for basic automation. There may be some threshold where your volumes are too small to justify automation at any level, but for the most part it’s surprising to me that this is news–fancy laser-guided robots aside, the business case for more automation in a DC is pretty much a slam dunk. The only question is whether you implement it with enough discipline to actually capture all of that business case.

Ed Dennis
Ed Dennis
16 years ago

UPS and FedEx have already solved all these problems. The grocery industry will, however, have to hire 50 consultants and enlist 12 competing software providers to arrive at a hodgepodge of systems that will be obsolete in 3 to 4 years. Spend a smart nickel and call FedEx or UPS and offer to license their systems for your warehouse. This could actually reduce your operating cost. Oh, and if you want to get a little more complicated by adding some direct delivery vendors then talk to Amazon.com. None of this is new–you just don’t know about it!

Bill Robinson
Bill Robinson
16 years ago

In the last 20 years Automation in the DC has had to take two enormous mid-course changes in direction. The first came in the 80s and 90s when it became apparent that vendors would actually collaborate if you asked them. This meant that mass shipments packed at the convenience of the supplier could be packed in a convenient way for retailers, marked, and ready for sale. This changed the mass holding and sorting areas into flow-through areas.

The second came in the late 90s when retail distribution centers suddenly had to begin to ship directly to their customers based on web orders. This required a manual intervention in the automation processes designed to send cartons efficiently to stores.

Based on this history lesson, retailers would be smart to design their distribution centers with the utmost flexibility.

Bill Bittner
Bill Bittner
16 years ago

The distribution center processes can be categorized into four major areas: Receiving and Putaway; Letdowns to picking areas; Outbound Order Selection (picking); and Outbound Order Assembly. The implementation of automation in each of these areas has varying levels of complexity and payback.

Receiving and putaway is probably the easiest area to implement automation as palletized receipts of merchandise are conveyed to their storage areas. Although it doesn’t use many labor hours, these are usually the most skilled (i.e. the forklift drivers) workers. All this assumes the merchandise arrives palletized or is palletized on the receiving dock by the shipper. Just thinking about this, do the recent investments in railroads by Warren Buffet mean all those railroad tracks that have been covered with concrete going to need digging out? Different topic.

Letdowns are tough because of the need to refill pick areas in anticipation of the shipments. The quantities are not always even pallet loads. This means in an automated approach there must be room to stage the letdown so that space in the picking area is available.

Picking is really difficult to automate. Here you get into handling the different case dimensions, weights, etc. Each product/manufacturer can have their own case configurations. The picking “mechanism” must be flexible enough to respond to variations in weight, fragility, dimensions, etc. Quite a challenge and an expensive area to automate. Unfortunately, it can also be the source of the most headaches as absentee rates and turnover are constant challenges.

The last area I identified is order assembly. I separated this from picking because this is where I think there is a great opportunity for automation. This is using the “Pick to Belt” approach. Instead of making the order assembly dependent on the sequence of the pick slots, “Pick to Belt” allows the outbound order to be sequenced for the store. This is a huge advantage for the store and should be a top priority for any self distributing retailer and independent retailers considering a wholesaler. I have always been disappointed with voice pick applications because they seem to only look at the warehouse as a cost center instead of a service center. By using picking labels, putting the cases on a conveyor belt, and sorting the cases for store delivery at the shipping dock, the warehouse can save significant amounts of store labor costs. The outbound loads can be assembled by store layout and priority in conjunction with a computer assisted ordering application that manages both selling area and backroom inventories.

So this is kind of a long way to reaching my opinion, which says I disagree that the middle of the road approach to automation will be squeezed out. “Pick to Belt” is a middle of the road approach that makes more sense than ever. Especially as retailers consider more banners with specific demographic targets and heterogeneous store layouts.

Kai Clarke
Kai Clarke
16 years ago

I disagree with the presumption of the article that Europe has better distribution systems than the US. This is clearly not true, and part of it goes to the disjointed ability of companies to cover all of the different countries in Europe. This lack of a consolidated EU means that companies don’t have to have state of the art automation, logistics, and international retail efforts. This allows for lost efficiencies and inefficient status quo management from both an electronic and manual perspective.

Compare this to the USA where Internet retailing requires not only sophisticated logistical support, but the ability to communicate between suppliers, customers and all facets of the logistics cycle within a company. Add to this a tremendous demand from the consumer (in dollars) that is literally thousands of times larger than a European effort, and it is no wonder that US corporations demand a level of excellence that is unparalleled in the rest of the world. We must (and do) reflect the state of technology in each step of the logistics, and retailing cycle to stay competitive. Anything less than this and the consumer rapidly shops elsewhere.

M. Jericho Banks PhD
M. Jericho Banks PhD
16 years ago

Automated picking and order bundling is just over the horizon in a revolutionary system invented by an American and being tested currently. It can accomplish the same tasks as a conventional DC in a footprint about two-thirds smaller and construction costs and times less than half as much. Additionally, this system–with its one-third sized footprint–is about twice as productive as current DCs. In brief terms, it does the same job and more of it, in a smaller facility that costs significantly less to build in a shorter period of time. Whew! I am constricted by an NDA, but can put the inventor in touch with you if your interests correspond with his. Let me know.