Inside Logistics

Patents pending

Are Amazon’s alternative DC designs sci-fi or hi-fi?


October 16, 2017
by Treena Hein

Nifty or nutty, innovative or idiotic, mind-blowing or mind-boggling? When news hit this summer about Amazon’s patents for a multi-level tower DC and airship DC—let alone a distribution centre located underwater—it was hard to tell whether the company was simply seeking attention, methodically covering its bases or if it really plans to turn these concepts into reality.

Jeff Bezos

Before we look further at the patents, let’s first catch up on Amazon’s evolution. As many know, founder Jeff Bezos launched the company as an online book store in 1994. He has since transformed it into the biggest online store on the planet, with an ever-growing supply chain and DC network able to deliver a rapidly expanding number of products across the world efficiently and economically. It is reported that Amazon grossed over US$135 billion last year and employs more than 350,000 people.

As it transforms, Amazon has been quietly expanding its focus on technology and logistics and also venturing into cloud computing, artificial intelligence, entertainment and more. It launched Amazon Robotics back in 2003, and its acquisition of Kiva Robotics in 2012 lit a fire under the robotics industry. In recognition of these innovations, Bezos is being inducted into the Logistics Hall of Fame this year.

Amazon has also been experimenting with drone delivery for several years. US-based tech research firm CB Insights reports that Amazon filed over 1,100 patents in each of 2013 and 2014, 660 in 2015 and 310 in 2016. And in 2016, 88 of its patents contained key words related to logistics, up from a mere seven in 2009.

The underwater DC relies on an acoustic signature to active balloons to lift product to the surface.

It’s clear that at least one of Bezo’s eventual goals is to expand his company’s global supply chain so that Amazon becomes the world’s primary interface between manufacturers of goods and the people who want to buy them. Amazon’s specific plans on that front are too numerous to summarize here, but it’s actively building its own delivery network rather than relying on existing entities like Purolator—for example, investing in ocean-based freight forwarding and building a US$1.5 billion air freight centre in Kentucky from which it operates dozens of leased cargo jets.

Along with expanding delivery, Amazon has also had to grow its distribution centres—which it calls fulfillment centres—in size and number. In its Amazon “strategy teardown” CB Insights notes “early on, Amazon realized its shipments were all unique combinations of goods, and that its warehouse techniques would actually be closer to manufacturing than shipping.”

Amazon has installed tens of thousands of robots in its DCs, and its total global number of DCs is exploding. In the US last year Amazon had a warehouse or delivery station located within 32 kilometres of 44 percent of the population, compared to only 26 percent two years before. This is expected to reach 80 percent in the next few years.

These fulfillment centres must be efficient, but whether Amazon’s futuristic DC concepts would actually be efficient—let alone feasible—is another matter entirely.

Warehousing, submerged and skyward

In a way, Amazon’s under water DC concept is no different from any other DC in that it involves stock placement, storage and removal. In this case however, stock is dropped into a natural or engineered body of water by parachute, truck or conveyor belt. Non-waterproof goods (the vast majority) would first be placed in weighted watertight containers, each programmed with a unique acoustic signature. When a given container is sonically ‘called’ by the DC system, it triggers an air canister that inflates a balloon, causing the container to move to the surface for removal. Artificial currents might be employed to move goods in desired directions.

Amazon argues in the patent that this concept addresses a serious problem: “For all of their technological advancements, today’s fulfillment centres are still plagued by the inefficient use of space.”

These centres, states the patent filing, “feature storage areas as large as one million square feet…Therefore, in order to prepare and ship an order that includes a large number or different types of items to a customer, a staff member or robot may be required to walk several thousand feet or even miles within a fulfilment centre in order to retrieve the items.”

The fallacy of unused space

Charles Fallon

Charles Fallon, principal at consulting firm LIDD in Montreal Quebec, isn’t buying this premise. “It isn’t correct to say that unused space is inefficient,” he says. “Unused space is fundamental to the distribution centre—you cannot distribute without the ability to move product around, to disassemble vendor deliveries and re-assemble that product into your customers’ needs. Calling that space ‘inefficient’ is misunderstanding the function of a DC.”

Fallon notes that most mega-DCs are typically several DCs under one roof—for example a grocery retailer having interconnected freezer, cooler and dry grocery DCs on one site.

“It is almost never the case that humans would be walking thousands of feet to complete a transaction within such a DC,” Fallon says. “If Amazon suffers from having humans walking thousands of feet…then you can conclude that Amazon’s DCs are probably too big for what they attempt to do and Amazon has the wrong materials handling systems.”

Author of its own problem

He believes Amazon’s problem is that in its “enormously expensive purchase of Kiva technology, it has committed to a solution that exacerbates space utilization problems rather than helps.” Amazon’s investment in Kiva, he explains, only solves part of their distribution problem while creating new problems, such as how to use clear height in a DC and how to make the shelving units readily accessible while maintaining the needed throughput.

“Wal-Mart mega-DCs do not have the same space utilization and handling challenges that Amazon does,” Fallon adds. “Amazon likely had to move so fast in creating its DC network that it couldn’t correct for fundamental design flaws.”

He believes there are basic solutions to this problem “that don’t require far-out thinking.”

Zoe Leavitt

But tech industry analyst at CB Insights Zoe Leavitt points out that since the Kiva robots travel underneath shelves and don’t require as much aisle space as humans, they have enabled Amazon to fit more inventory in a given warehouse, and that “even its underwater warehouse patent is focused on saving space.”

She adds that “if air bladders could push the packages around independently underwater, there would be no need for any aisles for either humans or robots, and Amazon could fit more packages into the same amount of space. Also, by storing goods underwater, Amazon could reduce their real estate footprint by going deep rather than wide.”

Under water may be all wet

Dana Magliola

But saved space or not, Fallon and Dana Magliola, owner of consulting firm Supply Chain Forward in North Carolina, are not at all convinced of the merits of an underwater DC. Fallon dismisses the concept as a fun idea that makes for great marketing with neither financial or environmental justification behind it, and “a non-solution to a non-problem”. For his part, Magliola calls the concept “an environmental disaster waiting to happen,” and “pretty far out there, even for Amazon”.

Amazon has also filed a patent for a fleet of airborne DC ‘blimps’ which could be stationed above cities or sporting/music events. Rapid delivery of items would be handled through large numbers of drones, as would delivery of items from Amazon’s multi-level ‘beehive’ tower DC concept.

Of these ideas, Magliola asks “What happens if capability leaps are achieved in additive manufacturing making delivery itself irrelevant? That said, I wouldn’t underestimate the potential for this future.”

A cross section of the proposed “beehive” drone DC.

Fallon points out that just one ‘beehive’ would likely need 50,000 drones to achieve a serious production level. “You end up with the same kind of space inefficiency Amazon purports to solve with this design,” he argues. “Like airplanes, making sure drones don’t collide means giving them a wide berth to manoeuvre.”

As for the blimp warehouse idea, in Fallon’s view it would require an “atrocious” carbon footprint that would earn Amazon a fierce public backlash.

There are other literally down-to-Earth DC solutions that Amazon is examining, and they are not new. These include mobile mini-warehouse trucks equipped with 3D printers for enroute manufacturing.

“It’s a fairly conventional idea that Amazon can use,” says Fallon. “As to 3D printing—perhaps a limited set of the most popular items or category of items could be 3D printed. I doubt it would be game-changing, but certainly could be useful in expanding the variety of SKUs that a truck-based mini-warehouse could deliver to its customers.” Magliola believes the concept of mobile 3D printing to be inevitable.

In the end, Fallon sees Amazon’s recent far-out DC patents as nothing more than a feeble attempt to shift the focus away from a grave misstep.

“Amazon has spent an incredible sum on Kiva materials handling technology that creates as many problems for it as it solves,” he explains. “This is a classic situation where a company has made a bad infrastructure investment and cannot see a way to correct it, so it flails about searching for a ‘paradigm shift’ to deliver a wholesale change in the way it does distribution. It’s a way of telling shareholders, ‘we didn’t make a mistake, we took current technology as far as it could and we have re-invent distribution to get where we need to go.’ At some point, somewhere a shareholder will grow tired of this story and say ‘maybe you made a mistake’. The real solution to Amazon’s space utilization dilemma is to admit it made a mistake. Then it can fix the problem without introducing modern-day Hindenburgs to the mix.”

“Amazon,” he declares, “is dreaming up the future, while not doing the hard work of distributing efficiently today.”