We have entered the warehouse robotics era. How far and how fast we move down that path are the real questions.
A glimpse at that future is nicely provided by a massive new Amazon Fulfillment Center in Shreveport, Louisiana.
As we reported in October, the new facility comes in at incredible 3 million square feet and uses more than ten times the number of robots than did the previous Amazon state-of-of-the-art FC design.
Gilmore Says.... |
 |
I also assume Amazon is still using some of the Kiva System “goods-to-person” technology it acquired for $750 million or so in 2012. With that, robots bring inventory carts to stationary pickers.
|
 |
What do you say? |
|
Click here to send us your comments |
|
|
|
|
The Wall Street Journal last week took a more detailed look at this new FC of the future – and just how human workers will be required.
Notable is this: Amazon says this is its first FC that uses automation and artificial intelligence at every step of the fulfillment process.
That many robots it seems to me must be replacing a large number of human workers – but we still appear a good distance away from a nearly completely robotics fulfillment operation.
The Shreveport FC currently employs about 1400 such human workers, on its way to as many as 2500 when the facility is fully operating.
What we don’t have, as far as I know, is some metric such as the number of human workers per order or line-item shipped for previous and this current FC design that would allow a better understanding of the impact of the FC automation on jobs.
Amazon and others are getting smart about positioning this, focusing on how the automation moves the most physically demanding and repetitive fulfillment tasks to robots, with a related reduction in injuries to workers.
Amazon has received much scrutiny over the past decade over allegedly higher than average injury rates at its FC, which Amazon has challenged as painting an unfair picture, while at the same emphasizing in the past few years how automation measurably reduces industry rates.
With all the robots, why does Amazon still need all the humans? The easy answer is that are some things robots still can’t do very well.
The Journal article notes the vast array of the incredible 400 million – that’s right 400 million – items that Amazon sells globally, of every size, shape and weight. That makes it very difficult to build robots that can handle them all.
The article quotes my friend Rueben Scriven, research manager for the warehouse automation sector at research firm Interact Analysis, as saying that “If you don’t know what items you’re going to be handling, it makes it very difficult to create an automated system that’s flexible enough to handle the various items.”
The good news, it seems to me: here humans still have some real advantages. The Journal notes that “Humans can easily look into a storage container packed full of goods, identify a particular item and know how to pick it up and handle it, whether it is a bottle of shampoo or a sweater.”
Meanwhile, Tye Brady, chief technologist at Amazon Robotics, adds “The tactile grasp that the human hand has, and the situational awareness and the perception of the human brain, is unmatched.”
How long will that advantage hold? Decades? Years? Not long at all?
The article also notes that truck loading and unloading remain areas where automation is challenged.
Amazon is using three different types of robotic arms but none of them are selecting items out of bins, but rather used for so-called “pick and place” applications where a robot picks the item from a conveyor belt and puts in a shipping or other container.
I also assume Amazon is still using some of the Kiva System “goods-to-person” technology it acquired for $750 million or so in 2012. With that, robots bring inventory carts to stationary pickers. Amazon has deployed tens of thousands of such mobile robots.
So I will say this. First, my guess is Amazon could have taken this even further, but felt this was the next technology increment that made sense, for both operational and political reasons.
Second, as with many manufacturing jobs that involve low skill, repetitive tasks, just how many of these FC jobs do we want to save in the end?
There have been many reports on how hard the jobs at Amazon are. I would just say for hundreds of thousands of workers it beats the alternative.
Any reaction to this colum on warehouse robotics from Gilmore? Let us know your thoughts at the Feedback button (email) or section below.
Your Comments/Feedback.
|