What happens inside those massive warehouses?
-
0:00 - 0:04I want to talk to you about,
-
0:04 - 0:07or share with you, a
breakthrough new approach -
0:07 - 0:10for managing items of
inventory inside of a warehouse. -
0:10 - 0:13We're talking about a pick,
pack and ship setting here. -
0:13 - 0:16So as a hint,
-
0:16 - 0:20this solution involves
hundreds of mobile robots, -
0:20 - 0:22sometimes thousands
of mobile robots, -
0:22 - 0:25moving around a warehouse.
And I'll get to the solution. -
0:25 - 0:26But for a moment, just think
-
0:26 - 0:29about the last time that
you ordered something online. -
0:29 - 0:31You were sitting
on your couch -
0:31 - 0:35and you decided that you
absolutely had to have this red t-shirt. -
0:35 - 0:37So — click! — you put it
into your shopping cart. -
0:37 - 0:39And then you decided
that green pair of pants -
0:39 - 0:41looks pretty good too — click!
-
0:41 - 0:43And maybe a blue
pair of shoes — click! -
0:43 - 0:45So at this point you've
assembled your order. -
0:45 - 0:47You didn't stop to think
for a moment that -
0:47 - 0:48that might not be a great outfit.
-
0:48 - 0:50But you hit
"submit order." -
0:50 - 0:54And two days later, this package
shows up on your doorstep. -
0:54 - 0:57And you open the box and you're
like, wow, there's my goo. -
0:57 - 1:00Did you ever stop to think about
how those items of inventory -
1:00 - 1:05actually found their way inside
that box in the warehouse? -
1:05 - 1:08So I'm here to tell you
it's that guy right there. -
1:08 - 1:12So deep in the
middle of that picture, -
1:12 - 1:14you see a classic
pick-pack worker -
1:14 - 1:18in a distribution or
order fulfillments setting. -
1:18 - 1:22Classically these pick workers will
spend 60 or 70 percent of their day -
1:22 - 1:24wandering around
the warehouse. -
1:24 - 1:26They'll often walk
as much as 5 or 10 miles -
1:26 - 1:29in pursuit of
those items of inventory. -
1:29 - 1:33Not only is this an
unproductive way to fill orders, -
1:33 - 1:37it also turns out to be an
unfulfilling way to fill orders. -
1:37 - 1:41So let me tell you where I
first bumped into this problem. -
1:41 - 1:45I was out in the Bay area
in '99, 2000, the dot com boom. -
1:45 - 1:49I worked for a fabulously
spectacular flame-out called Webvan. -
1:49 - 1:51(Laughter)
-
1:51 - 1:54This company raised hundreds of
millions of dollars with the notion that -
1:54 - 1:56we will deliver
grocery orders online. -
1:56 - 2:01And it really came down to the fact
that we couldn't do it cost effectively. -
2:01 - 2:05Turns out e-commerce was something
that was very hard and very costly. -
2:05 - 2:09In this particular instance we were trying
to assemble 30 items of inventory -
2:09 - 2:13into a few totes, onto a van
to deliver to the home. -
2:13 - 2:17And when you think about it,
it was costing us 30 dollars. -
2:17 - 2:20Imagine, we had an
89¢ can of soup -
2:20 - 2:23that was costing us one dollar to
pick and pack into that tote. -
2:23 - 2:27And that's before we actually
tried to deliver it to the home. -
2:27 - 2:29So long story short,
during my one year at Webvan, -
2:29 - 2:33what I realized by talking to
all the material-handling providers -
2:33 - 2:37was that there was no solution designed
specifically to solve each base picking. -
2:37 - 2:41Red item, green, blue, getting
those three things in a box. -
2:41 - 2:44So we said, there's just
got to be a better way to do this. -
2:44 - 2:47Existing material handling
was set up to pump -
2:47 - 2:51pallets and cases of
goo to retail stores. -
2:51 - 2:54Of course Webvan went out of business,
and about a year and a half later, -
2:54 - 2:58I was still noodling on this problem.
It was still nagging at me. -
2:58 - 3:00And I started
thinking about it again. -
3:00 - 3:05And I said, let me just focus briefly
on what I wanted as a pick worker, -
3:05 - 3:07or my vision for
how it should work. -
3:07 - 3:09(Laughter)
-
3:09 - 3:11I said, let's focus
on the problem. -
3:11 - 3:14I have an order here and what
I want to do is I want to put -
3:14 - 3:17red, green and blue
in this box right here. -
3:17 - 3:19What I need is a system where I put out
my hand and — poof! — -
3:19 - 3:22the product shows up
and I pack it into the order, -
3:22 - 3:24and now we're thinking,
-
3:24 - 3:28this would be a very operator-centric
approach to solving the problem. -
3:28 - 3:32This is what I need. What technology
is available to solve this problem? -
3:32 - 3:36But as you can see, orders can come
and go, products can come and go. -
3:36 - 3:41It allows us to focus on making the
pick worker the center of the problem, -
3:41 - 3:45and providing them the tools to make
them as productive as possible. -
3:45 - 3:47So how did I
arrive at this notion? -
3:47 - 3:51Well, actually it came from
a brainstorming exercise, -
3:51 - 3:54probably a technique
that many of you use, -
3:54 - 3:56It's this notion of
testing your ideas. -
3:56 - 3:58Take a blank sheet, of course,
-
3:58 - 4:02but then test your ideas
at the limits — infinity, zero. -
4:02 - 4:05In this particular case, we
challenged ourselves with the idea: -
4:05 - 4:08What if we had to build a
distribution center in China, -
4:08 - 4:10where it's a very,
very low-cost market? -
4:10 - 4:14And say, labor is cheap,
land is cheap. -
4:14 - 4:15And we said specifically,
-
4:15 - 4:18"What if it was zero dollars
an hour for direct labor -
4:18 - 4:21and we could build a million-
square-foot distribution center?" -
4:21 - 4:23So naturally that
led to ideas that said, -
4:23 - 4:25"Let's put lots of people
in the warehouse." -
4:25 - 4:27And I said, "Hold on,
zero dollars per hour, -
4:27 - 4:30what I would do is 'hire'
-
4:30 - 4:3410,000 workers to come to the
warehouse every morning at 8 a.m., -
4:34 - 4:37walk into the warehouse and
pick up one item of inventory -
4:37 - 4:39and then just stand there.
-
4:39 - 4:42So you hold Captain Crunch,
you hold the Mountain Dew, -
4:42 - 4:43you hold the Diet Coke.
-
4:43 - 4:45If I need it, I'll call you,
otherwise just stand there. -
4:45 - 4:49But when I need Diet Coke and I call it,
you guys talk amongst yourselves. -
4:49 - 4:53Diet Coke walks up to the front —
pick it, put it in the tote, away it goes." -
4:53 - 4:58Wow, what if the products
could walk and talk on their own? -
4:58 - 5:00That's a very interesting,
very powerful way -
5:00 - 5:03that we could potentially
organize this warehouse. -
5:03 - 5:05So of course,
labor isn't free, -
5:05 - 5:08on that practical
versus awesome spectrum. -
5:08 - 5:10(Laughter)
-
5:10 - 5:13So we said mobile shelving —
We'll put them on mobile shelving. -
5:13 - 5:18We'll use mobile robots and
we'll move the inventory around. -
5:18 - 5:22And so we got underway on that and
then I'm sitting on my couch in 2008. -
5:22 - 5:26Did any of you see the Beijing
Olympics, the opening ceremonies? -
5:26 - 5:29I about fell out of my
couch when I saw this. -
5:29 - 5:31I'm like, that was the idea!
-
5:31 - 5:35(Laughter and Applause)
-
5:35 - 5:39We'll put thousands of people on
the warehouse floor, the stadium floor. -
5:39 - 5:43But interestingly enough, this
actually relates to the idea -
5:43 - 5:48in that these guys were creating some
incredibly powerful, impressive digital art, -
5:48 - 5:50all without computers,
I'm told, -
5:50 - 5:53it was all peer-to-peer
coordination and communication. -
5:53 - 5:54You stand up,
I'll squat down. -
5:54 - 5:56And they made
some fabulous art. -
5:56 - 5:59It speaks to the
power of emergence -
5:59 - 6:03in systems when you let things
start to talk with each other. -
6:03 - 6:07So that was a little
bit of the journey. -
6:07 - 6:11So of course, now what became
the practical reality of this idea? -
6:11 - 6:12Here is a warehouse.
-
6:12 - 6:16It's a pick, pack and ship center
that has about 10,000 different SKUs. -
6:16 - 6:20We'll call them red pens,
green pens, yellow Post-It Notes. -
6:20 - 6:24We send the little orange robots
out to pick up the blue shelving pods. -
6:24 - 6:26And we deliver them
to the side of the building. -
6:26 - 6:29So all the pick workers now
get to stay on the perimeter. -
6:29 - 6:31And the game here is
to pick up the shelves, -
6:31 - 6:35take them down the highway and
deliver them straight to the pick worker. -
6:35 - 6:37This pick worker's life
is completely different. -
6:37 - 6:40Rather than wandering around
the warehouse, she gets to stay still -
6:40 - 6:42in a pick station like this
-
6:42 - 6:46and every product in the
building can now come to her. -
6:46 - 6:49So the process
is very productive. -
6:49 - 6:53Reach in, pick an item,
scan the bar code, pack it out. -
6:53 - 6:55By the time
you turn around, -
6:55 - 6:58there's another product there
ready to be picked and packed. -
6:58 - 7:01So what we've done is take
out all of the non-value added -
7:01 - 7:04walking, searching,
wasting, waited time, -
7:04 - 7:08and we've developed a very
high-fidelity way to pick these orders, -
7:08 - 7:12where you point at it with
a laser, scan the UPC barcode, -
7:12 - 7:15and then indicate with a light
which box it needs to go into. -
7:15 - 7:19So more productive, more
accurate and, it turns out, -
7:19 - 7:23it's a more interesting office
environment for these pick workers. -
7:23 - 7:26They actually complete
the whole order. -
7:26 - 7:28So they do red, green and blue,
not just a part of the order. -
7:28 - 7:32And they feel a little bit more
in control of their environment. -
7:32 - 7:35So the side effects
of this approach -
7:35 - 7:36are what really surprised us.
-
7:36 - 7:38We knew it was going
to be more productive. -
7:38 - 7:42But we didn't realize just how
pervasive this way of thinking -
7:42 - 7:47extended to other
functions in the warehouse. -
7:47 - 7:52But what effectively this approach
is doing inside of the DC -
7:52 - 7:57is turning it into a massively
parallel processing engine. -
7:57 - 7:59So this is again a
cross-fertilization of ideas. -
7:59 - 8:01Here's a warehouse
and we're thinking about -
8:01 - 8:05parallel processing
supercomputer architectures. -
8:05 - 8:07The notion here is that you have
-
8:07 - 8:1010 workers on
the right side of the screen -
8:10 - 8:14that are now all independent
autonomous pick workers. -
8:14 - 8:18If the worker in station three decides
to leave and go to the bathroom, -
8:18 - 8:21it has no impact on the
productivity of the other nine workers. -
8:21 - 8:26Contrast that, for a moment, with the
traditional method of using a conveyor. -
8:26 - 8:28When one person
passes the order to you, -
8:28 - 8:30you put something in
and pass it downstream. -
8:30 - 8:34Everyone has to be in place
for that serial process to work. -
8:34 - 8:37This becomes a more robust
way to think about the warehouse. -
8:37 - 8:41And then underneath the hoods gets
interesting in that we're tracking -
8:41 - 8:43the popularity
of the products. -
8:43 - 8:46And we're using dynamic
and adaptive algorithms -
8:46 - 8:50to tune the floor
of the warehouse. -
8:50 - 8:55So what you see here potentially
the week leading up to Valentine's Day. -
8:55 - 8:59All that pink chalky candy has
moved to the front of the building -
8:59 - 9:03and is now being picked into a
lot of orders in those pick stations. -
9:03 - 9:07Come in two days after Valentine's Day,
and that candy, the leftover candy, -
9:07 - 9:09has all drifted to the
back of the warehouse -
9:09 - 9:14and is occupying the cooler
zone on the thermal map there. -
9:14 - 9:17One other side effect of this
approach using the parallel processing -
9:17 - 9:20is these things can
scale to ginormous. -
9:20 - 9:22(Laughter)
-
9:22 - 9:24So whether you're doing
two pick stations, 20 pick stations, -
9:24 - 9:28or 200 pick stations, the
path planning algorithms -
9:28 - 9:30and all of the inventory
algorithms just work. -
9:30 - 9:35In this example you
see that the inventory -
9:35 - 9:37has now occupied all the
perimeter of the building -
9:37 - 9:39because that's where
the pick stations were. -
9:39 - 9:41They sorted it
out for themselves. -
9:41 - 9:44So I'll conclude with
just one final video -
9:44 - 9:47that shows how
this comes to bear -
9:47 - 9:50on the pick worker's actual
day in the life of. -
9:50 - 9:54So as we mentioned, the process is
to move inventory along the highway -
9:54 - 9:57and then find your way
into these pick stations. -
9:57 - 10:00And our software in the background
-
10:00 - 10:02understands what's going on
in each station, -
10:02 - 10:05we direct the pods
across the highway -
10:05 - 10:08and we're attempting to
get into a queuing system -
10:08 - 10:11to present the work
to the pick worker. -
10:11 - 10:14What's interesting is we can even
adapt the speed of the pick workers. -
10:14 - 10:18The faster pickers get more pods
and the slower pickers get few. -
10:18 - 10:21But this pick worker now is
literally having that experience -
10:21 - 10:23that we described before.
-
10:23 - 10:25She puts out her hand.
The product jumps into it. -
10:25 - 10:27Or she has to reach in and get it.
-
10:27 - 10:30She scans it and
she puts it in the bucket. -
10:30 - 10:34And all of the rest of the technology
is kind of behind the scenes. -
10:34 - 10:38So she gets to now focus on the
picking and packing portion of her job. -
10:38 - 10:41Never has any idle time,
never has to leave her mat. -
10:41 - 10:45And actually we think
not only a more productive -
10:45 - 10:48and more accurate
way to fill orders. -
10:48 - 10:52We think it's a more
fulfilling way to fill orders. -
10:52 - 10:55The reason we can say
that, though, is that workers -
10:55 - 10:57in a lot of these
buildings now compete -
10:57 - 11:00for the privilege of working
in the Kiva zone that day. -
11:00 - 11:03And sometimes we'll catch
them on testimonial videos -
11:03 - 11:05saying such things as,
-
11:05 - 11:09they have more energy after the
day to play with their grandchildren, -
11:09 - 11:14or in one case a guy said, "the
Kiva zone is so stress-free -
11:14 - 11:17that I've actually stopped taking
my blood pressure medication." -
11:17 - 11:19(Laughter)
-
11:19 - 11:23That was at a pharmaceutical distributor,
so they told us not to use that video. -
11:23 - 11:26(Laughter)
-
11:26 - 11:29So what I wanted to leave you
with today is the notion that -
11:29 - 11:32when you let things start
to think and walk -
11:32 - 11:37and talk on their own, interesting
processes and productivities can emerge. -
11:37 - 11:40And now I think next time
you go to your front step -
11:40 - 11:43and pick up that box that
you just ordered online, -
11:43 - 11:45you break it open and
the goo is in there, -
11:45 - 11:48you'll have some wonderment
as to whether a robot -
11:48 - 11:50assisted in the picking
and packing of that order. -
11:50 - 11:52Thank you.
-
11:52 - 11:57(Applause)
- Title:
- What happens inside those massive warehouses?
- Speaker:
- Mick Mountz
- Description:
-
We make millions of online purchases daily, but who (or what) actually puts our items into packages? In this talk, Mick Mountz weaves a fascinating, surprisingly robot-filled tale of what happens inside a warehouse.
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDTalks
- Duration:
- 12:06
Morton Bast edited English subtitles for What happens inside those massive warehouses? | ||
tom carter approved English subtitles for What happens inside those massive warehouses? | ||
tom carter accepted English subtitles for What happens inside those massive warehouses? | ||
Morton Bast edited English subtitles for What happens inside those massive warehouses? | ||
Morton Bast edited English subtitles for What happens inside those massive warehouses? | ||
Morton Bast edited English subtitles for What happens inside those massive warehouses? | ||
Morton Bast edited English subtitles for What happens inside those massive warehouses? | ||
Morton Bast edited English subtitles for What happens inside those massive warehouses? |