TEDxBoston 2011
Mick Mountz: What happens inside those massive warehouses?
Filmed:
Readability: 3.8
332,609 views
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.
Mick Mountz - CEO, Kiva Systems
Mick Mountz is the founder and CEO of Kiva Systems, making high-tech products for fast, cheap and efficient inventory fulfillment. In March 2012, Kiva was acquired by Amazon. Full bio
Mick Mountz is the founder and CEO of Kiva Systems, making high-tech products for fast, cheap and efficient inventory fulfillment. In March 2012, Kiva was acquired by Amazon. Full bio
Double-click the English transcript below to play the video.
00:11
I want to talk to you about,
0
0
3561
00:14
or share with you, a
breakthrough new approach
breakthrough new approach
1
3561
3561
00:18
for managing items of
inventory inside of a warehouse.
inventory inside of a warehouse.
2
7122
3342
00:21
We're talking about a pick,
pack and ship setting here.
pack and ship setting here.
3
10464
2558
00:24
So as a hint,
4
13022
2823
00:26
this solution involves
hundreds of mobile robots,
hundreds of mobile robots,
5
15845
4153
00:30
sometimes thousands
of mobile robots,
of mobile robots,
6
19998
2424
00:33
moving around a warehouse.
And I'll get to the solution.
And I'll get to the solution.
7
22422
2598
00:36
But for a moment, just think
8
25020
1220
00:37
about the last time that
you ordered something online.
you ordered something online.
9
26240
2526
00:39
You were sitting
on your couch
on your couch
10
28766
2064
00:41
and you decided that you
absolutely had to have this red t-shirt.
absolutely had to have this red t-shirt.
11
30830
4060
00:45
So — click! — you put it
into your shopping cart.
into your shopping cart.
12
34890
2296
00:48
And then you decided
that green pair of pants
that green pair of pants
13
37186
1898
00:50
looks pretty good too — click!
14
39084
1898
00:51
And maybe a blue
pair of shoes — click!
pair of shoes — click!
15
40982
1900
00:53
So at this point you've
assembled your order.
assembled your order.
16
42882
2050
00:55
You didn't stop to think
for a moment that
for a moment that
17
44932
1768
00:57
that might not be a great outfit.
18
46700
1768
00:59
But you hit
"submit order."
"submit order."
19
48468
1770
01:01
And two days later, this package
shows up on your doorstep.
shows up on your doorstep.
20
50238
4174
01:05
And you open the box and you're
like, wow, there's my goo.
like, wow, there's my goo.
21
54412
2954
01:08
Did you ever stop to think about
how those items of inventory
how those items of inventory
22
57366
2961
01:11
actually found their way inside
that box in the warehouse?
that box in the warehouse?
23
60327
4213
01:15
So I'm here to tell you
it's that guy right there.
it's that guy right there.
24
64540
3818
01:19
So deep in the
middle of that picture,
middle of that picture,
25
68358
3620
01:22
you see a classic
pick-pack worker
pick-pack worker
26
71978
2515
01:25
in a distribution or
order fulfillments setting.
order fulfillments setting.
27
74493
3225
01:28
Classically these pick workers will
spend 60 or 70 percent of their day
spend 60 or 70 percent of their day
28
77718
3882
01:32
wandering around
the warehouse.
the warehouse.
29
81600
1912
01:34
They'll often walk
as much as 5 or 10 miles
as much as 5 or 10 miles
30
83512
2711
01:37
in pursuit of
those items of inventory.
those items of inventory.
31
86223
2661
01:39
Not only is this an
unproductive way to fill orders,
unproductive way to fill orders,
32
88884
4421
01:44
it also turns out to be an
unfulfilling way to fill orders.
unfulfilling way to fill orders.
33
93305
4057
01:48
So let me tell you where I
first bumped into this problem.
first bumped into this problem.
34
97362
3631
01:51
I was out in the Bay area
in '99, 2000, the dot com boom.
in '99, 2000, the dot com boom.
35
100993
4054
01:56
I worked for a fabulously
spectacular flame-out called Webvan.
spectacular flame-out called Webvan.
36
105047
4062
02:00
(Laughter)
37
109109
1821
02:01
This company raised hundreds of
millions of dollars with the notion that
millions of dollars with the notion that
38
110930
2730
02:04
we will deliver
grocery orders online.
grocery orders online.
39
113660
2782
02:07
And it really came down to the fact
that we couldn't do it cost effectively.
that we couldn't do it cost effectively.
40
116442
4173
02:11
Turns out e-commerce was something
that was very hard and very costly.
that was very hard and very costly.
41
120615
3962
02:15
In this particular instance we were trying
to assemble 30 items of inventory
to assemble 30 items of inventory
42
124577
4251
02:19
into a few totes, onto a van
to deliver to the home.
to deliver to the home.
43
128828
3972
02:23
And when you think about it,
it was costing us 30 dollars.
it was costing us 30 dollars.
44
132800
3800
02:27
Imagine, we had an
89¢ can of soup
89¢ can of soup
45
136600
3086
02:30
that was costing us one dollar to
pick and pack into that tote.
pick and pack into that tote.
46
139686
3676
02:34
And that's before we actually
tried to deliver it to the home.
tried to deliver it to the home.
47
143362
3208
02:37
So long story short,
during my one year at Webvan,
during my one year at Webvan,
48
146570
2787
02:40
what I realized by talking to
all the material-handling providers
all the material-handling providers
49
149357
3236
02:43
was that there was no solution designed
specifically to solve each base picking.
specifically to solve each base picking.
50
152593
4638
02:48
Red item, green, blue, getting
those three things in a box.
those three things in a box.
51
157231
4023
02:52
So we said, there's just
got to be a better way to do this.
got to be a better way to do this.
52
161254
2985
02:55
Existing material handling
was set up to pump
was set up to pump
53
164239
2527
02:57
pallets and cases of
goo to retail stores.
goo to retail stores.
54
166766
3767
03:01
Of course Webvan went out of business,
and about a year and a half later,
and about a year and a half later,
55
170533
3797
03:05
I was still noodling on this problem.
It was still nagging at me.
It was still nagging at me.
56
174330
3580
03:08
And I started
thinking about it again.
thinking about it again.
57
177910
1868
03:10
And I said, let me just focus briefly
on what I wanted as a pick worker,
on what I wanted as a pick worker,
58
179778
5127
03:15
or my vision for
how it should work.
how it should work.
59
184905
2331
03:18
(Laughter)
60
187236
1414
03:19
I said, let's focus
on the problem.
on the problem.
61
188650
1922
03:21
I have an order here and what
I want to do is I want to put
I want to do is I want to put
62
190572
3855
03:25
red, green and blue
in this box right here.
in this box right here.
63
194427
2220
03:27
What I need is a system where I put out
my hand and — poof! —
my hand and — poof! —
64
196647
2852
03:30
the product shows up
and I pack it into the order,
and I pack it into the order,
65
199499
2523
03:33
and now we're thinking,
66
202022
1693
03:34
this would be a very operator-centric
approach to solving the problem.
approach to solving the problem.
67
203715
4014
03:38
This is what I need. What technology
is available to solve this problem?
is available to solve this problem?
68
207729
4227
03:42
But as you can see, orders can come
and go, products can come and go.
and go, products can come and go.
69
211956
3941
03:46
It allows us to focus on making the
pick worker the center of the problem,
pick worker the center of the problem,
70
215897
4683
03:51
and providing them the tools to make
them as productive as possible.
them as productive as possible.
71
220580
4624
03:56
So how did I
arrive at this notion?
arrive at this notion?
72
225204
2100
03:58
Well, actually it came from
a brainstorming exercise,
a brainstorming exercise,
73
227304
3533
04:01
probably a technique
that many of you use,
that many of you use,
74
230837
2840
04:04
It's this notion of
testing your ideas.
testing your ideas.
75
233677
2156
04:06
Take a blank sheet, of course,
76
235833
1730
04:08
but then test your ideas
at the limits — infinity, zero.
at the limits — infinity, zero.
77
237563
4030
04:12
In this particular case, we
challenged ourselves with the idea:
challenged ourselves with the idea:
78
241593
3182
04:15
What if we had to build a
distribution center in China,
distribution center in China,
79
244775
2976
04:18
where it's a very,
very low-cost market?
very low-cost market?
80
247751
2614
04:21
And say, labor is cheap,
land is cheap.
land is cheap.
81
250365
3213
04:24
And we said specifically,
82
253578
1646
04:26
"What if it was zero dollars
an hour for direct labor
an hour for direct labor
83
255224
3006
04:29
and we could build a million-
square-foot distribution center?"
square-foot distribution center?"
84
258230
2716
04:31
So naturally that
led to ideas that said,
led to ideas that said,
85
260946
2075
04:34
"Let's put lots of people
in the warehouse."
in the warehouse."
86
263021
1808
04:35
And I said, "Hold on,
zero dollars per hour,
zero dollars per hour,
87
264829
2545
04:38
what I would do is 'hire'
88
267374
2857
04:41
10,000 workers to come to the
warehouse every morning at 8 a.m.,
warehouse every morning at 8 a.m.,
89
270231
4127
04:45
walk into the warehouse and
pick up one item of inventory
pick up one item of inventory
90
274358
3119
04:48
and then just stand there.
91
277477
1733
04:50
So you hold Captain Crunch,
you hold the Mountain Dew,
you hold the Mountain Dew,
92
279210
2539
04:52
you hold the Diet Coke.
93
281749
1423
04:54
If I need it, I'll call you,
otherwise just stand there.
otherwise just stand there.
94
283172
2256
04:56
But when I need Diet Coke and I call it,
you guys talk amongst yourselves.
you guys talk amongst yourselves.
95
285428
3387
04:59
Diet Coke walks up to the front —
pick it, put it in the tote, away it goes."
pick it, put it in the tote, away it goes."
96
288815
4121
05:03
Wow, what if the products
could walk and talk on their own?
could walk and talk on their own?
97
292936
4652
05:08
That's a very interesting,
very powerful way
very powerful way
98
297588
2338
05:10
that we could potentially
organize this warehouse.
organize this warehouse.
99
299926
2594
05:13
So of course,
labor isn't free,
labor isn't free,
100
302520
2348
05:15
on that practical
versus awesome spectrum.
versus awesome spectrum.
101
304868
3308
05:19
(Laughter)
102
308176
1690
05:20
So we said mobile shelving —
We'll put them on mobile shelving.
We'll put them on mobile shelving.
103
309866
3117
05:23
We'll use mobile robots and
we'll move the inventory around.
we'll move the inventory around.
104
312983
4793
05:28
And so we got underway on that and
then I'm sitting on my couch in 2008.
then I'm sitting on my couch in 2008.
105
317776
4418
05:33
Did any of you see the Beijing
Olympics, the opening ceremonies?
Olympics, the opening ceremonies?
106
322194
3819
05:37
I about fell out of my
couch when I saw this.
couch when I saw this.
107
326013
3041
05:40
I'm like, that was the idea!
108
329054
1473
05:41
(Laughter and Applause)
109
330527
4499
05:46
We'll put thousands of people on
the warehouse floor, the stadium floor.
the warehouse floor, the stadium floor.
110
335026
3891
05:49
But interestingly enough, this
actually relates to the idea
actually relates to the idea
111
338917
4035
05:53
in that these guys were creating some
incredibly powerful, impressive digital art,
incredibly powerful, impressive digital art,
112
342952
5412
05:59
all without computers,
I'm told,
I'm told,
113
348364
2076
06:01
it was all peer-to-peer
coordination and communication.
coordination and communication.
114
350440
2446
06:03
You stand up,
I'll squat down.
I'll squat down.
115
352886
1576
06:05
And they made
some fabulous art.
some fabulous art.
116
354462
1645
06:07
It speaks to the
power of emergence
power of emergence
117
356107
2688
06:09
in systems when you let things
start to talk with each other.
start to talk with each other.
118
358795
3968
06:13
So that was a little
bit of the journey.
bit of the journey.
119
362763
3778
06:17
So of course, now what became
the practical reality of this idea?
the practical reality of this idea?
120
366541
4038
06:21
Here is a warehouse.
121
370579
1806
06:23
It's a pick, pack and ship center
that has about 10,000 different SKUs.
that has about 10,000 different SKUs.
122
372385
3776
06:27
We'll call them red pens,
green pens, yellow Post-It Notes.
green pens, yellow Post-It Notes.
123
376161
4137
06:31
We send the little orange robots
out to pick up the blue shelving pods.
out to pick up the blue shelving pods.
124
380298
3379
06:34
And we deliver them
to the side of the building.
to the side of the building.
125
383677
2168
06:36
So all the pick workers now
get to stay on the perimeter.
get to stay on the perimeter.
126
385845
3136
06:39
And the game here is
to pick up the shelves,
to pick up the shelves,
127
388981
2342
06:42
take them down the highway and
deliver them straight to the pick worker.
deliver them straight to the pick worker.
128
391323
3324
06:45
This pick worker's life
is completely different.
is completely different.
129
394647
2137
06:47
Rather than wandering around
the warehouse, she gets to stay still
the warehouse, she gets to stay still
130
396784
3417
06:51
in a pick station like this
131
400201
1588
06:52
and every product in the
building can now come to her.
building can now come to her.
132
401789
3998
06:56
So the process
is very productive.
is very productive.
133
405787
3278
07:00
Reach in, pick an item,
scan the bar code, pack it out.
scan the bar code, pack it out.
134
409065
4195
07:04
By the time
you turn around,
you turn around,
135
413260
1649
07:05
there's another product there
ready to be picked and packed.
ready to be picked and packed.
136
414909
3039
07:08
So what we've done is take
out all of the non-value added
out all of the non-value added
137
417948
3010
07:11
walking, searching,
wasting, waited time,
wasting, waited time,
138
420958
2684
07:14
and we've developed a very
high-fidelity way to pick these orders,
high-fidelity way to pick these orders,
139
423642
4278
07:18
where you point at it with
a laser, scan the UPC barcode,
a laser, scan the UPC barcode,
140
427920
4164
07:23
and then indicate with a light
which box it needs to go into.
which box it needs to go into.
141
432084
3194
07:26
So more productive, more
accurate and, it turns out,
accurate and, it turns out,
142
435278
3486
07:29
it's a more interesting office
environment for these pick workers.
environment for these pick workers.
143
438764
4306
07:34
They actually complete
the whole order.
the whole order.
144
443070
2437
07:36
So they do red, green and blue,
not just a part of the order.
not just a part of the order.
145
445507
2527
07:39
And they feel a little bit more
in control of their environment.
in control of their environment.
146
448034
3614
07:42
So the side effects
of this approach
of this approach
147
451648
3198
07:45
are what really surprised us.
148
454846
1568
07:47
We knew it was going
to be more productive.
to be more productive.
149
456414
1578
07:48
But we didn't realize just how
pervasive this way of thinking
pervasive this way of thinking
150
457992
4070
07:53
extended to other
functions in the warehouse.
functions in the warehouse.
151
462062
5230
07:58
But what effectively this approach
is doing inside of the DC
is doing inside of the DC
152
467292
4832
08:03
is turning it into a massively
parallel processing engine.
parallel processing engine.
153
472124
4464
08:07
So this is again a
cross-fertilization of ideas.
cross-fertilization of ideas.
154
476588
2623
08:10
Here's a warehouse
and we're thinking about
and we're thinking about
155
479211
1939
08:12
parallel processing
supercomputer architectures.
supercomputer architectures.
156
481150
3519
08:15
The notion here is that you have
157
484669
2397
08:18
10 workers on
the right side of the screen
the right side of the screen
158
487066
2797
08:20
that are now all independent
autonomous pick workers.
autonomous pick workers.
159
489863
4248
08:25
If the worker in station three decides
to leave and go to the bathroom,
to leave and go to the bathroom,
160
494111
3904
08:29
it has no impact on the
productivity of the other nine workers.
productivity of the other nine workers.
161
498015
3484
08:32
Contrast that, for a moment, with the
traditional method of using a conveyor.
traditional method of using a conveyor.
162
501499
4370
08:36
When one person
passes the order to you,
passes the order to you,
163
505869
2059
08:38
you put something in
and pass it downstream.
and pass it downstream.
164
507928
2499
08:41
Everyone has to be in place
for that serial process to work.
for that serial process to work.
165
510427
3117
08:44
This becomes a more robust
way to think about the warehouse.
way to think about the warehouse.
166
513544
3177
08:47
And then underneath the hoods gets
interesting in that we're tracking
interesting in that we're tracking
167
516721
4544
08:52
the popularity
of the products.
of the products.
168
521265
1738
08:54
And we're using dynamic
and adaptive algorithms
and adaptive algorithms
169
523003
2824
08:56
to tune the floor
of the warehouse.
of the warehouse.
170
525827
4454
09:01
So what you see here potentially
the week leading up to Valentine's Day.
the week leading up to Valentine's Day.
171
530281
4884
09:06
All that pink chalky candy has
moved to the front of the building
moved to the front of the building
172
535165
3843
09:10
and is now being picked into a
lot of orders in those pick stations.
lot of orders in those pick stations.
173
539008
3893
09:13
Come in two days after Valentine's Day,
and that candy, the leftover candy,
and that candy, the leftover candy,
174
542901
4163
09:18
has all drifted to the
back of the warehouse
back of the warehouse
175
547064
2273
09:20
and is occupying the cooler
zone on the thermal map there.
zone on the thermal map there.
176
549337
4293
09:24
One other side effect of this
approach using the parallel processing
approach using the parallel processing
177
553630
3483
09:28
is these things can
scale to ginormous.
scale to ginormous.
178
557113
2995
09:31
(Laughter)
179
560108
1527
09:32
So whether you're doing
two pick stations, 20 pick stations,
two pick stations, 20 pick stations,
180
561635
2734
09:35
or 200 pick stations, the
path planning algorithms
path planning algorithms
181
564369
3273
09:38
and all of the inventory
algorithms just work.
algorithms just work.
182
567642
2543
09:41
In this example you
see that the inventory
see that the inventory
183
570185
4356
09:45
has now occupied all the
perimeter of the building
perimeter of the building
184
574541
2326
09:47
because that's where
the pick stations were.
the pick stations were.
185
576867
2346
09:50
They sorted it
out for themselves.
out for themselves.
186
579213
2018
09:52
So I'll conclude with
just one final video
just one final video
187
581231
2287
09:54
that shows how
this comes to bear
this comes to bear
188
583518
2984
09:57
on the pick worker's actual
day in the life of.
day in the life of.
189
586502
3594
10:01
So as we mentioned, the process is
to move inventory along the highway
to move inventory along the highway
190
590096
4336
10:05
and then find your way
into these pick stations.
into these pick stations.
191
594432
2652
10:08
And our software in the background
192
597084
2471
10:10
understands what's going on
in each station,
in each station,
193
599555
2751
10:13
we direct the pods
across the highway
across the highway
194
602306
2611
10:15
and we're attempting to
get into a queuing system
get into a queuing system
195
604917
2714
10:18
to present the work
to the pick worker.
to the pick worker.
196
607631
2884
10:21
What's interesting is we can even
adapt the speed of the pick workers.
adapt the speed of the pick workers.
197
610515
3245
10:24
The faster pickers get more pods
and the slower pickers get few.
and the slower pickers get few.
198
613760
3926
10:28
But this pick worker now is
literally having that experience
literally having that experience
199
617686
3138
10:31
that we described before.
200
620824
1853
10:33
She puts out her hand.
The product jumps into it.
The product jumps into it.
201
622677
2544
10:36
Or she has to reach in and get it.
202
625221
2205
10:38
She scans it and
she puts it in the bucket.
she puts it in the bucket.
203
627426
2445
10:40
And all of the rest of the technology
is kind of behind the scenes.
is kind of behind the scenes.
204
629871
3796
10:44
So she gets to now focus on the
picking and packing portion of her job.
picking and packing portion of her job.
205
633667
3861
10:48
Never has any idle time,
never has to leave her mat.
never has to leave her mat.
206
637528
3274
10:51
And actually we think
not only a more productive
not only a more productive
207
640802
4140
10:55
and more accurate
way to fill orders.
way to fill orders.
208
644942
3270
10:59
We think it's a more
fulfilling way to fill orders.
fulfilling way to fill orders.
209
648212
3410
11:02
The reason we can say
that, though, is that workers
that, though, is that workers
210
651622
3208
11:05
in a lot of these
buildings now compete
buildings now compete
211
654830
2142
11:07
for the privilege of working
in the Kiva zone that day.
in the Kiva zone that day.
212
656972
3182
11:11
And sometimes we'll catch
them on testimonial videos
them on testimonial videos
213
660154
2663
11:13
saying such things as,
214
662817
2117
11:15
they have more energy after the
day to play with their grandchildren,
day to play with their grandchildren,
215
664934
4246
11:20
or in one case a guy said, "the
Kiva zone is so stress-free
Kiva zone is so stress-free
216
669180
4596
11:24
that I've actually stopped taking
my blood pressure medication."
my blood pressure medication."
217
673776
3115
11:27
(Laughter)
218
676891
1834
11:29
That was at a pharmaceutical distributor,
so they told us not to use that video.
so they told us not to use that video.
219
678725
3999
11:33
(Laughter)
220
682724
3568
11:37
So what I wanted to leave you
with today is the notion that
with today is the notion that
221
686292
3024
11:40
when you let things start
to think and walk
to think and walk
222
689316
2855
11:43
and talk on their own, interesting
processes and productivities can emerge.
processes and productivities can emerge.
223
692171
5291
11:48
And now I think next time
you go to your front step
you go to your front step
224
697462
2941
11:51
and pick up that box that
you just ordered online,
you just ordered online,
225
700403
2511
11:53
you break it open and
the goo is in there,
the goo is in there,
226
702914
2224
11:56
you'll have some wonderment
as to whether a robot
as to whether a robot
227
705138
2753
11:58
assisted in the picking
and packing of that order.
and packing of that order.
228
707891
2593
12:01
Thank you.
229
710484
1679
12:03
(Applause)
230
712163
4630
ABOUT THE SPEAKER
Mick Mountz - CEO, Kiva SystemsMick Mountz is the founder and CEO of Kiva Systems, making high-tech products for fast, cheap and efficient inventory fulfillment. In March 2012, Kiva was acquired by Amazon.
Why you should listen
With the growing popularity of online shopping, Mick Mountz and his company are finding ways to make picking, packing and shipping those purchases a flexible and inexpensive process. He first began working on this hidden challenge while at Webvan, an online grocer. In 2003, he took his experience there and founded Kiva systems, a company whose innovative solutions have led to it being listed by Inc. as one of the fastest growing companies.
In March 2012, Kiva was acquired by Amazon for $775 million.
Mick Mountz | Speaker | TED.com