I want to talk to you about,

我想談一下，

or share with you, a breakthrough new approach

或說分享一下一種突破性的新方法，

for managing items of inventory inside of a warehouse.

來做倉儲的存貨管理。

We're talking about a pick, pack and ship setting here.

我說的是取貨、包裝及出貨部分。

So as a hint,

我給點提示，

this solution involves hundreds of mobile robots,

這種方法要用上百個移動式機器人，

sometimes thousands of mobile robots,

有時甚至上千個移動式機器人，

moving around a warehouse. And I'll get to the solution.

在倉庫裡穿梭。我馬上會談方法。

But for a moment, just think

但現在，就請大家想一下

about the last time that you ordered something online.

你最近一次在網上購物的情景。

You were sitting on your couch

你坐在沙發上，

and you decided that you absolutely had to have this red t-shirt.

你看到一件紅色 T 恤， 想要的不得了。

So — click! — you put it into your shopping cart.

於是你一點，就把它放入購物車。

And then you decided that green pair of pants

然後你又看到一條綠色的褲子，

looks pretty good too — click!

也很不錯。又點！

And maybe a blue pair of shoes — click!

又發現一雙藍色的鞋子，點！

So at this point you've assembled your order.

這個時候你已經選好一整套服裝。

You didn't stop to think for a moment that

你一點也沒想過

that might not be a great outfit.

這樣的搭配可能不太好看。

But you hit "submit order."

但是你已下了訂單。

And two days later, this package shows up on your doorstep.

兩天之後，包裹出現在你家門前。

And you open the box and you're like, wow, there's my goo.

你打開箱子，哇，好貨來了！

Did you ever stop to think about how those items of inventory

你是否曾經想過這些東西的存貨

actually found their way inside that box in the warehouse?

如何從倉庫中找出再裝成包裹？

So I'm here to tell you it's that guy right there.

我現在就告訴你，就是那傢伙做的。

So deep in the middle of that picture,

在這張照片裡，有位小小的

you see a classic pick-pack worker

傳統取貨包裝工人，

in a distribution or order fulfillments setting.

在配送中心或履行訂單的地方工作。

Classically these pick workers will spend 60 or 70 percent of their day

傳統模式下，這些取貨員 每天要花六七成的上班時間

wandering around the warehouse.

穿梭在倉庫裡。

They'll often walk as much as 5 or 10 miles

他們累計要走 8-16 公里

in pursuit of those items of inventory.

來尋找所需的貨物。

Not only is this an unproductive way to fill orders,

這不但是沒什麼效率的發貨方式，

it also turns out to be an unfulfilling way to fill orders.

員工也沒什麼成就感。

So let me tell you where I first bumped into this problem.

所以讓我告訴你們 我第一次碰到這個問題的情況。

I was out in the Bay area in '99, 2000, the dot com boom.

1999-2000 年左右我在灣區， 當時正值電子商務潮。

I worked for a fabulously spectacular flame-out called Webvan.

我為一家曇花一現的網路雜貨宅配商 網路雜貨車 (Webvan) 工作。

(Laughter)

（笑）

This company raised hundreds of millions of dollars with the notion that

這家公司集資了幾億美元， 吸金的概念是

we will deliver grocery orders online.

網路超市宅配到家。

And it really came down to the fact that we couldn't do it cost effectively.

但美夢破碎， 因為這真的不符成本效益。

Turns out e-commerce was something that was very hard and very costly.

其實電子商務很難做，花費也大。

In this particular instance we were trying to assemble 30 items of inventory

在這個實例裡，我們試著將 存貨中的 30 件物品

into a few totes, onto a van to deliver to the home.

裝進袋子裡 ，裝上車並宅配到家。

And when you think about it, it was costing us 30 dollars.

你得瞭解，這要花我們 30 美元。

Imagine, we had an 89¢ can of soup

試想一下，一瓶罐頭湯才賣 89 美分，

that was costing us one dollar to pick and pack into that tote.

卻要花我們 1 美元的成本 來取貨及包裝。

And that's before we actually tried to deliver it to the home.

而且這還只是我們 宅配到家之前的花費。

So long story short, during my one year at Webvan,

所以長話短說， 我在網路雜貨車待的一年時間裡，

what I realized by talking to all the material-handling providers

跟很多物料運輸的廠家談過，

was that there was no solution designed specifically to solve each base picking.

我了解到並沒有特別 為基層取貨設計的方法。

Red item, green, blue, getting those three things in a box.

紅的物品、綠的、藍的， 把這三樣東西裝箱。

So we said, there's just got to be a better way to do this.

所以我們就想，一定有更好的辦法。

Existing material handling was set up to pump

現行的物料輸送法

pallets and cases of goo to retail stores.

是把貨板和貨箱發送到零售店。

Of course Webvan went out of business, and about a year and a half later,

當然網路雜貨車已經關門了， 在大概一年半後，

I was still noodling on this problem. It was still nagging at me.

我仍絞盡腦汁在想這個問題。 它一直困擾著我。

And I started thinking about it again.

我開始重新思考。

And I said, let me just focus briefly on what I wanted as a pick worker,

我說：讓我暫時簡化問題， 如果我是取貨員我要什麼。

or my vision for how it should work.

或是說我想怎麼做。

(Laughter)

（笑）

I said, let's focus on the problem.

我說，讓我們回到問題本身。

I have an order here and what I want to do is I want to put

我有一則訂單，我想做的是

red, green and blue in this box right here.

把紅、綠和藍放進這個箱子裡。

What I need is a system where I put out my hand and — poof! —

我需要的是一個系統， 讓我能一伸手，變！

the product shows up and I pack it into the order,

貨物就出現在我手上， 然後我就裝進箱子裡，

and now we're thinking,

現在我們在想，

this would be a very operator-centric approach to solving the problem.

這是以操作員為中心的解決辦法。

This is what I need. What technology is available to solve this problem?

這就是我想要的。 有什麼現成的科技能解決這個問題？

But as you can see, orders can come and go, products can come and go.

但是你也知道，訂單來來去去， 貨物也來來去去。

It allows us to focus on making the pick worker the center of the problem,

這讓我們專心把問題 重點放在取貨員上，

and providing them the tools to make them as productive as possible.

給他們工具以提高生產率。

So how did I arrive at this notion?

所以我是怎樣有這個概念的？

Well, actually it came from a brainstorming exercise,

其實是從一次腦力激盪而來，

probably a technique that many of you use,

你們大概都用過這個方法，

It's this notion of testing your ideas.

這是考驗你的想法的一種概念。

Take a blank sheet, of course,

當然，你要拿一張白紙，

but then test your ideas at the limits — infinity, zero.

但考驗你想法的限制是：無限、零。

In this particular case, we challenged ourselves with the idea:

這個例子，我們激盪一下這個想法：

What if we had to build a distribution center in China,

假如我們在中國 蓋一座配送中心會怎樣，

where it's a very, very low-cost market?

那裡是成本很低很低的市場？

And say, labor is cheap, land is cheap.

比如，勞力便宜，土地也便宜。

And we said specifically,

具體一點說，

"What if it was zero dollars an hour for direct labor

假如直接勞工費用是每小時零美元，

and we could build a million- square-foot distribution center?"

而且我們還能蓋一座占地 一百萬平方呎的配送中心？

So naturally that led to ideas that said,

所以我們很自然會想到一個方法：

"Let's put lots of people in the warehouse."

讓我們在倉庫裡放很多人。

And I said, "Hold on, zero dollars per hour,

然後我說：等等，每小時零元，

what I would do is 'hire'

那我要「雇」

10,000 workers to come to the warehouse every morning at 8 a.m.,

一萬名工人每天早上八點到倉庫，

walk into the warehouse and pick up one item of inventory

走進倉庫裡並從存貨中取一項貨品，

and then just stand there.

然後站在那。

So you hold Captain Crunch, you hold the Mountain Dew,

所以你拿桂格早餐片， 你拿山泉汽水，

you hold the Diet Coke.

你拿健怡可樂。

If I need it, I'll call you, otherwise just stand there.

我需要的時候會叫你們， 平常你們站著就好。

But when I need Diet Coke and I call it, you guys talk amongst yourselves.

但我叫我要健怡可樂時， 你們自己傳話。

Diet Coke walks up to the front — pick it, put it in the tote, away it goes."

拿可樂的人走到前面， 我拿、放進袋子、出貨。

Wow, what if the products could walk and talk on their own?

哇，假如產品能自己走過來 還能自己傳話？

That's a very interesting, very powerful way

那是非常有意思、非常強大的方法，

that we could potentially organize this warehouse.

以這種方法管理這座倉庫好像可行。

So of course, labor isn't free,

但是當然，勞工不可能免費，

on that practical versus awesome spectrum.

放在現實與夢想的秤上量量就知道。

(Laughter)

（笑）

So we said mobile shelving — We'll put them on mobile shelving.

所以我們想到移動櫃， 我們把產品放在移動櫃上，

We'll use mobile robots and we'll move the inventory around.

我們要用移動式機器人， 還要讓存貨移動。

And so we got underway on that and then I'm sitting on my couch in 2008.

所以我們就開始朝那方面發展。 然後 2008 年有一天我坐在沙發上，

Did any of you see the Beijing Olympics, the opening ceremonies?

你們看過北京奧運的開幕式嗎？

I about fell out of my couch when I saw this.

當我看到這幕， 差點沒從沙發上掉下來。

I'm like, that was the idea!

我就像，就是這樣啦！

(Laughter and Applause)

（笑聲及掌聲）

We'll put thousands of people on the warehouse floor, the stadium floor.

我們要在倉庫裡放上千人， 體育館地板上。

But interestingly enough, this actually relates to the idea

但是夠有意思了， 這其實與另一個點子有關，

in that these guys were creating some incredibly powerful, impressive digital art,

就是這些人創造了極為震撼、 令人讚嘆的數位藝術，

all without computers, I'm told,

卻不用電腦，人家告訴我，

it was all peer-to-peer coordination and communication.

這全靠同儕間的協調與溝通。

You stand up, I'll squat down.

你一站起來我就蹲下去。

And they made some fabulous art.

他們創造了極棒的藝術。

It speaks to the power of emergence

這就要來講講湧現的力量，

in systems when you let things start to talk with each other.

在你讓系統內的東西 開始自己對話後的結果。

So that was a little bit of the journey.

那就是產品開發的過程。

So of course, now what became the practical reality of this idea?

所以當然要看， 這個點子實踐出來是甚麼樣呢？

Here is a warehouse.

這是座倉庫。

It's a pick, pack and ship center that has about 10,000 different SKUs.

這是個取貨、裝貨送貨中心， 大約有一萬種不同的產品。

We'll call them red pens, green pens, yellow Post-It Notes.

我們就說他們是紅筆、 綠筆、黃色的便利貼。

We send the little orange robots out to pick up the blue shelving pods.

我們送這個小橘機器人 去取藍色的儲存盒。

And we deliver them to the side of the building.

然後我們再送他們去建築物邊上。

So all the pick workers now get to stay on the perimeter.

所以所有的取貨員都待在周圍。

And the game here is to pick up the shelves,

玩法是去拿貨架，

take them down the highway and deliver them straight to the pick worker.

從「高速公路」直送到取貨員那裏。

This pick worker's life is completely different.

這位取貨員的日子完全不一樣了。

Rather than wandering around the warehouse, she gets to stay still

她不用在倉庫裡亂走， 她只要待在某地，

in a pick station like this

待在一個像這樣的取貨站，

and every product in the building can now come to her.

建築物內的每個產品會走到她那裏。

So the process is very productive.

所以整個過程非常有生產力。

Reach in, pick an item, scan the bar code, pack it out.

伸進去、拿一個、掃描條碼、裝貨。

By the time you turn around,

一個轉身的時間，

there's another product there ready to be picked and packed.

另一項產品又在旁邊待命 等著取裝。

So what we've done is take out all of the non-value added

所以我們做的是把無附加價值的

walking, searching, wasting, waited time,

走路、搜尋、浪費、等待時間拿掉，

and we've developed a very high-fidelity way to pick these orders,

並發展一套非常準確的方法 來取訂單上的貨，

where you point at it with a laser, scan the UPC barcode,

你用雷射指出貨品， 掃瞄通用產品代碼，

and then indicate with a light which box it needs to go into.

然後以光指示產品該去那個箱子。

So more productive, more accurate and, it turns out,

所以生產力更高，更正確，結果是

it's a more interesting office environment for these pick workers.

這個工作環境對這些 取貨員而言更有趣。

They actually complete the whole order.

他們真的能取完整筆訂單的貨物。

So they do red, green and blue, not just a part of the order.

所以他們拿了紅綠藍， 不是只拿訂單上的部分貨物。

And they feel a little bit more in control of their environment.

他們覺得更能控制他們的環境。

So the side effects of this approach

所以這項方法的副作用

are what really surprised us.

真的嚇壞我們。

We knew it was going to be more productive.

我們早知道這會提高生產力。

But we didn't realize just how pervasive this way of thinking

但我們真的不知道 這個思考方式如何滲透，

extended to other functions in the warehouse.

延伸到倉庫的其他功能上。

But what effectively this approach is doing inside of the DC

但這個方法在配送中心有成效是因為

is turning it into a massively parallel processing engine.

這把它變成一個大規模 平行處理的引擎。

So this is again a cross-fertilization of ideas.

所以這又是一個互相影響的點子。

Here's a warehouse and we're thinking about

這雖是個倉庫，但我們在想的是

parallel processing supercomputer architectures.

平行處理的超級電腦結構。

The notion here is that you have

這裡的概念是你有

10 workers on the right side of the screen

10 名員工在螢幕右邊

that are now all independent autonomous pick workers.

現在都是獨立的自主取貨員。

If the worker in station three decides to leave and go to the bathroom,

如果三號站的人想要離開去上廁所，

it has no impact on the productivity of the other nine workers.

也不會影響到其他九個人的生產力。

Contrast that, for a moment, with the traditional method of using a conveyor.

相反的，如果用傳統的輸送帶，

When one person passes the order to you,

某人傳一個訂單給你，

you put something in and pass it downstream.

你放了某物進去再傳給下一個人。

Everyone has to be in place for that serial process to work.

每一個人都必須在崗位上， 因為那是串列處理的工作。

This becomes a more robust way to think about the warehouse.

這在設計倉庫時是更穩健的方法。

And then underneath the hoods gets interesting in that we're tracking

如果仔細看內部， 很有意思的是我們還能追蹤

the popularity of the products.

產品的受歡迎程度。

And we're using dynamic and adaptive algorithms

我們用動態可調適演算法

to tune the floor of the warehouse.

來調整倉庫樓面。

So what you see here potentially the week leading up to Valentine's Day.

所以你在這看見的是 情人節前一星期的倉儲狀況。

All that pink chalky candy has moved to the front of the building

所有的粉紅色糖都移到建築物的前面，

and is now being picked into a lot of orders in those pick stations.

而且現在已經開始取貨， 因為有大量訂單進入。

Come in two days after Valentine's Day, and that candy, the leftover candy,

情人節後兩天，這些糖，剩下來的糖，

has all drifted to the back of the warehouse

就會移到倉庫後面，

and is occupying the cooler zone on the thermal map there.

存放在溫度圖上比較冷的地方。

One other side effect of this approach using the parallel processing

使用平行處理方法還有一個副作用，

is these things can scale to ginormous.

就是這些東西的規模可以搞得極大。

(Laughter)

（笑聲）

So whether you're doing two pick stations, 20 pick stations,

無論你有 2 個取貨站、 20 個取貨站、

or 200 pick stations, the path planning algorithms

或 200 個取貨站， 路徑計畫演算法

and all of the inventory algorithms just work.

及所有的存貨演算都可應用上。

In this example you see that the inventory

在這個例子你看到存貨

has now occupied all the perimeter of the building

現在都放在建築物四周，

because that's where the pick stations were.

因為取貨站也在那裏。

They sorted it out for themselves.

他們自己會搞清楚該怎麼辦。

So I'll conclude with just one final video

所以我要最後要以一段影片總結。

that shows how this comes to bear

這段影片顯示這個方法如何影響

on the pick worker's actual day in the life of.

一名取貨員的實際工作狀況。

So as we mentioned, the process is to move inventory along the highway

所以就像我們之前提到的， 處理過程是在通道上移動存貨

and then find your way into these pick stations.

並找到去取貨站的路。

And our software in the background

在背後跑的軟體

understands what's going on in each station,

了解每個取貨站的狀況，

we direct the pods across the highway

我們指揮每一個儲存盒的流向，

and we're attempting to get into a queuing system

並試著為之排序，

to present the work to the pick worker.

把貨物依序交給取貨員。

What's interesting is we can even adapt the speed of the pick workers.

有趣的是我們還可以 適應取貨員的速度。

The faster pickers get more pods and the slower pickers get few.

動作快的取貨員排得儲存盒就多， 慢的就排得少。

But this pick worker now is literally having that experience

但這個取貨員現在真的在經歷

that we described before.

我們之前形容的景象。

She puts out her hand. The product jumps into it.

她伸出手，貨物就跑進去。

Or she has to reach in and get it.

好啦，她要伸手進去拿。

She scans it and she puts it in the bucket.

她掃描它，把它放進籃子裡。

And all of the rest of the technology is kind of behind the scenes.

其餘那些科技都在幕後。

So she gets to now focus on the picking and packing portion of her job.

所以她現在能專心 在取貨裝貨這部分。

Never has any idle time, never has to leave her mat.

絕不會閒著沒事， 絕不用離開工作區。

And actually we think not only a more productive

事實上我們認為 這不僅是更有生產力

and more accurate way to fill orders.

準確度更高的方法以屢行訂單。

We think it's a more fulfilling way to fill orders.

我們還認為這是更有成就感的方法。

The reason we can say that, though, is that workers

但是我們之所以這麼說，是因為

in a lot of these buildings now compete

很多在這種建築物內做事的員工

for the privilege of working in the Kiva zone that day.

都爭著要去有奇華系統的區域工作。

And sometimes we'll catch them on testimonial videos

有時候我們還在證言式廣告影片

saying such things as,