9 ms·

It’s shocking, that there are millions of people picking various objects from the boxes every day all over the world. I know, robotics is hard. But I would gues

by lnsru 4y ago

It’s shocking, that there are millions of people picking various objects from the boxes every day all over the world. I know, robotics is hard. But I would guess, that is couple orders of magnitude easier than self driving for example. Light is controlled, temperature above freezing point, size of the box known, shape and size with weight as well as weight distribution of the object are known. Why it is not happening?

pharmakom 4y ago
I think the tech will get there. For the shorter term though, imagine a telepresence robot so you can hire workers from anywhere in the world.
- iceTA 4y ago
  This is the plot of Sleep Dealer.
- monkeydust 4y ago
  This is doable today with a robotic arm at least. My friend is working on it here in UK. https://www.extendrobotics.com/ https://www.extendrobotics.com/
ben_w 4y ago
It is happening. The humans now on strike are what's left after most of it was automated.
kmlx 4y ago
https://www.ocadogroup.com/about-us/ocado-technology/ https://www.ocadogroup.com/about-us/ocado-technology/
LatteLazy 4y ago
If a robot cost £1mil (including install) and lasts for 10 years with zero maintenance or other costs, then it's cheaper to employ 4 people on 25k a year to do the same job. And that's without things like seasonal demand changes or government grants for hiring people. 1mil for a robot isn't that much. Does anyone make a robot that can work (at least) as fast as 4 people and never need maintaining etc?
- willcipriano 4y ago
  It's the same economics that made slavery not viable. Sure you don't have to pay labor costs with a slave but if they break their arm you have to call a doctor, otherwise you are out the money you paid for the slave (and that is many years of a typical man's wages), and either way you aren't getting any work until they are healed. It's much more economical to work humans that you've paid a wage, to the bone, and once they are broken, fire them for lack of performance and hire another. No upfront capital outlay, no ongoing costs when they are injured and can't work for you, hell you don't even need to pay them enough to survive, wage slaves beat actual slaves every time. Amazon's problem is they hit such a scale they are starting to run out of wage slaves. They either broke them already or the potential worker has friends or family broken by them so they are wise to the deal. If this wasn't the case they would have little interest in automation.
  - pessimizer 4y ago
    Slavery was economically viable.
    - willcipriano 4y ago
      After the post Civil War recession, economic growth rates increased dramatically[0]. While slavery made individual southerners wealthy, it's abolition resulted in a broader more dynamic economy that took millions out of poverty. [0]https://fee.org/articles/no-slavery-did-not-make-america-rich/ https://fee.org/articles/no-slavery-did-not-make-america-ric...
      - rovolo 4y ago
        > While slavery made individual southerners wealthy, it's abolition [...] took millions out of poverty. But your original argument was about the economics for the enslaver, not for the population. > It's the same economics that made slavery not viable. Sure you don't have to pay labor costs with a slave but if they break their arm you have to call a doctor [...] It's much more economical to work humans that you've paid a wage to the bone [...] wage slaves beat actual slaves every time. You're trying to make a point about capital vs operational expenses, but your example of US slavery's economic viability is both poor and distracting. Enslaving people is generally a financial boon for the enslaver. Focusing on the economic costs of owning your laborers and discounting the lack of wages sounds a lot like the "Positive Good" arguments from the mid 1800s: > Slavery is said to be an evil... But is no evil. On the contrary, I believe it to be the greatest of all the great blessings which a kind Providence has bestowed upon our glorious region... As a class, I say it boldly; there is not a happier, more contented race upon the face of the earth... Lightly tasked, well clothed, well fed—far better than the free laborers of any country in the world, ... their lives and persons protected by the law, all their sufferings alleviated by the kindest and most interested care.... https://en.wikipedia.org/wiki/Slavery_as_a_positive_good_in_the_United_States https://en.wikipedia.org/wiki/Slavery_as_a_positive_good_in_...
philipov 4y ago
There are a lot of places where manual labor is preferred to automation, not because automation is hard, but because automation is expensive, while labor can be found for dirt cheep.
- sct202 4y ago
  Also if you only need labor part of the year like for Christmas for Amazon, if you fully automated Amazon would have to have a bunch of expensive idle robots for the rest of the year. With people you can hire them as seasonal temps for just when you need them.
  - lotsofpulp 4y ago
    I would think that goes the other way, as a benefit to automation. People are a lot more volatile than machines, so scaling up and down in periods of high and low demand is more predictable with machines.
    - zdragnar 4y ago
      Machines are a capital expenditure. You basically can't scale them down below their initial investment cost to acquire. Order picking is, in the grand scheme of human employment, essentially unskilled labor. It takes a bit of practice to get efficient, and a certain level of perseverance and endurance to operate at top efficiency, but it doesn't take any form of higher study, credentialing or apprenticeship to become qualified at. Scaling seasonal workers in this role is relatively trivial in terms of expense compared to purchasing robots. Final note: humans take years to become obsolete and replaced by newer generations. The same isn't true with robots. If you can get full use out of one year round, it is much easier to justify investing now rather than 5 years from now than if you only get two months a year use out of it.
      - datavirtue 4y ago
        How about WFH drone robot operators? You could automate and macro some things and wouldn't need people onsite.
francisofascii 4y ago
"We're cheaper than droids and easier to replace." - Line from Andor, Episode 9.
monkeydust 4y ago
It is happening and Amazon are a big investors in this space afaik (https://amazon.jobs/en/teams/amazon-robotics https://amazon.jobs/en/teams/amazon-robotics). I think the issue is that there are some tasks that might seem easy for a robot to do but they are not, either too expensive vs human capital or the failure rate is not tolerable. The human body is a pretty amazing machine.
onion2k 4y ago
But I would guess, that is couple orders of magnitude easier than self driving for example. You would guess wrong then. Robots don't handle variability well at all. Making a robot that can pack anything like a human can is a huge challenge.
- ape4 4y ago
  The packages I receive seem like they are packed by a simple algorithm. Find the longest dimension of the object, get a box that holds that size. Fill the box with packing material.
  - onion2k 4y ago
    That's not what I mean. Picking a box isn't so hard (although box packing algorithms for non-uniform objects is still tricky). The robotics part is more like "how do I make a robot that can pick up a pen, a 50kg weight, and a silk blouse, and make it put all three in the same package without messing up?" In a factory that would be three distinct robots because that variability in weight, size, surface friction, malleability, etc makes it a hard problem. In a warehouse it would need to be one robot. Even if you compromise and accept additional packaging like putting soft items in boxes or bags it's still quite difficult.
    - ghaff 4y ago
      >Picking a box isn't so hard Though, frankly, Amazon often seems to do a terrible job of optimizing packing--both over and under. I'm guessing that within some range of the right size and some range of not having too much breakage, it's cheaper for them to half-ass it to some degree.
- js8 4y ago
  The solution will eventually be similar to shipping containers. Every producer will have to use one of several hundreds (because of different sizes) standard ways to package something, if they want to sell with Amazon.
detaro 4y ago
Even alone making grippers that reliably can pick up arbitrary objects is surprisingly hard (I believe Amazon did have some prizes offered to anyone who comes up with improvements there). So for now it's cheaper overall to have humans do the tricky part of actually picking stuff up and putting it in boxes, and where possible move other things to/from them. E.g. various "self-driving" shelving systems etc. And Amazon ships all kinds of stuff in various packaging formats etc, so is kind of the worst case here. I could see a company that produces and packages its own stuff choose differently here, and design packaging specifically with its logistics robot capabilities in mind. (For larger packaging sizes, that tend to be large-ish plain boxes, that's already what is happening)
hbgl 4y ago
In the places where automation is not happening, manual labor is still cheaper. The more these people strike though, the sooner they will get replaced by robots.
have_faith 4y ago
We think of ourselves so highly, yet mastery over boxes eludes us.
moonchrome 4y ago
We are fascinated by intelligence/reasoning but if you think about evolution - motion/sensory response were optimized for billions of years, human reasoning is a more recent development - I'm not surprised these things are harder to solve than generating text that appears to be written by another human.
- TremendousJudge 4y ago
  Same thing with chess, but written word is much older. Maybe spoken word is next, since chronologically it's much older. Moving through an environment, self-healing, reproduction, and being powered by pretty much anything that grows are last on the list.
pessimizer 4y ago
Because there's nothing easy about those things, and nothing cheap about building and maintaining a robot that would do them? When you ask the robot to move to another position on the line because they're caught up and the other positions have a backlog, they don't move, and they don't explain why. Also, when they break, you can't yell or threaten them into operating normally again.
blamazon 4y ago
Humans are, in many ways, in many places, the best, least expensive robot. To me the bottleneck is not technology, it's the labor market. Basic human labor planetwide is generally not expensive enough for full automation yet. This is not lost on an HR-heavy company like Amazon that employs over a million humans. Such an equilibrium is a long way away too. We are systemically undervaluing the life energy of our species, en masse.
- ghaff 4y ago
  There are a ton of things that, if labor were arbitrarily expensive, it would be replaced with some combination of: - Robots/automation - Self-service - Poorer service - A lot of people collectively deciding they just just don't need $X if it's going to cost that much Compare the US overall to many places in southeast Asia for example. There is generally a lot more personal service and staffing overall where the labor is cheaper.
  - amalcon 4y ago
    I've heard tell that in some cities with low labor costs, it's cheaper for a disabled person to pay someone to carry them up and down stairs than to rent an apartment with an elevator. That particular robot has been around in more or less its modern form for at least 100 years.
  - nicbou 4y ago
    > A lot of people collectively deciding they just just don't need $X if it's going to cost that much A lot of people decide that it's their God-given right to enjoy a certain service at a certain price, even if it requires exploited labour to achieve. They will actively support legislation that enables this sort of exploitation, because again, this service is too nice to pass on.
  - lawrenceyan 4y ago
    It's interesting to consider how basic income if implemented, could potentially greatly speed up the development/rollout of robotics in industry.
- deleted 4y ago
mabbo 4y ago
tl;dr: Robotic picking requires higher shipping costs, and shipping costs are much higher than the labour costs to pick the item. I spent years in Amazon Fulfillment Tech, a lot of in the Picking software space. It's all about tradeoffs, and shipping costs. Amazon will spend far more money getting the item from the warehouse to your front door than they will on labour costs getting the item into the box, ready to ship to you. Shipping is expensive, labour is relatively cheap when each item is touched by maybe 2-4 people for a few seconds each. Robotics are getting better, but the typical storage bin in an Amazon warehouse is optimized for density, not for easy picking. Humans do a great job of picking items despite the storage density. We fish it out, feel around, use our incredible vision powers. We're great at it! Robots suck at that sort of thing (today). Amazon could build a warehouse that is optimized for robotic picking. But if you built it with today's robots, you would have a fraction of the item storage density. That means you'd either need many times more warehouses or you'd need to reduce your selection. When you reduce selection, that means a different warehouse is handling those other orders, and it's probably further away from the customer. Further away from the customer => higher shipping costs.
- mymythisisthis 4y ago
  Do you think in the future some form of standardized shipping packages will become the norm? Much like the standard metal shipping container became the norm in the trucking industry. That way more automation can be incorporated across the whole shipping chain.
  - mabbo 4y ago
    But that's the same problem: you'd use more storage space per item, needlessly, decreasing storage density.
hourago 4y ago
> It’s shocking, that there are millions of people picking various objects from the boxes every day all over the world. I just read the article and does not talk about automation. Did I miss something?
fortituded0002 4y ago
> Why it is not happening? Amazon has been working on robotics for years and has been incorporating it into their warehouses for a long time. https://www.supplychaindive.com/news/warehouse-robotics-automation-innovations-amazon-showcased-2022/638115/#:~:text=Amazon%20is%20testing%20free%2Droaming,vision%2C%20cover%20this%20task%20instead https://www.supplychaindive.com/news/warehouse-robotics-auto.... Not sure how much I agree with this, but a buddy of mine for years talked about how problematic it was for warehouse workers to fight for higher wages and more support because their jobs would likely be automated away in the near-ish future. His argument was that these jobs are not intended to be long-term career jobs and that by pretending they are, you run the risk of doing people a disservice as they stay longer and they become less employable over time.
- Retric 4y ago
  The key takeaway here is: “Although many are prototypes, the company aims for the bots to eventually make an impact in the company’s logistics network.” They really don’t make significant use of robots.
  - fortituded0002 4y ago
    They do. I've seen it. It's been in the works for years. Each time I've worked in the warehouses things have changed. It's not fully automated, but improvements have been made.
    - Retric 4y ago
      Yet, their number of warehouse employees is still growing like an exponential curve. Compare the number of items shipped per 1,000 warehouse employees in 2012 vs 2022 and no it’s not wildly different.
      - fortituded0002 4y ago
        > Yet, their number of warehouse employees is still growing like an exponential curve. Compare the number of items shipped per 1,000 warehouse employees in 2012 vs 2022 and no it’s not wildly different. No idea where this data is coming from. But there are a lot of factors at play and this is pretty high-level data on what could be going on. It doesn't account for changes in how the company operates, the number of warehouses that are spinning up, how packaging and shipping has evolved (moving between warehouses to consolidate multiple orders into single shipments, etc), the massive spike in orders during the pandemic... But really, What's your goal here? It seems like you just want someone to say you are right instead of discuss? If you wanted to discuss, I'd expect that numbers thrown out would have some sort of backing behind them and displaying curiosity as to why things may be operating a certain way as opposed to already deciding how things work.
oliwarner 4y ago
What I find shocking is that they'll work for a billionaire for £10/h, and let him micro-monitor break time.
- netrus 4y ago
  Tells you something about the alternatives they have.
  - WeylandYutani 4y ago
    Other warehouses, despite what internet pundits will tell you, are much the same. Tech workers in air-conditioned buildings are out of touch on how the underclass live and work. BUT I PICKED TOMATOES FOR A SUMMER JOB WHEN I WAS 15
- rjh29 4y ago
  What choice do you think people have? At minimum wage pretty much all jobs are like that.
  - iso1210 4y ago
    Supermarkets pay more than that and while aren't exactly cushy, they aren't as bad as the reported warehouse conditions of an amazon factory.
dsfyu404ed 4y ago
There are many "lights off" warehouses and subsets of warehouses. Lots of stuff that ships direct from a manufacturer's warehouse is like this. The orders amazon picks and packages are far too diverse for that to be economical and current levels of technological progress.
zeroonetwothree 4y ago
Human hands are an amazing technology that we haven’t been able to artificially replicate. I expect that in the (near?) future when AI has replaced software engineers, lawyers, writers, and bloggers that humans will still be needed to be warehouse workers, fruit pickers, plumbers, and surgeons.
kansface 4y ago
> But I would guess, that is couple orders of magnitude easier than self driving for example. The more I think about this, the more I believe you to be incorrect! Identifying the correct item in the bin is hard - things are in random arrangements and orientations. They may be buried under other stuff. Packaging continually changes. Things will be in the wrong bin. Things will be packaged by the manufacturer incorrectly! Things expire but you can't rely on expiration dates, which you'd have to be able to read and parse in the first place. Anything in the bin could be extremely fragile. Some things are both fragile and heavy. The act of picking stuff up is hard. Grip too hard, you break it. Grip too soft, you drop it and break it or something else. Grip it the wrong way, you break it. Crush stuff around it rummaging, you break that. Some things must be pinched, others grasped, cradled or supported at multiple points. You break something, then what? You've contaminated who knows what around it roomba style - you can't send stuff out to customers thats covered in shards of glass or unknown liquids. How do you identify when something is broken or defective in the first place? Packaging may be damaged while the item is OK, or the item may be broken while the packaging is intact. Humans can do this because they have a model of the world with interactive feedback. Things with broken glass may sound wrong or just feel wrong when moved. The stakes are lower on average, but putting the wrong item in a cart will eventually kill someone. Consider allergies, medical supplies, the blind or illiterate. All the while, you have to do this for less than $15/hour? I can only see robots currently succeeding for hyper specific items - perhaps those that are low value and heavy for instance.

by lnsru 4y ago

pharmakom 4y ago

I think the tech will get there. For the shorter term though, imagine a telepresence robot so you can hire workers from anywhere in the world.

iceTA 4y ago
This is the plot of Sleep Dealer.
monkeydust 4y ago
This is doable today with a robotic arm at least. My friend is working on it here in UK. https://www.extendrobotics.com/ https://www.extendrobotics.com/

ben_w 4y ago

It is happening. The humans now on strike are what's left after most of it was automated.

kmlx 4y ago

https://www.ocadogroup.com/about-us/ocado-technology/ https://www.ocadogroup.com/about-us/ocado-technology/

LatteLazy 4y ago

If a robot cost £1mil (including install) and lasts for 10 years with zero maintenance or other costs, then it's cheaper to employ 4 people on 25k a year to do the same job. And that's without things like seasonal demand changes or government grants for hiring people. 1mil for a robot isn't that much. Does anyone make a robot that can work (at least) as fast as 4 people and never need maintaining etc?

willcipriano 4y ago
It's the same economics that made slavery not viable. Sure you don't have to pay labor costs with a slave but if they break their arm you have to call a doctor, otherwise you are out the money you paid for the slave (and that is many years of a typical man's wages), and either way you aren't getting any work until they are healed. It's much more economical to work humans that you've paid a wage, to the bone, and once they are broken, fire them for lack of performance and hire another. No upfront capital outlay, no ongoing costs when they are injured and can't work for you, hell you don't even need to pay them enough to survive, wage slaves beat actual slaves every time. Amazon's problem is they hit such a scale they are starting to run out of wage slaves. They either broke them already or the potential worker has friends or family broken by them so they are wise to the deal. If this wasn't the case they would have little interest in automation.
- pessimizer 4y ago
  Slavery was economically viable.
  - willcipriano 4y ago
    After the post Civil War recession, economic growth rates increased dramatically[0]. While slavery made individual southerners wealthy, it's abolition resulted in a broader more dynamic economy that took millions out of poverty. [0]https://fee.org/articles/no-slavery-did-not-make-america-rich/ https://fee.org/articles/no-slavery-did-not-make-america-ric...
    - rovolo 4y ago
      > While slavery made individual southerners wealthy, it's abolition [...] took millions out of poverty. But your original argument was about the economics for the enslaver, not for the population. > It's the same economics that made slavery not viable. Sure you don't have to pay labor costs with a slave but if they break their arm you have to call a doctor [...] It's much more economical to work humans that you've paid a wage to the bone [...] wage slaves beat actual slaves every time. You're trying to make a point about capital vs operational expenses, but your example of US slavery's economic viability is both poor and distracting. Enslaving people is generally a financial boon for the enslaver. Focusing on the economic costs of owning your laborers and discounting the lack of wages sounds a lot like the "Positive Good" arguments from the mid 1800s: > Slavery is said to be an evil... But is no evil. On the contrary, I believe it to be the greatest of all the great blessings which a kind Providence has bestowed upon our glorious region... As a class, I say it boldly; there is not a happier, more contented race upon the face of the earth... Lightly tasked, well clothed, well fed—far better than the free laborers of any country in the world, ... their lives and persons protected by the law, all their sufferings alleviated by the kindest and most interested care.... https://en.wikipedia.org/wiki/Slavery_as_a_positive_good_in_the_United_States https://en.wikipedia.org/wiki/Slavery_as_a_positive_good_in_...

philipov 4y ago

There are a lot of places where manual labor is preferred to automation, not because automation is hard, but because automation is expensive, while labor can be found for dirt cheep.

sct202 4y ago
Also if you only need labor part of the year like for Christmas for Amazon, if you fully automated Amazon would have to have a bunch of expensive idle robots for the rest of the year. With people you can hire them as seasonal temps for just when you need them.
- lotsofpulp 4y ago
  I would think that goes the other way, as a benefit to automation. People are a lot more volatile than machines, so scaling up and down in periods of high and low demand is more predictable with machines.
  - zdragnar 4y ago
    Machines are a capital expenditure. You basically can't scale them down below their initial investment cost to acquire. Order picking is, in the grand scheme of human employment, essentially unskilled labor. It takes a bit of practice to get efficient, and a certain level of perseverance and endurance to operate at top efficiency, but it doesn't take any form of higher study, credentialing or apprenticeship to become qualified at. Scaling seasonal workers in this role is relatively trivial in terms of expense compared to purchasing robots. Final note: humans take years to become obsolete and replaced by newer generations. The same isn't true with robots. If you can get full use out of one year round, it is much easier to justify investing now rather than 5 years from now than if you only get two months a year use out of it.
    - datavirtue 4y ago
      How about WFH drone robot operators? You could automate and macro some things and wouldn't need people onsite.

francisofascii 4y ago

"We're cheaper than droids and easier to replace." - Line from Andor, Episode 9.

monkeydust 4y ago

It is happening and Amazon are a big investors in this space afaik (https://amazon.jobs/en/teams/amazon-robotics https://amazon.jobs/en/teams/amazon-robotics). I think the issue is that there are some tasks that might seem easy for a robot to do but they are not, either too expensive vs human capital or the failure rate is not tolerable. The human body is a pretty amazing machine.

onion2k 4y ago

But I would guess, that is couple orders of magnitude easier than self driving for example. You would guess wrong then. Robots don't handle variability well at all. Making a robot that can pack anything like a human can is a huge challenge.

ape4 4y ago
The packages I receive seem like they are packed by a simple algorithm. Find the longest dimension of the object, get a box that holds that size. Fill the box with packing material.
- onion2k 4y ago
  That's not what I mean. Picking a box isn't so hard (although box packing algorithms for non-uniform objects is still tricky). The robotics part is more like "how do I make a robot that can pick up a pen, a 50kg weight, and a silk blouse, and make it put all three in the same package without messing up?" In a factory that would be three distinct robots because that variability in weight, size, surface friction, malleability, etc makes it a hard problem. In a warehouse it would need to be one robot. Even if you compromise and accept additional packaging like putting soft items in boxes or bags it's still quite difficult.
  - ghaff 4y ago
    >Picking a box isn't so hard Though, frankly, Amazon often seems to do a terrible job of optimizing packing--both over and under. I'm guessing that within some range of the right size and some range of not having too much breakage, it's cheaper for them to half-ass it to some degree.
js8 4y ago
The solution will eventually be similar to shipping containers. Every producer will have to use one of several hundreds (because of different sizes) standard ways to package something, if they want to sell with Amazon.

detaro 4y ago

Even alone making grippers that reliably can pick up arbitrary objects is surprisingly hard (I believe Amazon did have some prizes offered to anyone who comes up with improvements there). So for now it's cheaper overall to have humans do the tricky part of actually picking stuff up and putting it in boxes, and where possible move other things to/from them. E.g. various "self-driving" shelving systems etc. And Amazon ships all kinds of stuff in various packaging formats etc, so is kind of the worst case here. I could see a company that produces and packages its own stuff choose differently here, and design packaging specifically with its logistics robot capabilities in mind. (For larger packaging sizes, that tend to be large-ish plain boxes, that's already what is happening)

hbgl 4y ago

In the places where automation is not happening, manual labor is still cheaper. The more these people strike though, the sooner they will get replaced by robots.

have_faith 4y ago

We think of ourselves so highly, yet mastery over boxes eludes us.

moonchrome 4y ago

We are fascinated by intelligence/reasoning but if you think about evolution - motion/sensory response were optimized for billions of years, human reasoning is a more recent development - I'm not surprised these things are harder to solve than generating text that appears to be written by another human.

TremendousJudge 4y ago
Same thing with chess, but written word is much older. Maybe spoken word is next, since chronologically it's much older. Moving through an environment, self-healing, reproduction, and being powered by pretty much anything that grows are last on the list.

pessimizer 4y ago

Because there's nothing easy about those things, and nothing cheap about building and maintaining a robot that would do them? When you ask the robot to move to another position on the line because they're caught up and the other positions have a backlog, they don't move, and they don't explain why. Also, when they break, you can't yell or threaten them into operating normally again.

blamazon 4y ago

Humans are, in many ways, in many places, the best, least expensive robot. To me the bottleneck is not technology, it's the labor market. Basic human labor planetwide is generally not expensive enough for full automation yet. This is not lost on an HR-heavy company like Amazon that employs over a million humans. Such an equilibrium is a long way away too. We are systemically undervaluing the life energy of our species, en masse.

ghaff 4y ago
There are a ton of things that, if labor were arbitrarily expensive, it would be replaced with some combination of: - Robots/automation - Self-service - Poorer service - A lot of people collectively deciding they just just don't need $X if it's going to cost that much Compare the US overall to many places in southeast Asia for example. There is generally a lot more personal service and staffing overall where the labor is cheaper.
- amalcon 4y ago
  I've heard tell that in some cities with low labor costs, it's cheaper for a disabled person to pay someone to carry them up and down stairs than to rent an apartment with an elevator. That particular robot has been around in more or less its modern form for at least 100 years.
- nicbou 4y ago
  > A lot of people collectively deciding they just just don't need $X if it's going to cost that much A lot of people decide that it's their God-given right to enjoy a certain service at a certain price, even if it requires exploited labour to achieve. They will actively support legislation that enables this sort of exploitation, because again, this service is too nice to pass on.
- lawrenceyan 4y ago
  It's interesting to consider how basic income if implemented, could potentially greatly speed up the development/rollout of robotics in industry.
deleted 4y ago

mabbo 4y ago

tl;dr: Robotic picking requires higher shipping costs, and shipping costs are much higher than the labour costs to pick the item. I spent years in Amazon Fulfillment Tech, a lot of in the Picking software space. It's all about tradeoffs, and shipping costs. Amazon will spend far more money getting the item from the warehouse to your front door than they will on labour costs getting the item into the box, ready to ship to you. Shipping is expensive, labour is relatively cheap when each item is touched by maybe 2-4 people for a few seconds each. Robotics are getting better, but the typical storage bin in an Amazon warehouse is optimized for density, not for easy picking. Humans do a great job of picking items despite the storage density. We fish it out, feel around, use our incredible vision powers. We're great at it! Robots suck at that sort of thing (today). Amazon could build a warehouse that is optimized for robotic picking. But if you built it with today's robots, you would have a fraction of the item storage density. That means you'd either need many times more warehouses or you'd need to reduce your selection. When you reduce selection, that means a different warehouse is handling those other orders, and it's probably further away from the customer. Further away from the customer => higher shipping costs.

mymythisisthis 4y ago
Do you think in the future some form of standardized shipping packages will become the norm? Much like the standard metal shipping container became the norm in the trucking industry. That way more automation can be incorporated across the whole shipping chain.
- mabbo 4y ago
  But that's the same problem: you'd use more storage space per item, needlessly, decreasing storage density.

hourago 4y ago

> It’s shocking, that there are millions of people picking various objects from the boxes every day all over the world. I just read the article and does not talk about automation. Did I miss something?

fortituded0002 4y ago

> Why it is not happening? Amazon has been working on robotics for years and has been incorporating it into their warehouses for a long time. https://www.supplychaindive.com/news/warehouse-robotics-automation-innovations-amazon-showcased-2022/638115/#:~:text=Amazon%20is%20testing%20free%2Droaming,vision%2C%20cover%20this%20task%20instead https://www.supplychaindive.com/news/warehouse-robotics-auto.... Not sure how much I agree with this, but a buddy of mine for years talked about how problematic it was for warehouse workers to fight for higher wages and more support because their jobs would likely be automated away in the near-ish future. His argument was that these jobs are not intended to be long-term career jobs and that by pretending they are, you run the risk of doing people a disservice as they stay longer and they become less employable over time.

Retric 4y ago
The key takeaway here is: “Although many are prototypes, the company aims for the bots to eventually make an impact in the company’s logistics network.” They really don’t make significant use of robots.
- fortituded0002 4y ago
  They do. I've seen it. It's been in the works for years. Each time I've worked in the warehouses things have changed. It's not fully automated, but improvements have been made.
  - Retric 4y ago
    Yet, their number of warehouse employees is still growing like an exponential curve. Compare the number of items shipped per 1,000 warehouse employees in 2012 vs 2022 and no it’s not wildly different.
    - fortituded0002 4y ago
      > Yet, their number of warehouse employees is still growing like an exponential curve. Compare the number of items shipped per 1,000 warehouse employees in 2012 vs 2022 and no it’s not wildly different. No idea where this data is coming from. But there are a lot of factors at play and this is pretty high-level data on what could be going on. It doesn't account for changes in how the company operates, the number of warehouses that are spinning up, how packaging and shipping has evolved (moving between warehouses to consolidate multiple orders into single shipments, etc), the massive spike in orders during the pandemic... But really, What's your goal here? It seems like you just want someone to say you are right instead of discuss? If you wanted to discuss, I'd expect that numbers thrown out would have some sort of backing behind them and displaying curiosity as to why things may be operating a certain way as opposed to already deciding how things work.

oliwarner 4y ago

What I find shocking is that they'll work for a billionaire for £10/h, and let him micro-monitor break time.

netrus 4y ago
Tells you something about the alternatives they have.
- WeylandYutani 4y ago
  Other warehouses, despite what internet pundits will tell you, are much the same. Tech workers in air-conditioned buildings are out of touch on how the underclass live and work. BUT I PICKED TOMATOES FOR A SUMMER JOB WHEN I WAS 15
rjh29 4y ago
What choice do you think people have? At minimum wage pretty much all jobs are like that.
- iso1210 4y ago
  Supermarkets pay more than that and while aren't exactly cushy, they aren't as bad as the reported warehouse conditions of an amazon factory.

dsfyu404ed 4y ago

There are many "lights off" warehouses and subsets of warehouses. Lots of stuff that ships direct from a manufacturer's warehouse is like this. The orders amazon picks and packages are far too diverse for that to be economical and current levels of technological progress.

zeroonetwothree 4y ago

Human hands are an amazing technology that we haven’t been able to artificially replicate. I expect that in the (near?) future when AI has replaced software engineers, lawyers, writers, and bloggers that humans will still be needed to be warehouse workers, fruit pickers, plumbers, and surgeons.

kansface 4y ago

> But I would guess, that is couple orders of magnitude easier than self driving for example. The more I think about this, the more I believe you to be incorrect! Identifying the correct item in the bin is hard - things are in random arrangements and orientations. They may be buried under other stuff. Packaging continually changes. Things will be in the wrong bin. Things will be packaged by the manufacturer incorrectly! Things expire but you can't rely on expiration dates, which you'd have to be able to read and parse in the first place. Anything in the bin could be extremely fragile. Some things are both fragile and heavy. The act of picking stuff up is hard. Grip too hard, you break it. Grip too soft, you drop it and break it or something else. Grip it the wrong way, you break it. Crush stuff around it rummaging, you break that. Some things must be pinched, others grasped, cradled or supported at multiple points. You break something, then what? You've contaminated who knows what around it roomba style - you can't send stuff out to customers thats covered in shards of glass or unknown liquids. How do you identify when something is broken or defective in the first place? Packaging may be damaged while the item is OK, or the item may be broken while the packaging is intact. Humans can do this because they have a model of the world with interactive feedback. Things with broken glass may sound wrong or just feel wrong when moved. The stakes are lower on average, but putting the wrong item in a cart will eventually kill someone. Consider allergies, medical supplies, the blind or illiterate. All the while, you have to do this for less than $15/hour? I can only see robots currently succeeding for hyper specific items - perhaps those that are low value and heavy for instance.