5 ms·

The Robots Are The Chefs In This Japanese Restaurant

18 points by kkleiner 17y ago

ivankirigin 17y ago
Chefs are the last humans I'd automate in a fine restaurant.
- antonovka 17y ago
  You can not replace Omakase with robots, but I would gladly have my day-to-day food requirements met by robotic cooks that: - Allow me to arbitrarily adjust the variables involved in making my food, and thus discover the perfect bowl of noodles, the perfect burger, etc. - Reduce service costs to the minimum, allowing restaurants to further compete on ingredient quality and other factors. - Never burn my toast.
  - ivankirigin 17y ago
    Robots already make your toast, and consistently get it wrong. Robots for the next decade will certainly increase costs of chefs. You know short order cooks at most restaurants get paid close to nothing right? 10/hr = 20K/yr = 10% the cost of today's robots, and probably 2% factoring maintenance. And thats just for the hardware. The problems we're talking about require better hardware (for perception and dexterous hand) and unavailable software. Lower on the spectrum of dining, the "cook" is actually a microwave. People don't want to know how much food is just reheated. Ever eaten a "hot continental breakfast" at a hotel? Biscuits and gravy, bacon, scrambled eggs, sausage, etc. - all reheated from frozen. Disgusting. Finally, it's pretty arrogant to assume the perfect burger is just the right combination of ingredients and permutation of steps. The perfect burger is 100% an editorial creation.
    - antonovka 17y ago
      Robots already make your toast, and consistently get it wrong. Only because toasters require new instruction for every piece of toast, and I, like most everyone else, consistently give them the wrong instructions -- I place the toast in the wrong place, I set the toaster too hot, I set it to run too long, and I never perform sufficient experimentation to determine what the right variables are for a given type, size, and quantity of bread. Robots for the next decade will certainly increase costs of chefs. You know short order cooks at most restaurants get paid close to nothing right? 10/hr = 20K/yr = 10% the cost of today's robots, and probably 2% factoring maintenance. And thats just for the hardware. The problems we're talking about require better hardware (for perception and dexterous hand) and unavailable software. We have an long industrial history of steadily decreasing costs, steadily increasing power and features, and producing hardware and writing software that was hitherto unavailable. The question isn't whether we can do it, but whether we'll invest the resources necessary to do it. Lower on the spectrum of dining, the "cook" is actually a microwave. People don't want to know how much food is just reheated. Ever eaten a "hot continental breakfast" at a hotel? Biscuits and gravy, bacon, scrambled eggs, sausage, etc. - all reheated from frozen. Disgusting. People have varied tastes and acclimations. What does that have to do with a Robot Chef? Finally, it's pretty arrogant to assume the perfect burger is just the right combination of ingredients and permutation of steps. The perfect burger is 100% an editorial creation. Isn't cooking the perfect burger exactly a question of ingredients and permutations of steps adjusted for those ingredients and personal preference --- regardless of whether it's a robot or a human doing the cooking?
    - whatusername 17y ago
      "Robots already make your toast, and consistently get it wrong." Not really. Any large group toaster (the ones that look like a pizza cooker) with the conveyor belt produces pretty consistent toast pretty quickly. To use an old software excuse - the toaster is cooking toast exactly to the spec. Just because you put it down for too long - doesn't make it the toaster/robots fault. // agree on almost all else though
    - rsheridan6 17y ago
      >Robots for the next decade will certainly increase costs of chefs. You know short order cooks at most restaurants get paid close to nothing right? 10/hr = 20K/yr = 10% the cost of today's robots, and probably 2% factoring maintenance. Robots don't just work 40 hour weeks. They'll work as many hours as the store is open. They also don't require benefits, the employer doesn't have to pay unemployment insurance, and they don't have to pay social security taxes. In addition, there are harder to quantify benefits like the fact that robots don't call in sick, call in drunk, etc, and they'll probably be less likely to do things like drop or burn food. When you factor in all of that, a better estimate of the cost of a human compared to the cost of a robot would be more like 25%. The robot starts looking better. And the price of technology tends to come down over time.
      - ivankirigin 17y ago
        Asimo has a 15 minutes battery life, and batteries are improving at sub-linear rates. Tethering solves this, or some kind of ICE, but neither are useful everywhere. People require a pizza day, which is amazing. You're right about the price of technology, but the timeframe is what I'm talking about. I think the "robot revolution" is inevitable. But I still like to poke holes in theories that it is around the corner, but it isn't. Cheap restaurants and other cheap labor intensive activities are the _last_ things that will be automated. Your manager job will go first, along with your highly paid pilots, etc.
oldgregg 17y ago
This is a little gimmicky, but restaurant automation is fascinating -- payroll is still a massive expense for most restaurant. For fine dining, forget it. But for fast food I'm surprised more hasn't been done in this space. I know there have been a few attempts here and there, but the closest thing at scale is some of the crazy vending machines in japan. Fast food already seems so systematic that I'm kind of surprised that automation hasn't made much progress. It seems like automation costs are the prohibitive factor, but I just don't understand why. I feel like a concept with these constraints could work: a) vending machine with minimal real estate / high traffic b) a very limited menu where the food has been chosen based on ease of automation c) only uses products from sysco You still have upfront engineering costs, but assuming you can get food quality above McD's you'll be fine. I also wonder how much chefs have collaborated with engineers in previous attempts. For instance, if McDonalds built a machine it would HAVE to make a BigMac -- now you've already introduced bad constraints. It's also likely that the natural innovator, manufacturers of food service machinery, have strong financial incentives NOT to innovate. I'm mostly full of crap, but it is interesting. :)
biohacker42 17y ago
This makes me think of a recent Economist article about the industrial revolution. The question is why did the industrial revolution happen in the UK? Lots of other places had innovation earlier, had banking earlier, etc. But one theory is that the UK at that point in time uniquely had expensive labor combined with cheap energy. Then and only then did industrialization make sense, because only then was it economical - profitable. So I'm wondering when will advanced robotics, capable of completing complex tasks in the service industry be more economical then minimum wage humans? Sadly I don't think this is going to happen soon, human labor is cheap and industrial robots are expensive.

18 points by kkleiner 17y ago

ivankirigin 17y ago

Chefs are the last humans I'd automate in a fine restaurant.

antonovka 17y ago
You can not replace Omakase with robots, but I would gladly have my day-to-day food requirements met by robotic cooks that: - Allow me to arbitrarily adjust the variables involved in making my food, and thus discover the perfect bowl of noodles, the perfect burger, etc. - Reduce service costs to the minimum, allowing restaurants to further compete on ingredient quality and other factors. - Never burn my toast.
- ivankirigin 17y ago
  Robots already make your toast, and consistently get it wrong. Robots for the next decade will certainly increase costs of chefs. You know short order cooks at most restaurants get paid close to nothing right? 10/hr = 20K/yr = 10% the cost of today's robots, and probably 2% factoring maintenance. And thats just for the hardware. The problems we're talking about require better hardware (for perception and dexterous hand) and unavailable software. Lower on the spectrum of dining, the "cook" is actually a microwave. People don't want to know how much food is just reheated. Ever eaten a "hot continental breakfast" at a hotel? Biscuits and gravy, bacon, scrambled eggs, sausage, etc. - all reheated from frozen. Disgusting. Finally, it's pretty arrogant to assume the perfect burger is just the right combination of ingredients and permutation of steps. The perfect burger is 100% an editorial creation.
  - antonovka 17y ago
    Robots already make your toast, and consistently get it wrong. Only because toasters require new instruction for every piece of toast, and I, like most everyone else, consistently give them the wrong instructions -- I place the toast in the wrong place, I set the toaster too hot, I set it to run too long, and I never perform sufficient experimentation to determine what the right variables are for a given type, size, and quantity of bread. Robots for the next decade will certainly increase costs of chefs. You know short order cooks at most restaurants get paid close to nothing right? 10/hr = 20K/yr = 10% the cost of today's robots, and probably 2% factoring maintenance. And thats just for the hardware. The problems we're talking about require better hardware (for perception and dexterous hand) and unavailable software. We have an long industrial history of steadily decreasing costs, steadily increasing power and features, and producing hardware and writing software that was hitherto unavailable. The question isn't whether we can do it, but whether we'll invest the resources necessary to do it. Lower on the spectrum of dining, the "cook" is actually a microwave. People don't want to know how much food is just reheated. Ever eaten a "hot continental breakfast" at a hotel? Biscuits and gravy, bacon, scrambled eggs, sausage, etc. - all reheated from frozen. Disgusting. People have varied tastes and acclimations. What does that have to do with a Robot Chef? Finally, it's pretty arrogant to assume the perfect burger is just the right combination of ingredients and permutation of steps. The perfect burger is 100% an editorial creation. Isn't cooking the perfect burger exactly a question of ingredients and permutations of steps adjusted for those ingredients and personal preference --- regardless of whether it's a robot or a human doing the cooking?
  - whatusername 17y ago
    "Robots already make your toast, and consistently get it wrong." Not really. Any large group toaster (the ones that look like a pizza cooker) with the conveyor belt produces pretty consistent toast pretty quickly. To use an old software excuse - the toaster is cooking toast exactly to the spec. Just because you put it down for too long - doesn't make it the toaster/robots fault. // agree on almost all else though
  - rsheridan6 17y ago
    >Robots for the next decade will certainly increase costs of chefs. You know short order cooks at most restaurants get paid close to nothing right? 10/hr = 20K/yr = 10% the cost of today's robots, and probably 2% factoring maintenance. Robots don't just work 40 hour weeks. They'll work as many hours as the store is open. They also don't require benefits, the employer doesn't have to pay unemployment insurance, and they don't have to pay social security taxes. In addition, there are harder to quantify benefits like the fact that robots don't call in sick, call in drunk, etc, and they'll probably be less likely to do things like drop or burn food. When you factor in all of that, a better estimate of the cost of a human compared to the cost of a robot would be more like 25%. The robot starts looking better. And the price of technology tends to come down over time.
    - ivankirigin 17y ago
      Asimo has a 15 minutes battery life, and batteries are improving at sub-linear rates. Tethering solves this, or some kind of ICE, but neither are useful everywhere. People require a pizza day, which is amazing. You're right about the price of technology, but the timeframe is what I'm talking about. I think the "robot revolution" is inevitable. But I still like to poke holes in theories that it is around the corner, but it isn't. Cheap restaurants and other cheap labor intensive activities are the _last_ things that will be automated. Your manager job will go first, along with your highly paid pilots, etc.

oldgregg 17y ago

This is a little gimmicky, but restaurant automation is fascinating -- payroll is still a massive expense for most restaurant. For fine dining, forget it. But for fast food I'm surprised more hasn't been done in this space. I know there have been a few attempts here and there, but the closest thing at scale is some of the crazy vending machines in japan. Fast food already seems so systematic that I'm kind of surprised that automation hasn't made much progress. It seems like automation costs are the prohibitive factor, but I just don't understand why. I feel like a concept with these constraints could work: a) vending machine with minimal real estate / high traffic b) a very limited menu where the food has been chosen based on ease of automation c) only uses products from sysco You still have upfront engineering costs, but assuming you can get food quality above McD's you'll be fine. I also wonder how much chefs have collaborated with engineers in previous attempts. For instance, if McDonalds built a machine it would HAVE to make a BigMac -- now you've already introduced bad constraints. It's also likely that the natural innovator, manufacturers of food service machinery, have strong financial incentives NOT to innovate. I'm mostly full of crap, but it is interesting. :)

biohacker42 17y ago

This makes me think of a recent Economist article about the industrial revolution. The question is why did the industrial revolution happen in the UK? Lots of other places had innovation earlier, had banking earlier, etc. But one theory is that the UK at that point in time uniquely had expensive labor combined with cheap energy. Then and only then did industrialization make sense, because only then was it economical - profitable. So I'm wondering when will advanced robotics, capable of completing complex tasks in the service industry be more economical then minimum wage humans? Sadly I don't think this is going to happen soon, human labor is cheap and industrial robots are expensive.