Thursday, June 2, 2016

Elon Musk talks Machines of the Future and Model X

After certain owners raised concerns related to the Model X, Elon Musk was spending nights at the factory and personally supervising the production line to iron out some quality control problems plaguing the vehicle. At the shareholder's meeting, he even faulted himself for putting too much technology all at once into one product, "in retrospect the right thing to do with the Model X would have been to take a lot of the really awesome cool things and table them for future versions" Elon said. "This was a case of overconfidence," he added.
According to Musk, the vehicle's software in interpreting the signals from sensors and the operation of the doors has been particularly difficult to refine, "...but I think we're almost there in making the doors useful," he insisted, confirming the next couple of software updates will make the doors "better than normal doors".
During the last couple of months he spent at the Fremont factory, Elon realized that the true problem in accelerating and refining production is building "the machine maker", in other words the factory. Just like designing and building the Model S from scratch, the same approach should be used in building the machines at the factory. "I actually think the potential for improvement in the machine that makes the machine is more than a factor of 10 greater than the potential on the car side".

He then started rationalizing this by doing what he loves best - applying physics' first principles.

"When you think of a manufacturing facility - for a given size of factory - the output is going to be volume times density times velocity. Let's look at our factory and say what is the density of useful to non-useful volume. It’s crazy low. It’s like 2 or 3 percent if you look at it volumetrically – not on a plainer level... Car to non car volumetric ratio, that seems like a room for improvement.

As for velocity. What is a reasonable expectation for the exit velocity of vehicles from the factory. At first you may think that some of the most advanced car factories around the world are very good at making cars and that they may make a car every 25 seconds – that sounds fast, but actually, if you say the length of the car plus some buffer space is approximately 5 meters so it’s taking 25 seconds to move 5 meters. That’s 0.2 meter per second, basically not much faster than a tortoise. That doesn't seem fast."

Elon believes the rate of producing cars can be optimized: "a slow to medium walk  is 1 meter per second, a fast walk is 1.5 m/s, the best car factories in the world are doing 0.2 m/s. You should be able to have cars exit at least walking speed. This shouldn't be so crazy. The density improvement should go from 2-3 per cent to 20 or 30 percent of the volumetric density being optimal."

By comparing the design of a factory to that of a SoC or an advanced computer - where the focus lies on improving clock speeds and data transfer of RAM to SSD or internal CPU cache - the potential of improvement on production is very high. "With significantly less engineering effort, we can make dramatic improvements to the machine that makes the machine," concluding that computer engineers should work on building the factory of the future, "once you explain this to a first grade engineer student, their light bulb goes on... [manufacturers] spend a huge amount of effort trying to get a fraction of a percent of improvement on the product itself, but actually that same amount of effort would yield a greater result if you focus on building the machine that builds the machine. A lot of engineers don't realize that this is possible and they basically think they're operating according to these invisible walls.We just need to explain to them those walls don't exist. It's going to be pretty amazing."

Featured images: Teslarati, Fremont and Gigafactory.

1 comment :

  1. Very interesting to read. I learned lot of information. Thanks for sharing.

    Non woven bag making machine