Just announced. 500k cars by 2018 instead of 2020

[N5329K]

Every manufacturer has to define for itself what's "good enough." As buyers, we want 100% on time and on quality. A company that insists on 100% will spend its time, energy and resources in refining and inspecting, not delivering. A company that doesn't care just counts units and forgets about good enough. British Leyland in the 1970's comes to mind. And when GM ran NUMMI (what's now Tesla Fremont), that was their approach, too. We see how that ended.
Tesla has to find the right balance between quality and delivery. Too many units at too low quality and things come crashing down. Too few units at too high a quality works for fine art or Swiss watches, but not cars (unless you're making them for a tiny, tiny audience). Naturally, tiny audiences don't change history, and that's what Mr. Musk intends to do.
It's no wonder a very young company like Tesla is still hunting for that sweet spot.
Robin

 

[JeffK]

Every manufacturer has to define for itself what's "good enough." ...
It's no wonder a very young company like Tesla is still hunting for that sweet spot.
Robin

Their sweet spot is a moving target. With production lines it's entirely possible to increase throughput and quality at the same time. I don't think Tesla is going to settle for "good enough" in a culture of continual improvement.

 

[N5329K]

It's definitely possible. For Tesla to survive, it's also mandatory.
Robin

 

[Jleafs]

My point is this: The robots that make Tesla MS's have been programed for an eternity ( 4 years or so ). In other words...the "How to make an MS" question has been answered. There is no more design, nor programing robots necessary. I don't understand how or why they just can't ramp up production.

"How to make an MS", i.e. a singular vehicle, is a totally different question than how to make 1,000 or 10,000 or 1,000,000. It's not simply a matter of robots that need to be programmed, it's a complex process of supply management, financing, human resources, logistics, quality control, etc. etc.

Tesla obviously did not start out of the gate 4 years ago with a fully formed, perfected, and scaleable production process and they're not there yet 4 years later, either. Of course it's hard, if it was easy, then anybody could've done it by now.

 

[Garlan Garner]

This is just your opinion, you can change how it's done, just move the robot, optimize timing etc ( my job it's been this for a some years, just non in automovie industries ) and i've seend line doubling the production whitout big change, and with big change triple the production.
Since Elon stated that he see many point where he can improve thing, i would think that this is the case.
And we are talking about a huge line, with many robot ( and robot can be greatly customized and replaced for improve things, unlike other machine ), this at robot as people, you can always change you mind of how it's better to build the car, you can double the robot in the same line, just think like "one robot does the left side, the other the right side" while before there was only one robot, or you can place the seat while mounting the brake, things like that could greatly improve the overall line speed

Then just by changing something you can greatly reduce the "fault" where a car came out and has to be fixed ( this reduce the line speed ) and again you are improving the production.. i remember one time when simply dumping bad nails we got a +30% production.

If you can't conceive how this could be archived simply means that you havent put you hands on production line.

Its not my opinion that they have been making the MS for years now. Its a fact
Its not my opinion that the standard design of the MS is done. Its a fact
Its not my opinion that there is nothing more to engineer. Its done. Its a fact.

All Tesla has to do now is repeat success. The MS is a success. If robots are the predominant thing putting the MS together....then there has to be another issue. ( lack of parts/ paint/ slow delivery/ not enough robots...etc. )

If Tesla is running robots 24/7 ( production ) , then I don't know why they are running into delivery issues. It has to be a hang up after production.

 

[Garlan Garner]

Their sweet spot is a moving target. With production lines it's entirely possible to increase throughput and quality at the same time. I don't think Tesla is going to settle for "good enough" in a culture of continual improvement.

I"m not sure its a moving target. They haven't caught the target yet. They are vastly behind the target - behind their own target. It affects the stock price.

In other words, Tesla knows that if it makes 2000 MS's next week....they will still be behind. They know what the orders are...however they can't fulfill them. The question I have is why? Does anyone know....or are we going to keep speculation about what could possibly be the problem?

 

[Garlan Garner]

"How to make an MS", i.e. a singular vehicle, is a totally different question than how to make 1,000 or 10,000 or 1,000,000. It's not simply a matter of robots that need to be programmed, it's a complex process of supply management, financing, human resources, logistics, quality control, etc. etc.

Tesla obviously did not start out of the gate 4 years ago with a fully formed, perfected, and scaleable production process and they're not there yet 4 years later, either. Of course it's hard, if it was easy, then anybody could've done it by now.

Thanks. After reading the following article, the problem is not production. The robots must be working well.

Looks like the problem is delivery. In other words...it sounds like Tesla has cars sitting somewhere, however have a problem getting them delivered on time. This is the first paragraph of the article.

Tesla issued its Q2 production and delivery numbers, indicating that the company produced 20 percent more cars this quarter than the previous however fell short in delivering the vehicles to customers.

Can anyone confirm the facts? Q2, 18,345 cars produced - only 14,370 delivered.
4000 cars must be sitting in a lot somewhere.

Tesla increases Q2 production by 20% but falls short of deliveries

 

[EinSV]

Thanks. After reading the following article, the problem is not production. The robots must be working well.

Looks like the problem is delivery. In other words...it sounds like Tesla has cars sitting somewhere, however have a problem getting them delivered on time. This is the first paragraph of the article.

Tesla issued its Q2 production and delivery numbers, indicating that the company produced 20 percent more cars this quarter than the previous however fell short in delivering the vehicles to customers.

Can anyone confirm the facts? Q2, 18,345 cars produced - only 14,370 delivered.
4000 cars must be sitting in a lot somewhere.

Tesla increases Q2 production by 20% but falls short of deliveries

Hi Garlan,

Tesla produced 18,345 vehicles in Q2 and 15,510 in Q1, which is about an 18.2 percent increase in production. I expect the reason deliveries lagged is that Tesla reports that almost half of its production was in June, so many of those vehicles were still in transit on June 30, including overseas and to the East Coast.

 

[Garlan Garner]

Hi Garlan,

Tesla produced 18,345 vehicles in Q2 and 15,510 in Q1, which is about an 18.2 percent increase in production. I expect the reason deliveries lagged is that Tesla reports that almost half of its production was in June, so many of those vehicles were still in transit on June 30, including overseas and to the East Coast.

I agree, but the stockholders don't care where the cars are. They missed the deliveries target. Tesla promised 17,000 deliveries to it's investors.
They made the production target, however they can't deliver.... I now wonder why. Maybe its the overseas car deliveries..... I don't know. As a regular person....I don't care, however as a stock holder...I do care.

 

[EinSV]

I agree, but the stockholders don't care where the cars are. They missed the deliveries target. Tesla promised deliveries to it's investors.

I was just responding to your question and to your statement that undelivered cars might be sitting on a lot somewhere. Some of them are likely in transit, and others may now be on a lot waiting to be picked up by customers.

And I agree that some investors may focus on delivery numbers rather than production numbers -- I guess we will find out soon enough.

 

[Garlan Garner]

I was just responding to your question and to your statement that undelivered cars might be sitting on a lot somewhere. Some of them are likely in transit, and others may now be on a lot waiting to be picked up by customers.

And I agree that some investors may focus on delivery numbers rather than production numbers -- I guess we will find out soon enough.

Yep!

 

[cronosx]

Its not my opinion that they have been making the MS for years now. Its a fact
Its not my opinion that the standard design of the MS is done. Its a fact
Its not my opinion that there is nothing more to engineer. Its done. Its a fact.

All Tesla has to do now is repeat success. The MS is a success. If robots are the predominant thing putting the MS together....then there has to be another issue. ( lack of parts/ paint/ slow delivery/ not enough robots...etc. )

If Tesla is running robots 24/7 ( production ) , then I don't know why they are running into delivery issues. It has to be a hang up after production.

Since it seems you don't understanrd..just put in the pratical matter, if you have to make 5 holes, you can use a driller and make 1 hole at time, this would accomplish the task, the robot know how to do it, all done.
So what? this doesn't mean that i can make the drilling time 1/5, 1 drill time 10 second? time to move from one drilling position to another 1 sec? 10 sec * 5 + 1 sec * 4 -> 54 sec
But if i use 5 driller pre-positioned i kill the 1 sec*4 for positioning the driller, and since i do all the drilling at the same time, it doesn't took 54 second, it took 10 second.
So i changed my production rate from 54second piece to 10second piece. thus multiplying my production of a factor of 5.4. wich means that if i was building 66pieces in 1h, now i produce 360 piece in 1h, no change in desing or other. the piece it's the same. i've only changed the way the drilling line works.

Yes, i knew how to drill 5 holes, but i didn't it in a good way, i just did it. it's not the same thing.

Now put this in a weekly perspective, 66piece in 1h, 8h days, 5 days -> 2640 piece in a week produced in the last week
It took me 4 days to change the production line, so i've done 0 pieces while working on changing the production line -> i've worked only 1 days, but i've done it with the new production line, so 360 piece * 8 -> 2880 piece in a week

But the next week i'll producte 2880*5 -> 14400 piece, and if you just look at my previus week production you can say "meeh.. you produced only a small fraction more, and you costed me 4 days of work for what?", you just didn't see the big picture.

Now, this is an extreme case where there is a huge improvemenet, but the thing is, in a big production line you just have to tweak every single machinge to get maybe 10car/h more and it cost you maybe weeks of work.
This is not what you see on the outside, you just see a serius of falling/incresing in the production and not that in the next years you can produce maybe 2x the number of car you made the last years, maybe with less men on the line

 

[zenmaster]

If Tesla is running robots 24/7 ( production )

?

 

[Model 3]

#1.

Its not my opinion that they have been making the MS for years now. Its a fact

Sure. But they have not been producing it in the exact same way all this time. Its a fact.


#2.

Its not my opinion that the standard design of the MS is done. Its a fact

The basic design is done, true. But not all the designs. Just a few month ago it was a big "facelift" of the design, and Elon have said that the make numerous changes to the car etch week. Its a fact.

#3.

Its not my opinion that there is nothing more to engineer. Its done. Its a fact.

GOTO #2

 

[Az_Rael]

The basic design is done, true. But not all the designs. Just a few month ago it was a big "facelift" of the design, and Elon have said that the make numerous changes to the car etch week. Its a fact.

I wonder if the constant tweaking of the MS (and presumably the MX) design is one of the factors in the difficulty with the production ramp ups.

If they were more like a traditional auto manufacturer, they would make the same car for ~1 year before making any changes (barring safety recalls, etc). They would then incorporate all the minor updates into the yearly upgrade, a minor refresh every 2 or 3 years and a major refresh every 5 or so. This does allow the traditional auto makers to really stamp out cars as fast as they can since the supply chains, tooling, etc are all up and running for a longer period of time before changes are made.

If they constantly tweak the Model 3 just like they do the S and X, then those high production rates they are projecting are going to be extremely difficult to obtain.

 

[Model 3]

I wonder if the constant tweaking of the MS (and presumably the MX) design is one of the factors in the difficulty with the production ramp ups.

Probably.

If they constantly tweak the Model 3 just like they do the S and X, then those high production rates they are projecting are going to be extremely difficult to obtain.

I don't think they will. It is reasonable to do that as a new manufacture and/or with a new product. So they may do it the first months on the Model 3, and then just if it is some serious problems that needs to be fixed. And this is also hinted to by Elon. First starts delivery near the factory so any problems could be detected fast and be fixed in production. Then wait with new "cow bells" (and probably fixes) until next edition. But I still think they will not apply to the normal "model year", but maybe release new editions every 6 month or so?

I do expect to get 2018½ Model 3

 

[Burnardr]

Its not my opinion that they have been making the MS for years now. Its a fact
Its not my opinion that the standard design of the MS is done. Its a fact
Its not my opinion that there is nothing more to engineer. Its done. Its a fact.

All Tesla has to do now is repeat success. The MS is a success. If robots are the predominant thing putting the MS together....then there has to be another issue. ( lack of parts/ paint/ slow delivery/ not enough robots...etc. )

If Tesla is running robots 24/7 ( production ) , then I don't know why they are running into delivery issues. It has to be a hang up after production.

The production lines were designed to produce both cars right? If one line is down to fix X issues, Tesla is effectively at half capacity until it's fixed. Then they have to make those same changes to the other line at some point to maintain manufacturing flexibility.

 

[ohmman]

I expect the reason deliveries lagged is that Tesla reports that almost half of its production was in June, so many of those vehicles were still in transit on June 30, including overseas and to the East Coast.

And this is the story every quarter for Tesla. It's like they forget the pace, then someone shows them a calendar, and they freak out. "End of quarter! Production line up to 11!"

Quality for end of quarter vehicles is generally lower as a result of the rush, as well.

 

[JeffK]

Quality for end of quarter vehicles is generally lower as a result of the rush, as well.

Is this quantifiable or speculation I wonder? It makes sense that when humans are stressed or tired they make mistakes in judgement yet the robots don't get tired. err "machines that make machines"

If this quantifiable then maybe it can be improved.

 

[N5329K]

And this is the story every quarter for Tesla. It's like they forget the pace, then someone shows them a calendar, and they freak out. "End of quarter! Production line up to 11!"

Quality for end of quarter vehicles is generally lower as a result of the rush, as well.

This was a problem with Soviet production, too. When quota time came around, they drove screws with hammers to meet their quota number. They called it "storming." Back in the USSR, "consumers" were happy to get whatever they could, even if they knew that a quota rush TV might catch fire, or a stormed washing machine might need a few tweaks before it would actually work.
Expectations have changed.
Robin