3Q18 Model 3 Production - Take 2

[ValueAnalyst]

We now have further information, so let's run another poll.

Tesla achieved its goal of 5,000 weekly production rate by June, but Wall Street doubts its sustainability.

Tesla has also said it aims to achieve 6,000 weekly rate by the end of August.

Here is the previous poll for your reference. Per @generalenthu's request, I have added further granularity to options.

Good luck and Go Tesla!

 

[SlicedBr3ad]

Tesla achieved 5000/week in Q2 but the average was ~2500/week over 12 weeks. I expect about 45000-48000 model 3 produced in q3 based on 4000/week average rate with bursts up to 7000/week by end of q3 and some factory closures 1-2 weeks to improve production rates.

 

[SlicedBr3ad]

Tesla achieved 5000/week in Q2 but the average was ~2500/week over 12 weeks. I expect about 45000-48000 model 3 produced in q3 based on 4000/week average rate with bursts up to 7000/week by end of q3 and some factory closures 1-2 weeks to improve production rates.

Don't forget, they're doubling production rate from q1 to Q2 and again Q2 to q3. The main difference is whether Elon will be at the factory to motivate and solve problems. In q1, he wasn't and we know how that turned out.

 

[electracity]

The main difference is whether Elon will be at the factory to motivate and solve problems.

I'm speechless.

 

[Brovane]

Tesla sustained a burst rate of about 2,000+ Model 3 cars at the end of Q1 but was able to sustain about ~2,500 cars for all of Q2 as a rolling average.

I am going to with about 6,000 cars a week on a average for Q3 so about 70,000 cars for Q3. Especially considering that Tesla isn't going to hold back any shipments for Q3 and Q4 this year with the tax credit expiring.

 

[RABaby]

Tesla sustained a burst rate of about 2,000+ Model 3 cars at the end of Q1 but was able to sustain about ~2,500 cars for all of Q2 as a rolling average.

I am going to with about 6,000 cars a week on a average for Q3 so about 70,000 cars for Q3. Especially considering that Tesla isn't going to hold back any shipments for Q3 and Q4 this year with the tax credit expiring.

Can we refrain from the use of the term "burst rate" when speaking of production? This reference is used by those who question everything about Tesla without any effort to understand definitions. And if it doesn't suit the goal of the short, there will be no understanding.

A burst rate is a test of one area of the factory to produce at a specified rate.
All the areas of the factory each on their own have to be tested at a similar burst rate to verify that the factory as a whole can run at that rate. After each area passes the test, running all the systems together will reveal whether it is a synchronized run and then it continues at that rate until such time as a tweak is performed and a new set of burst tests are conducted .

The simplified definition of burst rate is running a system at a higher rate than it had been operating,
A burst rate does not apply to operation of the factory. You don't have a burst rate for the factory; you have a run or production rate.
Whatever the total produced in a week is the current rate.

And as they ramp up, there are constant tweaks and downtime and more burst tests so that the run rate can be increased. But the burst only is for an area and not the factory. Synchronizing all the areas will yield a production rate at the speed of the slowest area. But it is not a burst rate for the factory.

I don't know how many ways to describe the difference between burst and production, but the amount produced in a week is not a result of a burst test. If someone else can be more concise and put an end to the misunderstanding, that would be great.

 

[dhanson865]

Tesla sustained a burst rate of about 2,000+ Model 3 cars at the end of Q1 but was able to sustain about ~2,500 cars for all of Q2 as a rolling average.

I am going to with about 6,000 cars a week on a average for Q3 so about 70,000 cars for Q3. Especially considering that Tesla isn't going to hold back any shipments for Q3 and Q4 this year with the tax credit expiring.

It only takes 5,769 per week average to max out this poll, not sure why but ValueAnalyst chopped the poll short of the 80,000 option that was on the prior poll.

I guessed 70,000 last time, I had no issue with bumping it up to 75,000 plus given a months more time to think about it.

 

[ValueAnalyst]

It only takes 5,769 per week average to max out this poll, not sure why but ValueAnalyst chopped the poll short of the 80,000 option that was on the prior poll.

I guessed 70,000 last time, I had no issue with bumping it up to 75,000 plus given a months more time to think about it.

6000 per week is Tesla’s goal for the end of August. Given the usual one to two weeks of downtime per quarter, I don’t see more than 75,000 units produced in Q3. Having said that, I hope you prove right and I prove too conservative. I am often faulted for having answer ranges that are too optimistic, so it’s nice and refreshing to see your post. @generalenthu would be proud.

 

[Yggdrasill]

I'm thinking a ramp along these lines:

Week 1: 1500 (largely shut down 4. july)
Week 2: 3500
Week 3: 4000
Week 4: 4500
Week 5: 4500
Week 6: 5000
Week 7: 5000
Week 8: 5500
Week 9: 6000
Week 10: 6000
Week 11: 6000
Week 12: 6000
Week 13: 6500

64,000 total, minus an additional one week shutdown at some point, so minus ~5,000 cars, resulting in 59,000 Model 3s produced.

 

[SlicedBr3ad]

I'm speechless.

Just my take on what I saw from daily monitoring of this from July 2017 to July 2018.

I'm thinking a ramp along these lines:

Week 1: 1500 (largely shut down 4. july)
Week 2: 3500
Week 3: 4000
Week 4: 4500
Week 5: 4500
Week 6: 5000
Week 7: 5000
Week 8: 5500
Week 9: 6000
Week 10: 6000
Week 11: 6000
Week 12: 6000
Week 13: 6500

64,000 total, minus an additional one week shutdown at some point, so minus ~5,000 cars, resulting in 59,000 Model 3s produced.

I hope you're right and I'm wrong but it's possible there will be another shutdown or two...

 

[Yggdrasill]

I'm thinking there will be an additional long shutdown to bring the GA3 line from ~4000/wk to ~5000/wk. Other than that I don't really see what they would need to shut down for. GA4 is already at something like 1000/wk, which is sufficient, and can probably be improved to some degree without a longer duration shutdown. It should be possible to work on GA2 without shutting down GA3 and GA4. Improvements to the body-in-white line and the paint shop can probably be timed to coincide with the one week shutdown of general assembly.

But we'll just have to see. As long as the number is >50k, I won't be very disappointed.

 

[ValueAnalyst]

I'm thinking a ramp along these lines:

Week 1: 1500 (largely shut down 4. july)
Week 2: 3500
Week 3: 4000
Week 4: 4500
Week 5: 4500
Week 6: 5000
Week 7: 5000
Week 8: 5500
Week 9: 6000
Week 10: 6000
Week 11: 6000
Week 12: 6000
Week 13: 6500

64,000 total, minus an additional one week shutdown at some point, so minus ~5,000 cars, resulting in 59,000 Model 3s produced.

I believe Tesla will beat your first four weeks by 500 units per week; otherwise, we generally agree.

 

[defc0n]

These poll options are much more reasonable than usual, VA I chose 60k, expecting somewhere between 55k - 60k.

 

[EinSV]

60K exiting Q3 above 6K/week rate.

 

[mongo]

Are we playing closest to the pin, or Price Is Right rules?

 

[ValueAnalyst]

Are we playing closest to the pin, or Price Is Right rules?

Closest to the pin.

 

[electracity]

I would love to see the week by week forecast for those voting 75000 or more. That would presumably mean exiting September at upwards to 8000/week. That would put Fremont at a run rate of a half million total vehicles a year entering the fourth quarter.

 

[mongo]

I would love to see the week by week forecast for those voting 75000 or more. That would presumably mean exiting September at upwards to 8000/week. That would put Fremont at a run rate of a half million total vehicles a year entering the fourth quarter.

For the 75k bin, production needs to hit 72,500 (maybe 72,501)
At 11 weeks production, 6,591 average, with 5k at start, over 8k is needed at end.
At 12 weeks of production, that is 6,042 a week average. If they start at 4,800 and have a linear ramp, the needed end rate is 7,284.

 

[ValueAnalyst]

I would love to see the week by week forecast for those voting 75000 or more. That would presumably mean exiting September at upwards to 8000/week. That would put Fremont at a run rate of a half million total vehicles a year entering the fourth quarter.

Not necessarily. One can argue that Tesla took extra time to make major upgrades to its production lines in Q2, before hitting the 200,000-mark, and that no major downtimes will be needed in 3Q during the credit phase-down period. I don't believe Model S and X production line requires any major stoppage every quarter. In that case, 5,800 weekly average in a 13-week period would meet a 75,000+ forecast.

 

[electracity]

Typical practice is to shut down a week so that production workers take most of their summer vacation days at the same time. Worker desire for time off in July And August becomes problematic without a shutdown.