Tesla’s Everest


It’s good to be Tesla.  With pre-orders for the yet to be release Model 3 over 325,000, each with a $1,000 reservation fee, Tesla just “crowd-sourced” over $325 million.  Pretty Impressive.  If all of these orders transfer into sales, then Tesla will make somewhere around $14 Billion in revenue.  More Impressive.  However, while the analysts might be oohing and ahhing over Tesla’s production facility and these orders, actually producing them might be a little more daunting.  In what may be the understatement of the year, CEO Elon Musk tweeted, “Definitely going to need to rethink production planning…”


Rethink might be appropriate for a genius like Musk, but his team is probably freaking out.  I used to do capacity and production planning for Ford’s largest North American assembly plant, so I have an idea of what they need to do, and if they aren’t worried, they should be.  Forbes put some of this into perspective the other day with a “Be Careful What You Wish For” article, and at a high level, it doesn’t look pretty, but when you dive into the details, it’s downright ugly.  Here’s why:


Tesla has a current stated capacity of 100,000 units per year, which, we’ll assume is a straight time capacity (no overtime factored in).  For planning purposes, the number of production days in an automotive assembly plant is around 240 [365 days/yr – 104 days (weekends) – 8 days (holidays) – 4 days (typical Christmas/New Year break) – 9 days (summer shutdown) = 240].  This is the baseline number of days.  Tesla could choose to operate on certain weekends, not take shutdown periods, etc. but those would all be incremental to baseline capacity.  The baseline is what is used to set up operations, supply chain requirements, etc.  There must be a baseline.


Not knowing the shift pattern, let’s assume they run 2 shifts at 8 hours per shift, or 16 production hours per day.  16 hours per day at 240 days per year yields 3,840 hours per year available for production.  That means, Tesla currently can produce and report around 26 units per hour (UPH) in the final assembly area.  With 5% over speed (pretty standard in the industry), that means they are running at 27.3 UPH in final (faster in Paint, Body & Stamping – assuming the same shift pattern), or producing a vehicle every 2.19 minutes (131.67 seconds).  Everything is designed to support this volume – supply chain, equipment, work stations, staffing, maintenance schedules, etc.


Now, here’s where things get interesting.  If we say that Tesla needs to produce 300,000 units per year, the numbers change just a little bit…


Required UPH in Final is now 78.1, or 82 UPH with over speed.  That equates to producing a vehicle every 0.73 minutes, or every 43.89 seconds.  It’s one thing to produce a vehicle every 2 minutes or so, but it is a completely different game to produce one every 44 seconds. [Note: should they go to a full overtime model (2 shifts at 10 hours each), the UPH requirements drop from 78.1/82 to 62.5/65.6, or a vehicle every 55 seconds.  Either way, a significant change.]


Everything, and I mean EVERYTHING is faster.  And not just a little bit, but a lot.  The supply chain issues alone can bring a plant to it’s knees.  Delivery trucks are in and out faster, sequencing becomes critical, work station balances must be precise or people will stumble over each other, robots and other machines move faster, and there are more of them, quality, both internal and supplier, must be tight; everything must be designed to support this.


What before were minor issues quickly become major issues.  For example, a 5-minute breakdown today would result in about 2.3 jobs “lost” in production (around $80,500 in revenue).  In the future, 5 minutes means 6.8 jobs “lost” (around $238,000 in revenue). [note: breakdowns are why you have over speed, but clearly, the numbers add up much quicker]


It just happens faster.  Everything accelerates – including the stress on the workforce and the company to deliver.  The culture is different.  It is hard to understand just how much more difficult this environment will be to work in and produce the quality Tesla’s customers will expect.   I’ve worked in a plant that operated at 45 UPH, and the same one at 80 UPH, and let me tell you, life was much easier at 45 UPH…


Now, some of the pressure can be alleviated by modifying the production environment, for example, by adding a second final line (back departments may still struggle), which can spread the production risk.  Of course, that means more equipment, more synchronization, and more people.  There are also alternatives at a high level – Tesla could build another factory.  They could delay delivery to customers.  They could run a lot of over time, change shift patterns to operate 7 days a week, sacrificing preventive maintenance plans, and employee satisfaction (really only a short-term solution, and given the level of automation and number of robots, probably not a wise decision to reduce PM).  Or some other creative solution.


Regardless, this will be a challenge as large as Everest.


Now, Tesla has a lot of smart people to figure this out (I hope), and I do believe they will.  But if anyone thinks it will be easy or fast, they are sadly mistaken.


Musk stated in 2014, “People don’t quite understand how hard it is to manufacture something.  It is really hard.”  Well, if he thinks it’s hard now… just wait..

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s