Car Science: the sausage machine
Hello! You're reading Car Science; please consider subscribing, it's free and helps me.
Hey,
Working 9-5: it's a surprisingly aspirational way to be making a living, in retrospect, now that we live in horrendous sprint-grind-crunch culture. Gone are the days when the production line downed tools for anything short of a global pandemic and anyone wanting some extended switch-off might need to get in on the noble art of Suez shipping container jamming, these days.
Still, we make stuff and we make it in factories and we make it using production lines. Henry Ford was the first one to make a moving assembly line work, in 1913 - although he actually took the idea from Ransom Olds' Oldsmobile production line, the first to have a car travel from one end to the other being built.
In reality almost all production lines, as concepts, were invented by (mostly poor) women. The process of preparing food is probably where the first person-passing-to-person manufacturing started and there are plenty of places in the world still using traditional production lines to do just that; this is a diversion but I was watching this video about fellas making bread in Samarkand, Uzbekistan the other day and you can't deny that's a production line. (Uzbekistan baking YouTube might take over your recommendations after watching that but honestly, it's a good place)
These days car companies are reinventing the production line all over again. There's some wild stuff like Lotus' plan to have a 16-turn test track with autonomous cars driving at up to 140mph (230kph) as part of shaking themselves down on the production line.* Or the more actually-happening stuff where GM and Ford are competing to build more and bigger battery and assembly factories in the US right now. Federal and local government funding is accelerating their scale-up of new facilities in a way that isn't happening everywhere else and today we're gonna look at Ford's (potential) return to the cutting edge of assembly efficiency.
But before we get into all that: I'm trying to do more with Car Science, including some video stuff and a great justification for that would be bumping up subscription numbers. So if you like this newsletter and you're not subscribed (link at the bottom!) or you want to share it with other people who might like it, please do - honestly it amazes me so many of you already read this but these are challenging times for a freelance journalist and every sub gets me a bit more bargaining power with my own time to give to this and also, hopefully, scientists' to talk to me about the stuff that goes into it. And, y'know: it's free!
Right, youtuber spiel over. Time to get to grips (literally) with automotive assembly lines, production rigidity and the very big problem we call the global supply chain.
Warning: this edition is absolutely massive, so you might wanna take it in chunks
Ford announced in 2021 that it was putting $11.48bn into three new mega-campuses to build electric vehicles. The biggest, in Stanton, Tennessee is called Blue Oval City, the sprawling would be built on sustainable principles, with solar, wind and geothermal energy generation on-site (theoretically, at least) and is slowly emerging from the renderzone to be active in 2025, specifically to build F-Series electric pickups.
The Ford F-Series, for anyone who isn't American, is by far the company's best-selling product line. F-150s remain the bestselling car ("car" in European terms) in the US and the F-250 and F-350 super duties are just as popular. Get a pickup or get the fuck out of the parking lot, basically.
The first production line at Blue Oval City will be specifically constructed to make a new electric truck, currently being called Project T3. In fact, that might be the only truck made at Blue Oval City, if reports are currently to be believed - which is pretty wild when it comes to developing a completely new, bespoke facility.
So: why's Ford doing this, why is it complicated and why do OEMs have to reinvent the production line, anyway? It starts with, as everything, the supply chain.
Picking up a few bits
The F-150 Lightning has been an enormous success, partly because it came at a nearly-equivalent price point to its gasoline equivalent. Linda Zhang's team did everything they could to optimise the build, so that the company could call it non-loss-making even with a price point as low as $40k for the starting model.
Now - I realise - $40k is not a trivial amount of money for a truck. But compared to the way other OEMs have had to reach for the luxury label to make EVs they can price at anything short of cut-me-own-throat losses that was impressive. Unfortunately, the thing about the global automotive supply chain is that it's completely fucked so it doesn't really matter how much you try to optimise it, so by last month that price was sitting at $59k. Even allowing that Ford's confirmed the F-150 Lightning is eligible for the maximum $7,500 tax credit that's a big rise.
(because Reuters is annoying to sign into: the base Pro model that started at $39,974 in 2021 is now at $59,974 - a 34 percent increase; at the luxury end of things the rise has been smaller, from an original ~$90k starting point for the Platinum edition to $96,874 - more like 7 percent)
The price rises are, per Reuters "in response to current material costs, market factors, and supply chain constraints." Which figures: anyone following Formula 1 will remember the pleas from teams for some cost cap leaway last year after freight prices quadrupled. Some freight is still up at 800% of pre-2020 costs and it's not just more expensive, it's less reliable.
W Series' season came to a premature end in 2022, the series quitting in Singapore, without funds to make the final two rounds in the Americas. But it was on thin ice in 2021 - not because
"Obviously, COVID still has significant issues for us," series CEO Catherine Bond-Muir told me at the time, in the weeks just before the COTA finale. "There are logistical issues in the whole of shipping. So will our cars dock on time? Will they go to another port? These are very genuine concerns that we have. I think there are 40 ships off the coast of California, for example, they're not getting into port.
"So we live not having certainty in the way that we would have done historically. But I don't know the way that you deal with it. But actually you get used to the challenges and I stop stressing about it because I go, well, all we can do is our best and it will be what it will be. And so, for example, if our cars for some reason couldn't dock where they're supposed to in time to do the race, what can I do? It's pointless me worrying now about something. The time is beyond my control."
And that's taking one series' cars and equipment to a race location. When you're talking about the thousands of dependencies involved in constructing a car then things get even sketchier - and the companies don't have any better tracking, it's not like W Series were being lax.
It's alarmingly true that even a shipped finished vehicle worth tens of thousands of dollars has way less tabs kept on it than the Domino's pizza tracker, logistics just meant to work. Audi Mexico's supply chain manager described the process of trying to track delays and shipments as being completely manual, at a conference last year; he was simply phoning every day to make sure information was up to date and still searching for a system that could do the same.
Which is why there's been a reversal in globalisation, lately - re-localisation and onshoring have replaced looking internationally; it might be cheaper and more synergised to get chips from Taiwan than Indiana, if there's no global pandemic, shipping is running normally and you don't end up with months-long gaps in strangled supply.
So: everything costs more than it use to. And onshoring incentives like the Inflation Reduction Act, which makes tax credits for EVs dependent on having sourced stuff in the States or neighbouring countries.
That's the global supply chain for you. Car companies are adapting - Blue Oval City is a collaboration between Ford and battery company SK Innovations that will see batteries assembled on site and will work together with two other factories being built as part of the 11.6 billion dollar investment package as a whole.
Car companies aren't spending all this money because they particularly want to. Right now almost nowhere is making EVs at a profit margin suitable for mass production; it's where Tesla, which operated at nearly a 15% profit margin as of the end of 2022, has the biggest head start over Toyota or GM or Volkswagen, despite their manufacturing capacity being a fraction of the giants'.
Making manufacturing more efficient is one of the ways the car industry became profitable in the first place. Leaning out supply chains and having only just-in-time inventory was a way to carve out more margin in the early 21st century, now instability means that you have to go the other way because better to be able to make cars to sell than risk having huge redundancy at your facilities. Especially new, multi-billion dollar ones.
Powering Through
A niche thing that happened on a big scale last year was the US department of energy announced a new sub-department, the Office of Manufacturing and Supply Chains.
Energy has already been part of the manufacturing supply chain for a long time. It's just we haven't really thought too hard about it for awhile; most western grids haven't been under a lot of strain in the last three decades and industry has got good at generating its own power, so it's all become a bit more infrastructural than something you need to bicker for a share of.
But the argument's started: last year Tesla Berlin was the subject of scrutiny from the green party of its power usage, during a shortage of energy from Germany shutting down nuclear power stations and being forced to turn to coal, from natural gas. It's now in an ongoing row about water usage, expansion limited by the toll it will take on an already-fragile, drought-exposed region.
It's not just a bit of greenwashing that companies are installing solar farms and wind turbines and geothermal plants on their new campuses. It's because they have to. Environmental priorities are nice and increasingly something consumers care about, especially with regards to electric cars but there's also the plain truth that if the site can't be self-sufficient, in terms of power, it's going to be vulnerable to an extremely unstable period of energy competition. Which isn't what you spend £5.6 billion for.
It takes about 50kWh of energy to make 1kWh of storage at the minute. That's possibly a little bit toppish, given processes are improving and different manufacturers will have different ways of doing things but it's something around that. That's not bad value, energy-wise, given you're going to charge a car way more than 50 times and that'll come down as manufacturing gets more experienced and adapted to it.
There's a lot of industrial robots used in electric vehicle production (and generally in automotive design because human arms aren't particularly useful for dipping things into plasma chambers, for instance) because of the high voltage systems. Like a lot. To the point this 2021 paper called it the most important sector for industrial robotics.
There's bits of the production line that are too dangerous to have humans working on them and others where it's just too heavy and obviously all that moving stuff around and putting it together requires power both for the things doing the work and the computers telling them how to do it.
Heavy machinery, moving belts, pumping the water to cool them, paint-spraying, parts cleaning, scanning for QC, welding, moulding, packaging, drying - it's all energy and water-intensive. Which is why Blue Oval City will have its renewable electricity sources (and storage batteries to balance them) and also an on-site wastewater treatment plant that means it recycles water, with an aim to reach zero freshwater withdrawals, only water that needs treating or brine. (and there's not a lot of seawater near Stanton)
Which is all a very good way of making sure the plant can keep running. Bringing the battery supply chain on-campus is another factor, localising supplies, trying to optimise production, can all go a long way. But with only Project T3 announced for the site so far, there's a slight chance Ford's put $5.6 billion worth of eggs in one perfectly-designed basket.
Efficiency vs flexibility
Here's a thing about optimisation that anyone who's ever worked on a race car can tell you plenty about: you can't make something that does everything perfectly. It's got to be a balance of doing one thing perfectly and a bunch of other stuff badly or not at all or finding a tolerable medium ground where everything works reasonably well together.
Say: making cars. If a production line makes an F-250 perfectly, it won't make an F-350 very well or maybe at all, depending when the margins start falling apart. Production lines for car models are specific to that car, with factories running a number of production lines and shifts on each.
Once they're built, they're very much not designed to become anything else. Which surprised University of Washington robotics scientist Jeffrey Lipton when he worked on a 3D printable face shield that Ford wanted to assemble. Like most of us, he had a vision that the famed production lines would be flipped straight to making PPE and instead found people sitting in rooms passing them hand-to-hand - exactly the thing you're not meant to be doing during a pandemic.
“We worked on 3D printed face masks with Ford, who at that time had also started making disposable face shields. And I looked at what they were doing and said, ‘this is insane.’ We’ve got all this highly programmable infrastructure and Ford has this huge assembly line technology, they’re leaders in it and yet, when a pandemic hits and you need to switch what you’re producing, the solution is to throw bodies at the problem," Lipton told me last year. "There’s still nothing more flexible than a human. And the fact that we couldn’t, in a crisis, rapidly start up assembly lines, we saw that as a gigantic problem.”
As anyone who's celebrated 4/20 can tell you, the human hand is just, like, really amazing. Super adaptive. So's the human brain. If we weren't so goddamned squishy and electrocutable we'd be the perfect assembly line tools but alas, a 2700kg vehicle will not balance on our spindly limbs.
Which is why we have to make super rigid assembly lines that can avoid dropping things and complete processes efficiently. Within the one thing they can do, it's just that it's slightly like building a one-shot, disposable assembly line if you happen to need it to do something else - it'd need to be totally disassembled and substantially reconfigured even between different electric trucks.
This, understandably, surprised Lipton. So much his team of researchers ended up building a framework for reconfiguring robotic arms and 3D printing a gripper for them to take on a new task, since it seemed so obvious adaptability was the point of having all these robots. I don't know if the automotive industry has yet to take them up on the offer but it seems like a good idea.
After all, build a $5.6 billion megacampus and have a little hiccup in supply that means your Project T3s aren't going anywhere any time soon and you have a $5.6 billion oversized paperweight.
Small town problems
Also, a lot of bored people. Stanton, Tennessee where this 5.6 billion dollar campus is being built, isn't very famous. It's something to do with the fact that, as of the 2020 census, there were only 417 people living there. Almost 41% of whom live below the US poverty line.
It might slightly surprise you that there'll be 6,000 jobs at the plant, given that. So Blue Oval City really will be a city; occupying a stretch of land the state of Tennessee bought for $40 million in 2009 and has been looking for an industry to use ever since. In 2021 and 2022 its listing started getting answers in the form of the automotive and battery industry, who are getting further incentives from the state to shack up there.
$12.7 billion's been invested into Tennessee for EV projects since 2017. That's not a bad thing - clearly, what a town like Stanton needs is good jobs and industry nearby, an influx of people. Especially if the industrial sites are going to be built with a mind to not draining every water resource and drop of electricity in the region.
The point, at last
Anyway, now that we've discussed literally everything to do with automotive manufacturing, supply chains, EV investment in the state of Tennessee and probably a bunch of other stuff the thing that is interesting about Blue Oval City, for all these reasons, is that it's being purpose built to make the Project T3.
This report in Automotive News caught my eye because of something Ford's Lisa Drake said about the new site in it -
"When you walk in, it will not feel like a plant that you've been in before," Lisa Drake, Ford's vice president of EV industrialization, told Automotive News. "Our target was to actually build this truck more efficiently than the best truck we build — and we build trucks pretty well."
The plant will have fewer workstations, Drake said, and the vehicle built there, known internally as Project T3, will be offered to customers in significantly fewer configurations than today's models.
Both Blue Oval City and Project T3 are being designed together. The car to have the greatest construction simplicity it can, using more moulded parts and less tooling, the plant to be optimised to make it.
And there's one specific place they're saving on that, according to Ford CEO Jim Farley: people.
"The most important thing is to reduce the labor content," Farley said. "We're insourcing with batteries and things like that. The vehicle's designed in such a way — because it's second-generation — where we radically simplified the subassemblies. It's a really big change, the way this will be built."
All the complicated bits that can't be done by robotics have to be done by people. Some things, like wiring harnesses, aren't likely to change very soon from that (especially as wiring gets more and more complex) but the rest of the work is to try and move away from the inception of the production line - people passing work on from one another, after performing a specific part of it - to something more robotically driven.
Robots don't mind working long shifts, as well as being better at dipping into plasma and surviving electrocution. 24/7 running was something that car companies were experimenting with from 2009-onwards, in a bid for more efficiency, until the disruption of COVID.
Worrying about reducing labour isn't just a cost thing, though. Especially when you're investing in megaprojects in small towns in West Tennessee.
There's been a lot of OEMs building factories outside their usual stomping grounds. The state and federal governments are keen to offer incentives and even grants and investments for bringing high-paid, good jobs to these areas and OEMs need a lot of space and infrastructural potential for the plans they're currently undertaking. It's why the equivalent projects in Europe are happening in eastern Hungary and other places people don't look all that hard at the broader impacts of the construction.
Still: New York and Berlin (as proven by Tesla) don't want the projects, they have to go somewhere, there they go. Except that there don't go the workers.
The US has a crazily low unemployment rate right now. So the incentive for anyone to move from whatever they're doing, wherever they are, to pursue work in a totally new region, at a new factory, is relatively low. And car companies are struggling to find staff for their new plants.
So there's a real incentive, if you're going to keep the plant running, to try to make it as robot-reliant as possible. Shift work on production lines is less appealing to people, as something where you unavoidably have to go into a physical work location, especially given it's relatively technical, skilled and often not massively better compensated than other jobs. Ford says Blue Oval City will be bringing $33 per hour work, which is about standard for an automotive technician - but might not be enough to convince you, if you're already working in that, there's so many perks to moving to Stanton.
Shift work, to be fair, has become difficult to cover across the whole automotive industry. Audi has had success by turning its production line work flexible; workers sign up for the shifts they want and can do, rather than being assigned. It started as a trial in March 2022 and was successful enough it got a full-scale roll-out; Audi's a small manufacturer compared to Ford, however and not breaking ground on new factories. (it's F1 engine test beds aside which - let's not get into that, we're already 3,000 words in)
Labour, like power, hasn't factored into the supply chain before. Which has mostly been preoccupied with the important work of optimising container space for thousands of different types of fastener - which is and remains incredibly important - making factories sustainable for the people who work there is important, too. If the nearest town's a drive away, then solutions like BMW's private bus services between Munich's satellites and its factories might seem antithetical to a car company but mean the cost of transport is suddenly no longer a factor, something employees think about more than employers realised before the pandemic.
Anyway; I should stop talking about this before it turns into an even longer ramble. But building a factory: it's complicated. Ford's probably doing the right thing, unless what it actually builds turns out to be a bad fit for what it has to build next because this is all a learning process.
Creating a wonderfully bespoke factory is going to bring all kinds of immediate optimisations (hopefully) but the one thing you can say about the present is we sure don't know the future, so spending billions on anything less than extreme flexibility remains a gamble.
Right, see you on Friday for some stuff about electrolytes,
Hazel
x
*No, really. In an absolute masterpiece of burying the lede Lotus announced this about 16 paragraphs in to a release with the bland title of "Lotus global HQ established in China as future Lotus EV product plan revealed" where it managed to almost hide it was planning anything bonkers or that it had announced a new range of five cars. Automotive PR crafting in 5D.
Images:
Blue Oval City render (via Ford US)
Blue Oval City battery park render (via Ford US)