The Ford Motor Co. perfected the moving assembly line in 1913, and every automaker has been using it ever since. But Tesla announced it’s developing a new assembly process intended to revolutionize how cars are made and cut costs in half. Everybody in the industry better sit up and pay attention.
History books credit Henry Ford with inventing the moving assembly line, but Henry really wasn’t involved at all. If any one person deserves credit for the moving assembly line at Ford it’s probably a guy named Clarence Avery, whose name has largely been lost to history. He and his crew were the ones who figured out how to make it work.
They got the idea from two different places, grain mills and slaughterhouses. They observed how grain mills moved grain – like corn, wheat and barley – in buckets on a conveyor line into the mill to get ground down into meal and flour. At slaughterhouses, they observed how animal carcasses traveled on conveyors where workers along the line would cut off different parts of the animals.
That’s how they got the idea to start using conveyors to move parts, and to run the slaughterhouse process in reverse—instead of cutting parts off, the line workers would bolt them on.
It all started in 1913 at Ford’s assembly plant in Highland Park, MI. Avery and his crew experimented with engine parts, magnetos and finally with an entire car.
They literally tied a rope onto a Model T chassis and started winching it across the floor. They put piles of parts along the line and had workers pick up the parts and bolt them on the car. They used stopwatches to do time-and-motion studies until they finally figured out the right number of parts, the right number of people and the right speed for the line to move. Amazingly, today’s assembly lines pretty much move at the same speed those guys figured out in 1913.
But doing it that way also required a lot more people on the line, because a moving assembly line is actually 33% inefficient. Here’s why:
In a typical assembly plant, a car rolls off the line every minute. That means each worker on the line has 60 seconds to complete the tasks in their workstation. They pick up parts at the beginning of their station and walk along with the car as it goes down the line, installing the parts as they go.
When they get to the end of their workstation, they walk back to the beginning, pick up more parts and start the process all over again. Typically, they have 40 seconds to do their tasks and 20 seconds to walk back.
Those 20 seconds are waste. There is no value being added to the car as they walk back. The 20 seconds come to 33% of workers’ assembly time. And none of that time is added value.
At many stations it’s even worse. If they have to climb into the car, or reach under the dash, they may only have 10 seconds or so to do their task, which means their inefficient, non-value-added time could be as high as 80%. The only way to compensate for that wasted time is to add more stations and more people, and that adds cost.
Manufacturing experts have known about this inefficiency but just accepted it. Except for Tesla, that is. It wants to put cars together in a completely different way.
Today, automakers stamp out panels, weld them together, paint the entire structure and then install all the components.
Tesla doesn’t want to do it that way. It’s going to take a modular approach, with essentially five modules. The left side of the car is one module, the right side is another. There’s a front module, a rear module and a floor module (pictured, below). All those modules get put together in one final station and then it drops a glass roof on top of it all.
Here’s more detail. The left-hand side of the car gets stamped and painted and so does the right side. Next, the interior parts get attached, like door trim panels, the door seals, door windows, hinges, etc. And because the sides of the car and closure panels are separate, it’s very easy to install parts on them. Probably a lot of it will be automated.
The front and rear modules are gigacastings with all the needed components attached to them. The front module gets the powertrain, the suspension, brakes, steering and the instrument panel including the steering wheel. The rear module gets the rear suspension components, the brakes and trunk. Since they’re assembled separately from the car, there’s more opportunity for automation. And these subassemblies can be fully tested before they all go together.
The floor module will be assembled just like Tesla does now. It’s a battery pack with the carpeting, seats and center console mounted on top of it.
The final step is for all these modules to come together. The sides, floor, front and rear modules get put together, I’m guessing with structural adhesives, then the glass and the closure panels are added.
Theoretically, at least in my mind, Tesla doesn’t need a final assembly line. It might have multiple final assembly stations with feeder lines bringing in the modules. That way, if one station had to stop for some reason, the rest of them could keep on making cars. On today’s assembly lines, if you have a problem in just one station on the final line, the line comes to a stop.
The payoff with Tesla’s new approach could be huge. It claims the manufacturing footprint will be 40% smaller, which means it can build new plants faster with less investment. And it says it can cut assembly costs in half. This is going to be the key way Tesla will be able to offer a $25,000 electric car and still make a tidy profit on it.
To me, if Tesla can pull off this new way to make cars, it’s going to be the most significant change in automotive assembly since those guys at Ford tied a rope on the chassis of a Model T and started winching it across the floor 120 years ago.
John McElroy (pictured, left) is the President of Blue Sky Productions, which produces “Autoline Daily” and “Autoline After Hours” on www.Autoline.tv and the Autoline Network on YouTube. The podcast “The Industry” is available on most podcast platforms.