Robots are deployed in automotive manufacturing for various applications, including heavy lifting, welding, robotic vision, and many more. Yet, how much of the automation worked?
In part 1, we discussed how Mercedes-Benz did with automation. The luxury car manufacturer integrated robots into their workplace but failed to realize that mass customization can be done faster by a human than a robot.
In this article, we will discuss what Tesla has done to automate and how efficient it was.
Elon Musk, the founder of Tesla, has a grand scheme to automate the entire production line for its electric cars. Many touted it as too ambitious. After 2 years of its inception, how much of it actually worked?
What Worked: Rapid Development Using Agile Framework
Situated in Silicon Valley, Tesla has adopted the notion of “scale first, fix later”. This relatively new car manufacturer forgo-ed trial productions, long development cycles and dived straight into full-scale production. Essentially, they took agile software development ideals into the automotive industry. 
Tesla Factory Automation
With this rapid scaling progress, Tesla factories became highly automated in a short amount of time and are now on par with its competitors.
What Didn’t: Finer Details Still Require Humans
While the agile framework allowed Tesla to scale at a phenomenal speed, it led to them being too ambitious in their automation process.
“Excessive automation at Tesla was a mistake,” Musk tweeted. “To be precise, my mistake. Humans are underrated.”
Some tasks are easy for humans to execute but tough for a robot with today’s technology. The details, such as tough wiring harnesses, required manoeuvring across the car body frame, which is a simple task for a human but tough to program on a robot. Unfortunately, robots are unable to orientate themselves to attach items like nuts and bolts or to move between the car frame. For Tesla, even a small issue can cause the assembly line to halt.
Tesla engineer alongside Robot
The missing piece to full automation in Tesla is in the robots’ software and programming.
Machine learning and artificial intelligence today is still not capable of completely replicating the intelligence and adaptability of a human. With further advancements in the systems that run these robots, the current difficult tasks faced by robots will be made more manageable eventually. Overestimating the capability of the robots was the main mistake in Tesla’s execution.
As a result, Tesla experienced numerous production delays to fix the areas where the robots failed.
Tesla made the choice to scale quickly and fix problems later, adopting the popular Agile framework, to move fast and break things. This brought about faster technological advancement, and rapid scaling to be on equals with competitors that have existed for hundreds of years.
At the same time, this method led to the failure of optimising the process before applying it, causing many losses. Unlike software which can be updated with a click of a button, these physical cars have to be brought back to the factories to be installed with the latest updates.
Whether Tesla is ahead of its time for envisioning a fully automated factory – deploying only robots, only time will tell. As technology progresses, the software and programming needed for 100% automation may not be too far into the future after all.
Part 3: Toyota
In the last part of this series, we will look into Toyota, and how they prioritised process over automation and whether that worked for them.
Augmentus offers a no-code and fully-integrated robotic programming software that enables anyone, even those with no robotic experience, to program dynamic industrial robots in minutes. Our proprietary technology incorporates an intuitive graphical interface on an iPad that eliminates the need for coding and CAD files in robot teaching. Companies using Augmentus have experienced up to 70% cost reduction and 17 times faster deployments across a wide variety of applications, such as spraying, palletizing, welding, and inspections. Augmentus ushers in a new era of human-machine interface, democratizing robotic automation.
Book a demo now to experience it.
Robots were used in the automotive industry since the 1970s and today, it is no surprise to see machines in car factories for manufacturing and logistics. How has automation transformed automobile factories? As technology progresses, machines and robots get more sophisticated. Robots are here to stay and they will get smaller and more flexible. The […]
16 Feb 3 mins readRead More
Pinxi Tan is a second-year undergraduate from the National University of Singapore, studying a Bachelor of Computing in Computer Science. Pinxi has been working closely with the technical team since January 2020 as a software engineer intern, developing our prototype from a simple proof of concept to a full-fledged functioning product. We sat down with […]
05 Oct 3 mins readRead More
In today’s age of automation, Oxford Economics, a UK-based research firm, estimates that 20 million jobs will be wiped out by robots globally by 2030. While this may be bad news for some, it also means that more jobs would be created in the robotics and automation industry. Roles such as robotics technician or robotics […]
30 Jul 3 mins readRead More
On a global scale, robots for agriculture is one of the biggest trends for 2020, as the world struggles to reduce its vulnerability to volatility. As a small country that relies so much on foreign imports, Singapore has to adapt. With Singapore’s “30 by 30” goal to have 30% of its food supply onshore, agri-food […]
15 Jul 3 mins readRead More
04 Jul 3 mins readRead More
Looking to further enhance your manufacturing process with robots? Lean robotics is a guide by Samuel Bouchard, the CEO of Robotiq, to help you efficiently deploy collaborative robots in your factory. It primarily integrates lean principles and collaborative robotics into your existing process design as straightforward as possible. Lean robotics may be outlined by the […]
17 Jun 3 mins readRead More