One of the most interesting points of the entire event for me concerns Tesla’s new engineering approach, a synergistic product/process development approach. The so-called “over-the-wall,” or working in silos or watertight compartments, does indeed make it extremely difficult to integrate the skills needed to develop a physical product. However, it is still very difficult to eliminate in many companies.

Tesla, through strongly multidisciplinary teams, completely reshapes the entire process of product design and production system, based on an agile principle: by dividing the assembly system into smaller work areas, it is simpler and faster. This leads to greater “space-time density,” as it is defined in minute 1:07:42: multiple people or robots can work simultaneously on a product part. As a result, greater process efficiency is achieved. As will be reported later, OEE in Tesla in fact reaches about 90%.

Another advantage of this approach is the possibility of introducing more automation. In fact, one of the traditional limitations when assembling a car concerns ergonomics and bulk: a robot simply does not have the flexibility of a human operator and cannot reach the gaps needed when setting up the interior parts of the vehicle. When Tesla tried to fully automate the assembly process, the result was the so-called “Model 3 Production Hell” (minute 1:01:34). “A great idea, buti it wasn’t the right timing.” Tesla’s new approach to vehicle design aims to remove this limitation.

Normally, automakers purchase power electronics from their suppliers. Tesla designs critical elements in house (minute 1:13:34). For the record, the same approach is used at SpaceX, where about 80 percent is designed and manufactured in house. Tesla also designs the software and part of the manufacturing system facilities in house, such as the automated facilities for producing the powertrain unit (minute 1:16:05).

This leads to a number of advantages, mainly 4:

• Performance: software and hardware are synergistic and optimized for the product, improving performance
• Resilience: by not having to rely on sub-suppliers, there is greater responsiveness in component production
• Greater control over quality (better), time and cost (lower)
• Know-how: key know-how is ensured to remain in-house

Another interesting point is the different working relationship with their suppliers. What is not produced in house is produced by a network of suppliers (Tier-1 and Tier-2). Tesla co-develops its components together with its suppliers. This means that Tesla employees also work with suppliers for several months to support the development phase, improving communication between the parties and consequently speeding up the decision-making process. A highly vertical and integrated supply chain leads to a number of benefits, including:

• Quick response during the product development phase
• Involvement of key partners, leading to a greater sense of ownership of the final product
• As before, greater control over quality, time and cost

As reported from minute 1:28:00, Tesla, through its telemetry system, collects a huge amount of data with 2 goals:

• Improve the product
• Make the driving experience adaptive

Based on the data collected, Tesla’s software is updated in real time, without the need to go to specialized workshops. Making the driving experience adaptive means adapting the car’s characteristics based on specific terrain characteristics, for example, changing the stiffness of the car’s suspension. The data collected also help to improve the design of subsequent vehicles (not models!) based on real user feedback.

In summary, through data collection, Tesla creates a true digital twin of its fleet, through which it monitors and improves the performance of its vehicles.

From minute 1:34:00, how software integration occurs during the assembly process is described. “Test early, test often.” Software is integrated early in the assembly process and verifies that all components are assembled, correctly. In fact, each component is equipped with sensors that send information to the system during assembly. If anomalies occur, it is then possible to intervene before the car is fully assembled. The risks of anomalies are thus reduced and the “reaction plan” in case these anomalies occur is improved.

Tom Zhu reports some numbers, specifically 5:

• 4 global plants
• 65k employees working in manufacturing
• Annual capacity of 2M vehicles
• Cycle time: 45 seconds
• OEE at 90%

The latter value is from my point of view extremely interesting, as it is higher than the typical values typically aspired to, about 75-80%. Regarding cycle time, BMW in its plant in Munich, winner of several Lean awards, claims a cycle time of 57 seconds in the production of endothermic vehicles (read my article here), while for production of the i3 model at the Leipzig plant, a cycle time of about 100 seconds (source: ChatGPT).

Tom explains the philosophy used for cycle time reduction, the same one Elon used for building rockets at SpaceX (minute 2:28:55):

In short, Tesla’s continuous improvement approach is based on reducing waste (“delete”), thus a lean approach. Only after this stage is it advisable to “accelerate” by introducing new technology and then “automate.”

In the last part of the Q&A event, Elon emphasizes how the company is trying hard to reduce production costs so as to ensure greater affordability for future buyers. According to Elon, there is no doubt that people want to buy a Tesla, the problem is that it is still too expensive. The points highlighted above go in this direction:

• Vehicle redesign to provide greater accessibility for robots
• Vertical integration of the supply chain
• Increased efficiency and reduced cycle times through elimination of waste (lean)

In the last minutes of the event, Drew Baglino, SVP Powertrain and Energy Engineering, reports this phrase, often repeated to the team by Elon: “It’s OK to scrap equipment or money, it’s NOT OK to scrap time.” This approach once again demonstrates how speed and agility-the ability to make decisions quickly based on objective feedback-are among the company’s guiding approaches. If learning means making mistakes, even investing money, then that is fine. What is not fine is not trying, not testing, because that does not allow learning, and therefore is a waste of time.

By virtue of the information so far, there would seem to be no room for doubt: Tesla is setting itself up as a role model for electric vehicle production. Is this really the case?

A final paranthesis worth making is to ask how Tesla stands in the current market, particularly in reference to its fierce German competitors. In this regard, I invite you to read an interesting article that recently appeared in Seeking Alpha: A Comparison Of The German Premium Auto Manufacturers And Tesla. In particular, in reference to superior profitability, which is often cited, the article points out that when considering “carbon credit” subsidies, Tesla appears to be less profitable than its German competitors.