“For the BMW Group, the use of industrial data is a key factor in translating artificial intelligence into value creation,” said Dr. Franz Decker, CIO and Senior Vice President of the BMW Group. “By combining our engineering datasets with Mistral AI’s model training capabilities, we are building specialized AI which supports complex development tasks.”  

Complexity and Data Volume in Crash Simulation  

The scale and complexity of crash simulation at the BMW Group underline the need for domain-specific AI. Each week, the company runs thousands of virtual crash simulations, generating vast amounts of engineering data. Over time, this has resulted in a historical dataset of over one petabyte of crash simulation data. It provides highly detailed insights into vehicle structures and material behaviour, forming a unique foundation for training an industrial AI model.  

“As Industrial AI becomes the new frontier for AI, we are proud to partner with the BMW Group” said Marjorie Janiewicz, Chief Revenue Officer of Mistral AI. “This collaboration shows how industry specific AI models can help solve complex engineering challenges such as crash simulation.”  

Large Industry Model as technical foundation  

To scale this approach, the BMW Group is focusing on so-called Large Industry Models (LIM). These are AI systems trained on industry specific engineering and simulation data from vehicle development and safety testing. Unlike general‑purpose AI systems, LIMs embed domain‑specific knowledge directly into the AI model. This requires not only industrial data, but also deep domain expertise and technical environments that allow AI systems to learn directly from BMW’s development processes.  

The partnership highlights the importance of industrial data for the next phase of data‑driven value creation and strengthens the BMW Group’s AI and innovation ecosystem.  

Debrecen. The BMW Group has entered the final phase before the first Neue Klasse model rolls off the assembly line. Series production of the new BMW iX3 will get underway in late October at the BMW Group plant in Debrecen, Hungary, as the newest, most innovative manufacturing facility in the company’s global production network officially begins work.

“The start of series production for the BMW iX3 signals a new era of automotive manufacturing. Our new plant in Debrecen has been designed and built fully in line with our strategic vision of the iFACTORY. Digital from the very beginning, the plant will offer a new dimension in efficient production without fossil fuels,” says Milan Nedeljković, member of the Board of Management of BMW AG responsible for Production.

The BMW iFACTORY represents a strategic vision of production, built on streamlined, efficient structures, responsible resource management, economical use of cutting-edge digital innovations and a strong focus on employees. The BMW Group has also registered new patents for a large number of innovative systems and processes at Plant Debrecen.

“We have taken on the challenge of building a completely new vehicle as the first job at an entirely new plant – and doing so in the leanest, most efficient way possible. We have simplified processes, reduced complexity, digitally validated each individual operation and systematically leveraged expertise across our global network. The result is this plant, with its efficient production, innovative processes and the flexibility to integrate additional models,” says Hans-Peter Kemser, head of BMW Group Plant Debrecen.

As with any new model, capacity for the new BMW iX3 will be gradually ramped up in stages after the start of production. Spearheading the arrival of the Neue Klasse, the iX3 showcases the latest design and technology innovations that will define the brand's entire model range in the future. Between now and 2027, the technologies in the Neue Klasse will be integrated into 40 new models and model updates.

From virtual factory to reality

The guiding principles of the BMW iFACTORY have been fully implemented in the technologies employed at the new BMW Group plant in Debrecen. For example, Plant Debrecen was planned and built digitally from the outset, and celebrated its virtual start of production back in March 2023 in the BMW Group’s Virtual Factory. This approach allowed every operation to be tested virtually in advance, enabling production lines to be installed in the buildings exactly as in the digital twin.

Optimal processes for all plant technologies

With an optimised value stream and efficient systems, the press shop guarantees a high level of productivity. The BMW Group continues to follow its proven strategy of using the same tools and presses throughout its global network. This optimises capacity utilisation within the network by allowing press tools to be deployed at multiple locations and enabling staff to be trained in-network.

The body shop, in particular, benefits from prior digital planning and validation. Every process detail was digitally simulated in advance to determine the ideal value stream and ensure that each of the nearly 1,000 robots was optimally positioned. From the outset, close collaboration between development and production teams led to high production efficiency and maximum customer benefit. For example, the number of joining methods was significantly reduced, lowering overall complexity. Design features integrated early in development are also evident in the details of the Neue Klasse, such as the invisible door seal. This creates a visual connection between the window and the door, resulting in a unique look. In addition, maximising the space available for the battery in the body structure – and thus the size and performance of the battery – directly benefits the customer.

At BMW Group Plant Debrecen, the paint shop plays a critical role in significantly reducing the CO₂e footprint of the BMW iX3. Production of the new BMW iX3 will generate a total of about 80kg CO₂e (scope 1/2 emissions). This figure covers CO₂e emissions from Plant Debrecen, as well as in-house parts production at other BMW Group facilities, including components manufactured in Landshut, for example. This represents a reduction of approx. two-thirds compared to production of existing BMW derivatives. For Plant Debrecen alone, this approach will reduce CO₂e emissions from the manufacture of a vehicle, including its high-voltage battery, by around 90% – to about 34 kg CO₂e (when operating at full capacity, compared with other BMW Group facilities).

Paint shops typically rely on gas to reach the high temperatures (up to 180 degrees Celsius) required. Plant Debrecen will be the BMW Group's first car plant supplied exclusively with electricity from renewable energy sources during normal operation, i.e. without using fossil fuels such as oil or gas. Because of its high energy demands, the paint shop is the most important factor in significantly lowering Plant Debrecen’s CO₂e emissions. In the paint shop alone, using electricity from renewable sources reduces annual CO₂e emissions by up to 12,000 tonnes. About a quarter of the plant's annual power needs will be supplied by the 50‑hectare on-site photovoltaic system. Surplus solar energy, such as that generated on non‑working days, is stored in a 1,800 m³ thermal storage system with a capacity of 130 MWh.

The paint shop also uses an energy recovery system. A heat grid was successfully implemented in the design of the new paint shop, delivering additional energy savings of up to 10%. This innovative concept combines multiple measures for efficient energy recovery from the compressed air supply, drying ovens and cooling systems. The recovered waste heat is then used to preheat the water circuit.

In addition, a thermal storage system – holding 1,800 m³ of water, with a capacity of 130 MWh – stores surplus energy from the photovoltaic system during off-peak periods and supplies it as heat during peak demand.

Full digitalisation of production processes further boosts assembly efficiency. The AIQX (Artificial Intelligence Quality Next) IT platform developed in-house at the BMW Group is a key component of the BMW iFACTORY. AIQX uses sensors and camera systems along the production line to automate quality processes. AI evaluates the data and provides real-time feedback to employees on the line. Looking ahead, vehicles on the line will also become active, connected participants in the industrial IoT (Internet of Things) ecosystem. They will conduct self-analysis, interact with plant employees in real time and automatically share and document relevant messages, using tools such as on-board cameras and IT systems. The equipment, tools, components and every BMW in the assembly halls are already digitally connected to the BMW production system.

The building's “finger structure” – an optimised version of the design employed at BMW Group Plant Leipzig – allows 80% of parts to be delivered directly to the correct point of assembly on the line. All-electric in-house logistics also benefit from autonomous tugger trains, which transport high-voltage batteries directly from production to installation points, as well as Smart Transport Robots that autonomously deliver smaller components to the assembly line. By linking all internal and external databases, the entire logistics process achieves unprecedented digital depth, automating many previously manual, interconnected analysis tasks. Well-structured information is therefore available at any time and can be analysed instantly “at the push of a button”.

The BMW Group has created highly intelligent production processes for its new high-voltage batteries developed in-house. Plant Debrecen will be the first of five plants worldwide to begin series production of Gen6 high-voltage batteries. The production processes were initially developed and tested at pilot plants. Artificial intelligence, data analytics and constant knowledge sharing within the production network also play a key role in the ramp-up of production. Digital twins of production and comprehensive AI databases ensure optimisation of processes and employee training. A consistent zero-defect approach enables seamless in-line quality inspections and 100% end‑of-line monitoring. In line keeping with the “local for local” principle, high‑voltage battery assembly takes place directly on site, allowing the production teams to take advantage of infrastructure efficiencies and short distances.

Best of the best:
Plant Debrecen benefits from expertise at all production sites worldwide

BMW Group Plant Debrecen is the first production site in the company's network not tied to a specific primary plant. Instead, it serves as a network plant, combining best practices from various locations worldwide. A major advantage of this approach is that the plant’s current workforce of more than 2,000 employees can be trained within the network, allowing employee know-how from different locations, such as China, South Africa, Mexico, the US and, of course, Germany, to be shared. At the same time, employees throughout the global network gain the opportunity to share their technology insights in Debrecen and bring expertise in production of the Neue Klasse models back to their own location – creating a win-win situation for all the plants.

 

A long version with further details on all technologies is available as a PDF file in the downloads section.

WLTP energy consumption combined: 17,9 - 15,1 kWh/100 km; WLTP CO₂ emissions combined: 0 g/km; CO₂-class: A; WLTP electric range: 679 - 805 km.