Driving Innovation: NVIDIA Supports BMW Group’s Leap in Production Efficiency With Generative AI

With a history spanning over 100 years, the BMW Group stands as a leading manufacturer of premium as well as luxury and high-performance cars globally. Renowned for its precision engineering, the BMW Group has earned a reputation for delivering high-quality vehicles. Notably, the brand received excellent reliability ratings in J.D. Power’s 2023 U.S. Vehicle Dependability Study.¹

The manufacturer is known to offer maximum expression of an individual’s personality, allowing for customization on options from selecting rare and unconventional colors to changing the trim or even swapping out or adding parts to improve functionality and performance. For its MINI alone, at its height, the BMW Group offered 15 trillion different combinations. Despite the manufacturing complexity, on average, the BMW Group produces one new vehicle every minute.

Situated in Munich, the BMW Group Technology Office stands as an advanced R&D facility dedicated to emerging technologies and product design, actively shaping the future of BMW Group products. This technology hub collaborates with various BMW Group divisions on AI-based projects, guiding them on goal definition, data procurement, model development, and solution deployment.

In pursuit of optimizing numerous production processes, the BMW Group Technology Office spearheaded the development of SORDI (Synthetic Object Recognition Dataset for Industries). This groundbreaking initiative aims to accelerate AI training in production by offering the largest, most realistic open-source dataset for the industrial environment, comprising over 800,000 photorealistic images spanning 80 categories, from pallets to forklifts.

The BMW Group is using NVIDIA Omniverse™ to create virtual factories and simulate complex scenes, along with NVIDIA DGX systems to create synthetic data based on those simulations.

To manage the entire lifecycle management of deep learning models, from development and training to deployment and maintenance, the BMW Group turned to DGX systems with NVIDIA Hopper™ architecture. DGX systems are first used to train deep-learning-based synthetic data generation models. Then, using the synthetic data generated, they’re used to train deep learning models, which include tasks like object detection, image segmentation, image classification, and 6D pose estimation. Lastly, the DGX systems are used in evaluating and testing the trained models.

A BMW Group IT leader commented, “Our transformative journey began with the introduction of the first DGX systems. Over time, we consistently embraced innovation by either upgrading to the latest generation or seamlessly integrating new systems into our cluster. Transitioning from an initial focus on R&D, we now deploy DGX systems in production. With dedicated clusters for specific teams and projects, we efficiently manage jobs, assigning priorities and quotas. Initially employed by R&D teams, DGX now plays a key role in running integral parts of our business.”

“With numerous developers running training jobs seven days a week and managing large datasets exceeding 500K images, the enhanced computing power of the DGX systems enables us to train larger and more complex models. This, in turn, allows for testing more iterations and increasingly diverse parameters to achieve optimal results. DGX systems delivered an 8X boost in our data scientists’ productivity by optimizing resource utilization; we can run a single large training session or launch multiple parallel ones, resulting in a more efficient workflow that supports rapid iteration. Compared to our prior legacy systems, we consistently achieve improvements ranging from 4–6X,” a BMW Group IT leader said.

The SORDI.ai dataset, composed of synthetic images, has significantly impacted downstream AI applications. The team developed LabelTool Lite, which is a pretrained image recognition system refined by employees with suitable photos for specific tasks. For example, the AI training for door sills detection takes less than an hour, requiring no more than five images per task. The AI pipeline processes and enhances this data by adding synthetically generated images and labels, eliminating manual effort. The AI system can then recognize different types of door sills, sounding an alarm if the wrong model is installed. It also detects missing or incorrectly colored stitches in leather products, automating visual inspection with a strong focus on quality assurance.

“Thousands of photos used to be manually categorized to reflect infinite possible variations in the manufacturing process. Using deep learning models trained on DGX, we can now automatically generate hundreds of thousands of images at the push of a button. The time it takes for our employees to implement AI automation in quality assurance has been slashed by over two-thirds. Every possible case, every conceivable combination, including different lighting conditions, is taken into account and covered by our SORDI dataset. The employee can automatically load this data into LabelTool Lite and begin training immediately without any further manual effort, enabling no-code AI,” added the BMW Group IT leader.

The BMW Group utilizes TAO, part of the NVIDIA AI Enterprise software suite, for inference. TAO incorporates AutoML scripts used by the BMW Group for hyperparameter optimization, ensuring optimal accuracies in various applications. An illustrative example includes real-time detection in computer vision models, enabling them to assess and identify faulty parts with precision in milliseconds.

In addition to optimizing production processes and improving quality control, the SORDI dataset is helping with BMW Group’s sustainability strategy. The dataset contains information like an object’s CO2 footprint, age, and energy consumption. Using this data, the BMW Group is able to perform simulations on its DGX systems to optimize energy and CO2 savings for the factory’s products and the components that go into them.

The BMW Group IT leader added, “NVIDIA’s experts, particularly in SORDI AI and AI integration in Omniverse, have been remarkably supportive. The swift responses and comprehensive support were particularly impressive, especially during the initial server or cluster installation and setup. NVIDIA’s assistance extended beyond routine support, providing valuable insights, tricks, and optimizations that greatly contributed to our success and efficiency.”

In the era of Industry 5.0, the BMW Group is pioneering automation to enhance the efficiency of knowledge workers by using generative AI to improve quality control, simulate various production scenarios, and improve supply chain management. From using large language models to write the code to develop strategic factory plant layouts, to text-to-image solutions that accurately generate industrial objects, to tools capable of understanding user-custom prompts and produce appropriate data, to the publication of the SORDI of highly photorealistic dataset, the BMW Group is democratizing artificial intelligence for the manufacturing industry.

“Our commitment includes the ongoing expansion of our DGX infrastructure, a valuable complement to cloud GPU usage. This is particularly crucial in the realm of SORDI AI, where the development of numerous new APIs and networks demands the reliability and performance of the DGX platform,” said the BMW Group IT leader.

[1] J.D. Power.J.D. Power 2023 U.S. Vehicle Dependability Study. February 2023.