Bosch Creates an Industry-wide Automotive Quality Management Service Framework on AWS

In the automotive industry, it’s critical to identify the causes of problems in new vehicle series as early as possible. When they don’t understand these causes, manufacturers can face high—and sometimes needless—expenses for vehicle repairs during the warranty period. The industry estimates that about 50 percent of all components replaced in auto repair shops are essentially intact.

After a vehicle’s market launch, information about detected malfunctions can be slow to reach companies across the automotive supply chain. Defective parts are usually first sent to car manufacturers for an intial inspection, then to parts suppliers. In some cases, up to 6 months can pass between the time that auto repair shops first notice recurring problems and the time those reports make their way to car manufacturers.

Something often stands in the way of targeted troubleshooting at the supplier level, and that’s the absence of contextual information about how individual components or integrated modules interact with other parts of the vehicle. As modern cars become more complex and the networked functionality of installed components rises, diagnostics need digital development, test, and operating data more than ever. So improving how data is exchanged benefits everyone involved.

The problem is that detailed product data, gathered over years of development work, is often protected by patents held by suppliers or manufacturers. That means that data for quality assurance must be exchanged very selectively—both in terms of which product and operational data is shared and in terms of which companies can access that information. Data also needs to be exchanged quickly, securely, in compliance with governance guidelines, and with as little bureaucratic red tape as possible.

COVANTO holds the potential for major impact in data-driven quality management. Car manufacturers and suppliers can save on time and costs by detecting errors and emerging problems early. “By merging data, it is possible to significantly accelerate error detection, by months in some cases,” Sekot says. “This can reduce external error costs by up to 20 percent. The latter can quickly amount to three-digit millions for suppliers in the automotive sector. The effort for troubleshooting in auto repair shops could even be reduced by up to 50 percent.”

Using COVANTO, suppliers will also be able to perform more targeted troubleshooting in the future. The ability to view operational data that’s relevant to their own components often offers suppliers faster visibility into the context of the error they’re investigating. “When suppliers receive a defective part back for analysis, for instance, they often cannot understand why it failed,” Sekot says. “But if they see that—a few weeks before their own part failed—another component that interacts with that part always shows signs of malfunction, they can recognize the pattern and save a lot of valuable time when troubleshooting.”

Such error detection is possible because suppliers have access to the operational data of all components that are functionally connected to their own. At some point, Sekot believes, it would make sense to merge this data with diagnostic data from repair shops so physical error patterns could be matched with data patterns. This could involve monitoring parts to determine whether they continue to fail after they have been modified. In the long term, it might even be possible to digitally map all mechanical and electronic components to enable purely data-driven quality management.

Bosch knows that COVANTO’s success depends on the participation of as many manufacturers and suppliers as possible. This means that the platform must be beneficial to everyone participating. Companies must also be able to trust the technology’s data security and integrity. That’s why Bosch plans to publish the source code for the DAC.

In the meantime, COVANTO's operations have been transferred to Caruso GmbH, the automotive industry's joint data services venture for connected vehicles, which will operate the platform as an independent third party. COVANTO is also being evaluated as a solution module for Catena-X, the European automotive industry’s data network. For now, however, the short-term priority is to add new partners, populate the platform with more data, and keep scaling the service framework.