Key Takeaway
The automotive industry is under pressure to speed up vehicle development as cars become more complex and software-defined. BMW Group has transformed its development process by utilizing cloud computing and AI, specifically Microsoft’s Azure platform, to enable real-time data analysis. Previously reliant on physical data transfer, BMW now employs an Internet of Things (IoT) device, the Mobile Data Recorder (MDR), which monitors over 10,000 signals and transmits data to the cloud. This shift has doubled data processing volume and increased analysis speed significantly, allowing for faster feedback loops and improved accessibility for non-technical staff through tools like Microsoft Power BI.
The automotive industry is under immense pressure to speed up vehicle development cycles.
As vehicles grow increasingly technologically sophisticated and the industry transitions to software-defined models, manufacturers must handle larger volumes of test data while shortening time to market.
This shift means that traditional development methods, which depended on physical data transfers and manual analyses, are no longer sufficient to meet the complexities of modern vehicles—prompting manufacturers to demand real-time access to test data.
BMW Group has tackled this challenge by revolutionizing its vehicle development process through cloud computing and AI, transitioning from manual data collection to real-time analysis using Microsoft’s Azure platform.
BMW implements Azure AI for development fleet
Prior to 2018, BMW’s development fleet depended on physical hard drives for data transfer, causing engineers to wait at least 24 hours for information to analyze.
Now, Sebastian Heinz, Mobile Data Recorder (MDR) Co-Creator at BMW Group, highlights the integration challenge BMW faces: “BMW is a master of integration. Our goal is to create a car that truly meets customer needs.
“This requires the integration of thousands of digital components.”
The MDR system, an Internet of Things (IoT) device embedded in test vehicles’ networks, monitors over 10,000 signals twice per second, with the data sent to a centralized cloud platform powered by Microsoft Azure’s suite of AI services.
The implementation utilizes various Azure services, including Azure AI Foundry and Azure OpenAI Service, to process the data.
The system features an MDR copilot powered by GPT-4, accessible through a web application, which translates natural language queries into technical database searches.
Additionally, the system architecture incorporates Azure IoT Hub for device management and Azure Databricks for data processing.
Both front-end and back-end services function as microservices on Azure Kubernetes Service, ensuring responsive user interaction.
“For our time-series analytics, Azure Data Explorer was the only option,” Sebastian states, referring to Microsoft’s data analysis service that efficiently processes large volumes of streaming data.
BMW democratises data access through AI
The deployment started with 50 test vehicles and expanded as the team discovered new applications. Now, two-way transmission capabilities allow for real-time configuration changes to test vehicles.
The system stores data in Azure Database for PostgreSQL, which retains copilot interactions, conversations, and feedback to enhance machine learning capabilities over time. This includes vectors for database queries and chat patterns.
“The Azure AI operates seamlessly, significantly lowering the entry barrier and opening up a much broader range of applications,” Sebastian adds.
The company also employs Microsoft Power BI to visualize data for non-technical staff.
“We can transform very complex raw data into an understandable and comprehensive web interface, allowing many BMW Group employees who aren’t engineers to access it,” explains Christof Gebhart, Mobile Data Recorder Co-Creator at BMW Group.
BMW development process transformation
The cloud-based system has doubled the volume of vehicle data processed while increasing analysis speed by a factor of 10, cutting the time needed for insights from days to hours or even minutes.
“The MDR system has been essential for the development of all our cars since 2020,” Christof states.
The platform also allows engineers to send configuration changes back to development vehicles for further testing, creating faster feedback loops.
The company continues to explore additional Microsoft services, including Microsoft Fabric for unified data management and visualization.
“There are always new and innovative Azure features—and our development process greatly benefits from that,” Sebastian notes.
“Azure acts as a turbocharger for delivering the right data to the right person at scale,” Christof adds. “BMW has always combined luxury and performance with cutting-edge technologies. With Microsoft’s cloud and AI leadership, we can harness the potential of our MDR system to accelerate the development of our innovative cars.
“The MDR, powered by Microsoft Azure, guarantees BMW reliability and quality long before the cars hit the road.”








60 Comments