BMW Group Plant Regensburg is setting new standards in assembly line efficiency with its groundbreaking smart analysis system, aimed at preventing unplanned stoppages and optimizing vehicle production flow.
The cutting-edge predictive maintenance solution employs artificial intelligence (AI) to proactively identify and address potential equipment faults, resulting in significant improvements in production uptime and cost savings.
The smart monitoring system at BMW Group Plant Regensburg focuses on the assembly process, where vehicles are attached to mobile load carriers or skid systems. These carriers traverse production halls in a chain, and any technical fault in the conveyor systems can disrupt the assembly line, leading to increased maintenance efforts and costs.
To circumvent these issues, BMW’s innovation team developed a system capable of early fault detection, ensuring uninterrupted production.
Remarkably, this monitoring system harnesses existing data from installed components and conveyor element control, eliminating the need for additional sensors or hardware. It actively evaluates various data points, including power consumption fluctuations, conveyor movement irregularities, and barcode legibility, to identify anomalies.
When such anomalies are detected, an alert is immediately sent to the maintenance control center, enabling swift action to address the issue.
Project manager Oliver Mrasek emphasizes the system’s continuous operation: “The surveillance monitors at our control center run 24/7, enabling us to respond quickly to any kind of fault report and take the affected vehicle out of the cycle.”
Implementation: AI-supported, standardized, and cost-effective
Predictive maintenance is not just a stand-alone solution; it’s a collaborative effort. The system’s standardization, in cooperation with BMW Group’s central shopfloor management and other plant sites, facilitates its rapid deployment to BMW Group locations worldwide.
A notable advantage is its cost-effectiveness, as it doesn’t require additional sensors, with expenses limited to storage and computing power.
In-house machine learning models are integrated into the system, employing heatmaps with different color codes to visualize fault patterns in various components. This visual representation allows maintenance technicians to respond precisely to the identified issues.
The system’s success is underscored by continuous improvement efforts. The team is currently expanding its capabilities by connecting additional installations, optimizing the system, and integrating recommended actions into fault messages. This enhancement aims to simplify troubleshooting for maintenance technicians by highlighting similar problems that have occurred in the system.
Deniz Ince, the team’s data scientist, emphasizes the broader benefits of optimal predictive maintenance: “Optimal predictive maintenance not only saves us money, it also means we can deliver the planned quantity of vehicles on time – which saves a huge amount of stress in production.”
Future goals: Enhancing predictability and patents
BMW Group Plant Regensburg’s journey in data-driven monitoring of conveyor technology spans six years, with approximately 80% of the main assembly lines now being monitored using this system. While not every fault can be predicted, the system has already prevented around 500 minutes of downtime per year in vehicle assembly alone. Given the plant’s production rate, this translates into significant operational efficiency gains.
The team’s future objectives include enhancing predictability by estimating the time remaining between fault detection and potential stoppage. This feature will help technicians prioritize maintenance tasks based on urgency. Additionally, the system is being explored for use in other areas of the plant, such as equipment used for filling vehicles with brake fluid and coolant.
Remarkably, BMW Group Plant Regensburg’s integrated learning system is a pioneer in its field, earning recognition from equipment manufacturers and leading to two registered patents by the BMW Group.
