BMW's AI-Supported Predictive Maintenance For Assembly Line Efficiency

Mobility Outlook Bureau
28 Nov 2023
12:58 PM
1 Min Read

The AI-supported system does not require additional sensors or hardware; instead, it evaluates existing data from installed components and conveyor element control.


BMW Group Plant Regensburg has implemented a cutting-edge AI-supported predictive maintenance system for its assembly lines, contributing to an average avoidance of around 500 minutes of disruption annually. The smart analysis system focuses on early identification of potential faults in conveyor technology, ensuring optimal vehicle production flow and preventing unplanned stoppages.

The assembly process at BMW Group Plant Regensburg involves vehicles attached to mobile load carriers, passing through production halls in a chain. Any technical faults in the conveyor systems can lead to assembly line standstills, incurring additional maintenance effort and higher costs. To address this, the innovation team at the plant developed a system that identifies potential technical defects early, removing affected conveyor elements for repairs away from production.

An alarm is triggered if anomalies are detected. For instance, data from load carriers is transmitted to the BMW Group's predictive maintenance cloud platform, where an algorithm searches for irregularities such as power consumption fluctuations or conveyor movement abnormalities.

The surveillance system operates 24/7, enabling a quick response to fault reports. Project manager Oliver Mrasek emphasises the system's integration with the BMW Group's central shopfloor management, ensuring standardization for efficient rollout to other plant locations worldwide.

Machine-learning models, developed in-house, utilize heatmaps with colour-coded abnormalities to visualize findings. Continuous improvement is driven by practical insights, and the team is working on connecting additional installations, optimizing the system, and integrating recommended actions into fault messages.

The long-term goal is predictability, with the system learning to estimate the time between fault detection and potential stoppage. The integrated learning system is a first of its kind, earning two patents for BMW Group's in-house development. The success of this innovative approach is evident, with approximately 80% of main assembly lines at BMW Group Plant Regensburg already benefiting from data-driven monitoring. The system is also being deployed at other plant sites, contributing to improved production efficiency and reduced downtime.

Also Read

BMW India Sores High In Jan-Sept Period

Share This Page