When it comes to the maintenance of production systems, the solution chosen by the BMW Group is the use of sensors, data analysis and artificial intelligence (AI). Instead of the previous service regime of regular, rule-based service intervals, preventive maintenance is carried out according to the current state of the system. This not only prevents unplanned interruptions in production, but also contributes significantly to sustainability and resource efficiency by ensuring optimal system availability. Innovative cloud-based predictive maintenance solutions are now being implemented across the global production network.
Preventive maintenance acts as an early warning system in production
The increasing digitalisation of maintenance work makes prevention increasingly important. By monitoring equipment and condition data, predictive maintenance can anticipate system failures before they actually happen. To improve system maintenance, data is used as part of a proactive response to prevent unnecessary interruptions in production. Predictive maintenance also increases efficiency and sustainability by ensuring that intact components are not replaced sooner than they should be.
Predicting situations through a platform based on cloud
Predictive maintenance uses an advanced cloud platform to receive early warnings about potential production outages. The data comes directly from the production systems themselves, connecting to the cloud only once, via gateway, for monitoring and then for continuous data transfer - typically once a second. Independent software functional components (modules) within the platform can be flexibly activated and deactivated as needed, for immediate adaptation to changing requirements. And with a high degree of standardization of independent components, the system is globally accessible, highly extensible and allows for easy implementation of new scenarios and rapid implementation of existing solutions.
Preventive maintenance allows maintenance and repair work to be carried out according to the actual requirements of the systems and to be integrated into the already planned production stoppages. Repairs can be targeted more accurately and more cost and resource efficient. In addition, increased uptime significantly extends the life of tools and systems. These solutions are developed once and are gradually being rolled out across the BMW Group's entire production network.
Variety range applications
Flexible, highly automated mechanical drive train production systems build a conventional motor or the shell of an electric motor every minute. To keep these machines in good working order, predictive maintenance uses simple statistical models - or predictive AI algorithms, in more complex cases - to detect any anomalies. It then issues visual warnings and alerts to let workers know that maintenance work is required.
In the bodywork department, the welding heads perform around 15,000 electrodes per day. To prevent possible downtimes, data from the welding heads around the world is collected by means of specially developed software. This is then sent to the cloud for correlation and analysis using algorithms. All data is visualized in a digital dashboard supporting maintenance processes for global use.
In vehicle assembly, preventive maintenance helps to avoid belt failures. At the BMW Group's plant in Regensburg, for example, the belt system control units operate 24/7 to send data on issues such as electricity, temperatures and locations to the cloud, where they are continuously evaluated. Data specialists can identify the location, status and activity of each belt at any time. Proactive AI models use the data to detect any anomalies and identify technical problems.
